MALDI and LDI Mass Spectrometry of Solids Sachinkumar Dagurao

MASARYK UNIVERSITY Faculty of Science

Czech Republic

MALDI and LDI Mass Spectrometry of Solids

Sachinkumar Dagurao Pangavhane

Doctoral Thesis

Director: Prof. RNDr. Josef Havel, DrSc.

September 2012

Bibliographic identification

Author’s name Sachinkumar Dagurao Pangavhane

Director Prof. RNDr. Josef Havel, DrSc.

Thesis title MALDI and LDI Mass Spectrometry of Solids

Study program PřF D-CH4 Chemistry

Study domain Analytical Chemistry

Year of defense 2012

Keywords LDI/MALDI TOF MS, clusters, binary/ternary/multi-component chalcogenide glasses, phosphorus nitride, clusters structure

Declaration of authenticity

I declare that all material presented in this thesis is my own investigation, except where otherwise stated or fully acknowledged wherever references adapted from other sources.

Brno, September 2012 Sachinkumar D. Pangavhane

Cordial reverence

I feel very deeply my cordial reverence to my mother Mrs. RATNAMALA DAGURAO PANGAVHANE and my father Mr. DAGURAO KASHINATH PANGAVHANE for their blessings and love, plus I’m thankful to my siblings for their support and love during my abroad stay.

Eyes afraid, hands work! - My mother

Acknowledgments

Foremost, I owe an immense gratitude for director of this thesis, Prof. Josef Havel; his substantial comments, insightful criticisms, and dense spirit during discussion as to fight against ‘why and what?’ are profoundly honoured.

I would like to thank Prof. Petr Němec and Prof. Tomáš Wagner from Centre for Material Science, and Department of General and Inorganic Chemistry and Research Centre, Faculty of Chemical Technology, University of Pardubice (Pardubice, Czech Republic), for their steadfast support in the preparation of glass samples.

I would also like to thank Prof. Manuel Valiente, Director, Group of Separation Techniques, Department of Chemistry, University Autònoma of Barcelona (Barcelona, Spain) for his collaboration.

Abstract PhD thesis deals with identification of clusters generated using UV laser (337 nm) from solids. Mass spectra were measured with TOF MS. First of all chalcogenide glasses and their nano-layers were studied. Further, the clusters formed by laser desorption of phosphorus nitride (P3N5) were investigated while clusters stoichiometry was determined via isotopic envelope analysis and computer modeling. Main part of PhD thesis deals with mass spectrometric study of binary (As-Se), ternary (As-S- Se) and Er (III) doped multicomponent (Ga-Ge-Sb-S) chalcogenide glasses in bulk form but also in the form of nano-layers prepared by pulsed laser deposition technique. Several combinations of chalcogenide glasses of different composition (binary, ternary and quaternary systems) were studied. Clusters structure observed in plasma indicates structural motifs of glasses and yields fundamental information on glass structure. The results obtained were completed with Raman spectroscopy. Technique of clusters analysis was also applied to other compounds like phosphorus nitride, which is used in electronics. The P3N5 compound is stable solid. The aim was to evaluate possibility to detect high nitrogen clusters (proposed as fuel of future). It was partially successful, in addition to numerous P-N clusters; the formation of high nitrogen clusters up to

N15 was proved.

Abstrakt (in Czech) PhD Disertace se zabývá určením stechiometrie klastrů generovaných UV laserem (337 nm) z pevných látek. Hmotnostní spektra byla měřena v režimu TOF. Především byla studována chalkogenidová skla a jejich nano-vrstvy, dále pak klastry vznikající laserovou desorpcí a ionisací nitridu fosforu (P3N5). Stechiometrie klastrů byla určována analysou isotopových obálek a počítačovým modelováním. Hlavní náplní distertace je hmotnostní spektrometrie binárních (As-Se), ternárních (As-S-Se) a Er (III) dopovaných vícesložkových chalkogenidových skel (Ga-Ge-Sb-S) v makro formě ale také ve formě nano-vrstev deponovaných pulsní laserovou deposicí. Bzlo studováno několik kombinací chalkogenidových skel o různém složení (binární, ternární a kvarterní systémy). Struktura klastrů pozorovaných v plasmatu odráží do jisté miry chemické strukturní motivy skel a podává tak fundamentální informaci o struktuře skel. Výsledky byly doplněny Ramanovou spektrometrií. Technika klastrové analýzy byla aplikována také na analýzu klastrů tvořených z nitridu fosforu, sloučeniny hojně používané v elektronice. Sloučenina P3N5 je stabilní látkou v pevné fromě. Cílem bylo vyhodnotit možnosti tvorby vysokých dusíkových klastrů (navrhovaných jako palivo budoucnosti). Výsledky jsou částečně úspěchem, kromě četných P-N klastrů byla potvrzena tvorba vysokých dusíkových klastrů – až do N15.

Content Bibliographic identification…………………………………………………………………….....2 Copyright……………………………………………………………………………………….....3 Declaration of authenticity………………………………………………………………………...4 Cordial reverence……………………………………………………………………………….....5 Acknowledgments…………………………………………………………………………..…...... 6 Abstract ……………………….…………………………………………………………..……....7 Abstrakt (in Czech)…………………………………………………………...... 8 Content……………………………………………………………………………...... 9

Chapter 1. INTRODUCTION…………………………………………………………….……...12

1.1 Chalcogenide glasses….………………………………………………...... 12 1.2 Phosphorus nitride…….………………………………………………...... 14

Aims of Thesis……..………………………………………………………………………….....16

Chapter 2. MASS SPECTROMETRY………………………………………………………...... 17

2.1 Introduction……………………………………………………………………..17 2.2 Instrumentation………………………………………………………...... 17 2.3 Brief History…………………………………………………………………....18

Chapter 3. MALDI TOF MS…………………………………………………………………….20

3.1 Introduction……………………………………………………………………..20 3.2 Principle………………………………………………………………………...21 3.3 Sample Preparation……………………………………………………………..22 3.4 Calibration………………………………………………………………………24 3.5 Instrumentation………………………………………………………………....25 3.5.1 Ionization chamber………………………………………………………..25 3.5.2 Mass analyzer……………………………………………………………..26 3.5.3 Detector…………………………………………………………………...28 3.6 Delayed extraction……………………………………………………………...29 3.7 Post source decay…………………………………………………………….....29 3.8 Isotopic envelope…………………………………………………………….....29 3.9 Cluster ion………………………………………………………………………30

Chapter 4. EXPERIMENTAL…………………..……………………………………………….31

4.1 Chemicals & apparatus………………………………………………………....31 4.1.1 Binary As-Se glasses………....………………………………………...... 31 4.1.2 Ternary As-S-Se glasses…………………………...... ………31 4.1.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses…………………...31 4.1.4 Phosphorus nitride P3N5 …………………………………………………32

4.2 Sample preparation…………………………………………………………...... 32 4.2.1 Binary As-Se glasses…………………………………………………...... 32 4.2.2 Ternary As-S-Se glasses….…………………………………...... 32 4.2.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses………………...... 33 4.2.4 Phosphorus nitride P3N5 …………………………………………………34

4.3 Instrumentation…………………………………………………………………34 4.3.1 Binary As-Se glasses…………………………………………………...... 34 4.3.2 Ternary As-S-Se glasses….………………………………...... 35 4.3.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses………………...... 35 4.3.4 Phosphorus nitride P3N5 …………………………………………………36

4.4 Software & computation………………………………………………………..37

Chapter 5. RESULTS AND DISCUSSION…………………………………………………...... 38

5.1 Binary glasses As-Se…………………………………………………...... 38 5.1.1 Effect of laser energy………………………………………………….....38 5.1.2 Positive ion mode………………………………………………………...39 5.1.3 Negative ion mode…….……………………………………...... 40 5.1.4 Raman scattering of As-Se bulk glasses…………….…………………....42 5.1.5 Structure of AspSeq clusters and of As-Se glasses…………………...... 44

5.2 Ternary glasses As-S-Se……………………………………………………...... 47 5.2.1 Negative ion mode…………………………………………...... 47 5.2.2 Positive ion mode………………………………………………………..49 5.2.3 Analysis of nano-layers………………………………………………….52 5.2.4 Structure of binary and ternary clusters………………..……………...... 53

5.3 Rare earth (Er3+) doped quaternary glasses Ga-Ge-Sb-S …………………....…54 5.3.1 Chemical composition and amorphous state of prepared samples………54 5.3.2 Glass structure supported by Raman analysis……………………………55 5.3.3 Mass spectrometry...... 56 5.3.4 Effect of laser energy…………………………………………………….57 5.3.5 Positive ion mode...... 58 5.3.6 Negative ion mode...... 61 5.3.7 Germanium species and relationship with Hydrogen……………………65 5.3.8 Comparison of glass structure and identified clusters…………………...66

5.4 Phosphorus nitride P3N5 ………………………………………..………………67 5.4.1 LDI of P3N5 …...………………………………………..…………….....68 5.4.2 MALDI of P3N5 ………………………………………...…...... 71 5.4.3 Stoichiometry of clusters and the crystal structure of solid P3N5 ….…....74

5.5 Publications……………………………………………………………...……...76

Chapter 6. CONCLUSIONS……………………………………………………………...…….110

6.1 As-Se glasses...... 110 6.2 As-S-Se glasses...... 111 6.3 Ga-Ge-Sb-S glasses...... 111 6.4 Phosphorus nitride…………………………………………………………….112 6.5 General conclusions…..………………………………………………………112 6.6 Obecné závěry (in Czech)...... 113

References………………………………………………………………………………..……..115 List of abbreviations and acronyms………………………………………………………...... 125

Appendix………………………………………………………………….………………...... 127

List of publications Poster presentation List of oral presentations Curriculum vitae

CHAPTER 1. INTRODUCTION

Materials with at least one dimension in the size of nano-meters can be defined as nano materials (NMs). Over the two decades, NMs have been the subject of enormous interest. NMs have the potential for wide range of industrial, biomedical, and electronic applications; they can be metals, ceramics, polymeric materials, etc. NMs might form nano-layers, nano-fibres, etc. Several analytical techniques have been used to characterize/analyze NMs; mass spectrometry is one of them. In this dissertation several NMs were studied using laser desorption ionization time of flight mass spectrometry. The main part of dissertation deals with analyses of chalcogenide glasses in bulk form but also in the form of their nano-layers deposited by pulsed laser deposition technique.

The dissertation also contains results of other materials analyses like phosphorus nitride (P3N5).

1.1 Chalcogenide glasses

Glass is a super-cooled liquid; for centuries glasses were prepared only for windows and drinking vessels but in science the glasses are used in wider sense, for example metallic glasses, polymeric glasses, chalcogenide glass, etc. Chalcogenide glasses (ChG’s) have been studied extensively over the past 60 years.1 ChG’s are the glasses containing one or more chalcogenide elements as substantial constituents; chalcogenide is a binary compound consisting of a chalcogen and one or more electropositive elements or radicals. The chalcogens are the chemical elements in group 16 of the Mendeleev periodic table. Group 16 elements are oxygen (O), sulfur (S), selenium (Se), tellurium (Te), the radioactive element polonium (Po), and the synthetic element ununhexium (Uuh). The term chalcogenide is more commonly reserved for sulfides, selenides, and tellurides, rather than for oxides.2-5 The word ‘chalcogen’ is literally taken from Greek and the general meaning is ‘ore- former’. ChG’s posses many inherent characteristics, but the most crucial is IR transmission range which extends to wavelengths far beyond the range of silica and other glasses. Like traditional

12 glasses, ChG’s can be polished and shaped in lenses and bulk optics, drawn into optical fibres and deposited as thin and thick films. They can transmit and focus the light. Nevertheless, in contrast to ordinary glasses, and more analogous to crystals and semiconductors, chalcogenides can also actively interact with electrons and photons. Such blend of active and passive properties makes them unique in optical and electronic materials. ChG’s are todays exciting field for innovation, their widespread technological applications (either the glass in nano or in bulk form). They are holding scientists attention and forcing them to look for upcoming challenges and demands in the field of infrared optics, photonics, development of next generation computer memory, and recently emerging their applications in medical, military, and aerospace fields.6-12 Chalcogenide glasses drawn into hair-thin threads called optical fibers, which have changed the world and the way we communicate. They have allowed the internet to expand and grow to every corner of the globe.13,14 Such applications of chalcogenide glasses are based on their ideal properties than classical glass (silica): acousto- optics, electro-optics, wide IR transparency (0.8-16 µm), low phonon energy (~350 cm-1 for sulfides and ~250 cm-1 for selenides), photosensitivity, high linear/non linear refractive index, and high chemical durability.15, 16, 12 Rare earth doped chalcogenide glasses are intensively studied for several years. Rare earth ions in optical absorption or light emission show direct relation to the energies in ground and excited states of the electron system.17,18 Low phonon energy of ChG’s affords low probability of multiphonon relaxations among the energy levels of rare earth ions. Therefore, radiative efficiencies of the most rare earth emissions in near and mid-IR are improved. Many radiative transitions in the near/mid IR have been observed in bulk chalcogenide glasses doped with rare earth ions, for example, Pr3+, Tb3+, Dy3+, etc.19 Recently, mechanism of the compositional dependence of blue emission from Nd3+/Tm3+ co-doped Ga-Ge-S-CsBr chalcogenide glasses was 20 investigated. Erbium doped Ga5Ge20Sb10S65 glass posses suitable thermo-mechanical properties for optical fibre drawing, it contains gallium, which allows a better solubilization of rare earth ions, and erbium possesses a mid IR transmission around 4.5 µm.21,22 ChG’s of different compositions, binary As-Se,23 ternary As-S-Se,24 and rare earth Er3+ doped quaternary Ga-Ge-Sb-S glasses were studied via LDI TOF MS. It was shown recently that the LA or LDI TOF MS is powerful tool to study the formation of clusters of solid materials and can be used to characterize their structures.25-28 TOF MS was successfully used to analyze bulk and

PLD-produced nano-layers of binary As-Se, ternary As-S-Se, and multi-component Ga-Ge-Sb-S glasses and it was found that this technique is suitable for the determination of the structural fragments of concerning glasses. The aim of this work was to study the formation of clusters from various glasses by laser ablation in order to elucidate structural fragments and plasma processes during the PLD manufacturing of nano-layers of ChG’s.

1.2 Phosphorus nitride

A solid substance with the composition P3N5 was obtained for the first time by Briegleb and Genther in 1862 by the reaction of magnesium nitride with phosphorus pentachloride.29 Pure, 30 stoichiometric, hydrogen free and crystalline α-P3N5 was reported in 1996 by Schnick et al.

Phosphorus nitride (P3N5) has the potential for various ceramic applications such as sintering additives,31 pigments,32 ionic conductors,33 microporous materials34 and for the doping of 35 36 semiconductors. Recently, nanoscale hollow spheres of P3N5 were synthesized and thin films 37 of amorphous P3N5 were prepared in a low pressure plasma.

The structure of α-P3N5 has a three-dimensional network, consisting of corner-sharing PN4 tetrahedra and it contains two different types of nitrogen atoms (in the ratio 3:2).38 Of the five nitrogen atoms, two are bonded to three phosphorus atoms and three are bonded to two phosphorus atoms. Via high-pressure synthesis, another form, γ-P3N5, has been prepared and, in contrast to the α-P3N5 phase, the larger proportion of the nitrogen in γ-P3N5 is three-fold 30 coordinated with phosphorus. The γ-P3N5 structure consists of one-third of PN4 tetrahedra and two-thirds of tetragonal PN5 pyramids. The refined geometries, properties and the relative stabilities of α-P3N5 and γ-P3N5 and possible high-pressure phase δ-P3N5 with kyanite-type 39 structure have been studied. A nuclear magnetic resonance (NMR) study of solid P3N5 shows the deshielding of nitrogen by the second nearest neighbour effect.40 Recently, crystal structures 41 42 of HP4N7, P4N6O, MP4N7 (M = Na, K, Rb, Cs), and M3P6N11 (M = Na, K, Rb, Cs) 43,44 compounds have been described where the last two show corner-sharing PN4 tetrahedra. 45 Theoretical models of nitrogen-rich species P(Nn)m (n = 3, 4; m = 1-4) were proposed recently. It is known that phosphorus and nitrogen can form clusters. Phosphorus exists in several allotropic modifications, such as white, red, yellow, and black. Laser ablation of red phosphorus leads to the formation of several homo-atomic singly charged poly-anions and cations.46,47 A

14 theoretical study of the isomers of neutral, cationic, and anionic phosphorus clusters Pm (m = 5, 7-9, 20, 25, 33, 41, 49) was performed using B3LYP and DFT in order to predict the most stable isomers of these ions generated via laser ionization.48-52 Phosphorus clusters at higher masses have been reported and it was shown that laser ablation of red phosphorus leads to the formation + ‾ 53 of singly charged Pn (n = 1-189) and Pn (n = 1-225) clusters. Recently, red phosphorus 54 clusters were proposed for calibration in mass spectrometry.

Nitrogen clusters are of great interest because they are theoretically considered to be the best energy-conserving fuel and because of their use in propulsion and explosive applications.55 The energies of the N–N, N=N, and N≡N bonds are 39, 100, and 228 kcal/mol, respectively, at 298 K, with the N≡N bond being the strongest.56 Stable clusters of nitrogen have been searched for 57 intensively after the discovery of fullerene C60 and the possible existence of nitrogen fullerene has been proposed and discussed.58,59 The geometric structure and properties of a energy-rich 60 nitrogen cluster, N60, were studied theoretically, by Wang and Zgierski. Several theoretical investigations of neutral, cationic, and anionic nitrogen clusters Nn (n = 3-13, 16, 18, 20, 28, 60) have been reported using methods such as ab initio, Gaussian-3 studies.61-66 The structures and +/– +/– stabilities of N10, N11 and N13 clusters have been investigated with ab initio, second-order Møller-Plesset, and DFT methods.67-70 The formation of nitrogen clusters has been reported60 71 but, experimentally, only the compounds containing N10 ions have been prepared. Neutral nitrogen clusters and positive cluster ions of frozen nitrogen were desorbed by keV He+ ion + 72 bombardment and a large number of cluster ions Nn (n = 30) were identified. Interestingly, even higher nitrogen clusters, up to n = 40, have been reported to be present in Triton's atmosphere.73

Laser ablation ionization of P3N5 and mass spectrometric analyses of various clusters formed from solid P3N5 might be of interest as this material has also been studied by CVD. The aim of +/– this work was to study the laser ablation ionization of solid P3N5 and to analyse the PmNn clusters formed in order to understand the formation of phosphorus‐nitrogen clusters and/or also to determine whether nitrogen‐rich compounds can be generated.

Aims of Thesis

- To study the possibility and limitations of TOF mass spectrometry for the characterization of nano-materials and/or inorganic precursors of NMs. - To evaluate power and limitations of TOF mass spectrometry for structure elucidation of chalcogenide glasses. - To evaluate possibility of TOF mass spectrometry for identification of clusters formed in plasma plume. From all studied material, the Thesis will try to answer the question ‘Do mass spectra reflect condensed phase chemistry of solids?’

CHAPTER 2. MASS SPECTROMETRY

2.1 Introduction

Mass spectrometry is analytical technique by which chemical substances are identified by the sorting of gaseous ions in electric and magnetic field according to their mass-to-charge ratios. Mass spectrometry conducts many investigations. These include an identification of the isotopes of chemical elements and determination of their precise masses and relative abundances, analysis of inorganic and organic chemicals especially for small amounts of impurities, dating of geologic samples, structural formula determination of complex inorganic and organic substances, the strength of chemical bonds and energies necessary to produce particular ions, identification of products of ion decomposition and the analysis of unknown materials, for example lunar samples, for their chemical and isotopic constituents. Mass spectrometers are employed to separate isotopes and to measure the abundance of concentrated isotopes when used as tracers in chemistry, medicine, and biology.

2.2 Instrumentation

There are five major parts to a MS: the inlet, the ionization chamber, the mass analyzer, the detector, and the electronic readout device. The principle components of MS are shown in Figure 1.

The sample to be Ion source Mass analyzer Detector analyzed enters the Ions generation Ions separation Ions detection instrument through the inlet. In the ionization Electron Ionization (EI) Magnetic sector field Faraday-cup Chemical Ionization (CI) Electric sector field Secondary electron chamber, the sample is multiplier Fast atom Bombardment (FAB) Ion trap ionized and fragmented. Scintillation country Electrospray Ionization (ESI) Quadrupole This can be accomplished Multichannel plate Matrix Assisted Laser Time Of Flight (TOF) in many ways: electron Desorption Ionization (MALDI) bombardment, chemical Figure 1. The principle components of mass spectrometer

17 ionization, laser ionization, electric field ionization, etc. The choice is usually based on how much the analyst wants the molecule to fragment. A milder ionization (lower electric field strength, less vigorous chemical reaction) will leave many more molecules intact, whereas a stronger ionization will produce more fragments. In the mass analyzer, the particles are separated into groups by mass, and then the detector measures the mass-to-charge ratio for each group of fragments. Finally, a readout device usually a computer records the data. Mass spectrometers are often used in combination with other instruments. Since a MS is an identification instrument, it is often paired with a separation instrument like a chromatograph. Sometimes two mass spectrometers are paired, so that a mild ionization method can be followed by a more vigorous ionization of the individual fragments.

2.3 Brief history

MS was born in the field of physics over hundred years ago.74 The foundation of MS was laid in between 1886 and 1898, when Eugen Goldstein, a German physicist, discovered positively charged rays in 1886 and Wilhelm Wien, a German physicist, who discovered that beams of charged particles could be deflected by a magnetic field.75 In more refined experiments carried out between 1898 and 1913, the British physicist J. J. Thomson, who had already discovered the electron and observed its deflection by an electric field, when he passed a beam of positively 76 charged ions through a combined electrostatic and magnetic field. Thomson was able to separate particles of different mass-to-charge ratios. He had separated 20Ne and 22Ne isotopes, 22 77 and correctly identified the m/z = 11 signal as a doubly charged Ne particle. Thomson was awarded the Nobel Prize in Physics "in recognition of the great merits of his theoretical and experimental investigations on the conduction of electricity by gases" in 1906.78 Francis Aston, a British chemist, constructed the first velocity focusing mass spectrograph with mass resolving power of 130 in 1919 and extended the resolving power up to 2000 in 1937.79 He won the Nobel Prize in 1922 for his discovery of isotopes in a large number of non- radioactive elements, and for his enunciation of the whole-number rule.80,81 Ernest Lawrence, an American physicist discovered cyclotron (cyclotron is a type of particle accelerator) and won the Nobel Prize in 1939.82 He developed calutron (calutron is a MS used for separating the isotopes of uranium) in 1942.83 Before 1939, Josef Mattauch, a German physicist and Richard Herzog,

18 invented double-focusing mass spectrograph in 1934.84,85 Such instrument uses both direction and velocity focusing, and therefore an ion beam of a given mass/charge is brought to a focus when the ion beam is initially diverging and contains ions of the same mass and charge with different translational energies. Arthur Dempster, a Canadian-American physicist developed the spark ionization source in 1936 and discovered uranium isotope 235U.86 In 1943, Westinghouse electric international company proclaimed ‘a new electronic method for fast, accurate gas analysis’. The concept of a TOF MS was developed in 1946 by William Stephens.87 In 1956, Fred 88-90 McLafferty, an American chemist, proposed hydrogen transfer reaction observed in MS. Researchers at Dow Chemical interface a gas chromatograph to a MS in 1959. Chemical ionization has been developed by American chemists F.H. Field and M. S. B. Munson in 1966.91- 93 Malcolm Dole was an American chemist; he discovered ESI in 1968.94,95 It is especially useful in producing ions from macromolecules because it overcomes the propensity of these molecules to fragment when ionized. 96 Field desorption refers to the ion source of MS first reported by H. D. Beckey in 1969. In 1974, Melvin B. Comisarow and Alan G. Marshall of the University of British Columbia, Canada revolutionized ICR by developing FT-ICR MS.97 Technique, PDMS was developed by American chemist Ronald MacFarlane and his co-workers in 1976.98,99 In PDMS, the ionization of any species by interaction with heavy particles (which may be ions or neutral atoms) formed as a result of the fission of a suitable nuclide adjacent to a target supporting the sample. John B. Fenn and co-workers (USA) used electrospray to ionize biomolecule in 1984. Franz Hillenkamp, Michael Karas and co-workers (Germany) described and coined the term MALDI in 1985. Koichi Tanaka, Japanese scientist, uses the “ultra fine metal plus liquid matrix method” to ionize intact proteins in 1987. Wolfgang Paul, a German physicist, received the Nobel Prize in Physics "for the development of the ion trap technique" in 1989. Alexander Makarov, a Russian physicist, presented the orbitrap MS in 1999.100 In 2002, John Bennett Fenn and Koichi Tanaka were awarded one-quarter of the Nobel Prize in chemistry each "for the development of soft desorption ionisation methods for mass spectrometric analyses of biological macromolecules." Application of ‘soft ionization’ ESI and MALDI allowed MS to evolve into the realm of biology.101

CHAPTER 3. MALDI TOF MS

3.1 Introduction

In 1985, Franz Hillenkamp, Michael Karas and their colleagues successfully demonstrated the use of a matrix (a small organic molecule) in laser desorption to circumvent the mass limitation.102 These researchers discovered amino acid alanine could be ionized more easily if it was mixed with the amino acid tryptophan and irradiated with a pulsed 266 nm laser. Matrix is sublime compound with strong absorbance at laser wavelength.103 During MALDI sample preparation, analyte of low concentration is mixed with matrix, deposited on a probe or metal plate, and introduced into a pulsed laser beam.104 A substantial burst of ions is produced with each laser pulse. This is the foundation of MALDI. Later developments by Koichi Tanaka and co-workers at Shimadzu corp. demonstrated the application of MALDI to a large biomolecule analysis. Koichi Tanaka and John Fenn have had received 2002 Noble prize in chemistry for their work on developing the soft ionization techniques suitable for large biomolecule analysis.105 The first commercial MALDI instruments introduced in the early 1990.106 Picture of the instrument of MALDI TOF MS is shown in Figure 2. Presently, MALDI technology has widespread applications in the field of biochemistry,107-112 organic chemistry,113-115 inorganic chemistry,116 polymer chemistry,117-120 microbiology,121 etc. It used to monitor and optimize enzymatic digests, characterize proteins, qualitative analysis during peptide synthesis. MALDI has also been used for N- Figure 2. Picture of MALDI TOF MS Model- AXIMA CFR from Kratos Analytical, Shimadzu terminal and C-terminal

20 protein/peptide sequencing. There are also applications in the rapid conformation of post- translational modifications and the quantitation of drugs and chelators conjugated to monoclonal antibodies.

3.2 Principle

Mixing an analyte on a metal sample probe with a suitable matrix compound of a 1.000 up to 10.000 times molar excess, while the matrix is absorbing at the used laser wavelength, causes a co-crystallization process of both matrix and analyte material after evaporation of the solvent. The incorporation of the To mass analyser sample molecules into the + lattice structure of the matrix is supposed to be Analyte ion precondition of the functioning of the LDI + Matrix ion process. The crystallized + + surface of the prepared + sample is then exposed to - an intensive pulse of short Analyte/matrix mixture deposited on target wavelength laser irradiation in the high Figure 3. Principle of the MALDI process vacuum area inside the ion source of the MS. The coupling of the energy, which is necessary for the ions, is performed on UV radiation by resonant excitation of the matrix molecules, e.g., into the -electron system of aromatic compounds. Theoretical calculations suggest that the excitation energy being stored in the matrix molecules relaxes in extremely short time periods into the solid-state lattices causing here a strong distortion of the expansion. This process is followed by transition into a phase in which a part of the solid-state surface is vaporized explosively long before a thermal balance occurs. In this step matrix molecules and sample molecules as well are released into the gas phase (Fig. 3). Obviously, the internal degree of freedom of the molecules sharing this process is low enough and thus even thermal labile macromolecules such as proteins endure this process.

However, this is only valid within a limited range of irradiation power between 106 to 107 W/cm2 on the sample. To a great extent, too high radiation can destroy the sample. The matrix is also considered an agent ionizing the sample molecules, according to photoionization, radical species from matrix molecules caused by transferring protons a high yield of electrically charged sample molecules. An electrode, which is mounted some mm apart opposite to the sample position, is used to generate an electrostatic field in the range of some 100 V/cm to some 1000 V/cm. Depending on the polarity positively or negatively charged ions are accelerated from the sample surface towards the analyzer. Inherent characteristics of a MALDI matrix: -the matrix strongly absorbs the laser light at a wavelength at which the analyte is only weakly absorbing -the matrix reduces intermolecular contacts beyond analyte-matrix interactions thereby reducing the desorption energy -the matrix acts as a protonating (positive ion detection) or deprotonating (negative ion detection) agent, either in solution/solid phase or in the gas phase and is therefore essential in the ion formation process

3.3 Sample Preparation

Sample preparation is one of the most important aspects of MALDI TOF MS.122,123 It is not as simple as just applying a matrix material (Table 1) and an analyte on the sample target. Eppendorf tube for sample storage and micropipette are shown in Figure 4. Several variables influence on the homogeneity of sample and may include the concentration of the matrix and analyte, choice of matrix, analyte sample history (i.e. exposure to strong ionic detergents, formic acid), hydrophobicity or hydrophilicity character, contaminants and compatible Figure 4. Eppendorf tube for sample storage and micropipette

22 solubilities of matrix and analyte solutions, just to mention a few (Table 1). Briefly, the technique involves mixing the analyte of interest with a large molar excess of a matrix compound, usually a weak organic acid, such as CHCA, SA, and DHB, which are the most common matrices. This mixture is placed on a vacuum probe and inserted into the TOF MS for laser desorption analysis. Table 1. Some common MALDI matrices.

Matrix Applications

DHB Peptides, proteins, lipids, and oligosaccharides

CHCA Peptides, proteins, and glycoproteins

SA Peptides, proteins, lipids, and oligonucleotides

During MALDI analysis, the sample (peptide/protein) concentration of about 10 µM and matrix about 10 mM can be premixed in a small eppendorf tube and 1 μl aliquot applied directly to the sample plate. Figure 5. ZipTip pipette tip for sample enrichment Homogeneous sample is needed, then it is important that neither the matrix nor analyte precipitates during the mixing. ZipTip pipette tip for sample enrichment on target is shown in Figure 5. Once the sample is applied on the sample plate, the sample is allowed to evaporate. However, heating is not allowed because to increase the evaporation rate, due for changing the temperature, will alter the crystal growth and protein incorporation. The dried sample is quite stable and can be stored at room temperature in the dark or in a vacuum for several days or more. Besides that some other desirable practices for preparing good MALDI samples Figure 6. Deposition of sample and matrix on MALDI plate

23 include the use of fresh matrix solutions. The use of 0.1% TFA usually eliminates adverse effects caused by a pH above 4. Nonvolatile solvents should be avoided because they can interfere with the crystal growth. Sample and matrix deposition on MALDI plate are shown in Figure 6. Poor MALDI mass spectra can be produced even though the analyst believes the sample is of good MALDI quality and the instrument is performing perfectly (Table 2).124

Table 2: Contaminant concentration tolerated in MALDI -TOF-MS.125

Maximum allowable contaminant concentration (approx.)

Urea 0.5M

Guanidine-HCl 0.5M

Dithiothreitol 0.5M

Glycerol 1%

Alkali metal salts <0.5M

Tris buffer 0.05M

NH4HCO3 0.05M

Phosphate buffer 0.01M

Detergents (not SDS) 0.1%

SDS 0.01%

(http://www.abrf.org/ABRFNews/1997/June1997/jun97lennon.html)

3.4 Calibration

A flight time of ions is converted into the molecular mass which is usually calibrated with standard molecules of known masses.126-132 There are two basic types of calibration methods: external and internal. For the external calibration, one or more spots of standard sample are deposited on target, and the calibration parameters are obtained from this standard sample.126-129 These calibration parameters are used to calibrate samples in other spots. For the internal calibration, the standard molecules are mixed with the sample, and a MALDI TOF MS spectrum of the mixture is acquired or measured.126,130-132 The peaks of the standard molecules are

24 identified and masses of the standard molecules are used to calibrate the entire spectrum. The routine external calibration is convenient but often not accurate, especially for MALDI mass spectrometers of early models.126 The internal calibration could be accurate but it has the following drawbacks:126 A prior knowledge of the sample quantity is needed to determine how much standard molecule should be added to the sample for the internal calibration. If the standard molecule quantities are insufficient, the standard MS peaks may not be identified for the calibration. If too many standard molecules are added, the sample MS peaks will be suppressed. The standard molecule MS peaks may overlap with the sample MS peaks. Unexpected or unknown molecules in the standard sample also add interfering MS peaks to the spectrum. The internal calibration can also be performed using matrix or enzyme autolysis MS peaks. However, these peaks are suppressed when the sample signal is strong.

3.5 Instrumentation

Instrumentation of MALDI TOF MS includes, matrix assisted laser desorption ionization as an ionization source, time of flight tube as an analyzer, and multichannel plate as a detector. The basis of the method is that the molecule is vaporised and ionised by a laser pulse, its mass is then measured by the time taken for the ions to pass down a time of flight analyzer.

3.5.1 Ionization chamber

Detector During the MALDI analyses, a short UV- or IR- laser pulse of a few nano seconds irradiates the + sample (crystal) and leads to + desorption/ionization of the matrix and analyte Time of flight tube molecules in to the gas phase (Fig. 7). The + wavelength of the laser should be close to the maximum absorption to the matrix molecules. + - - + + - + - The processes of material desorption and + ionization of the matrix and analyte molecules Ionization chamber twist together and take place on a micrometers Figure 7. A schematic diagram of MALDI TOF MS

25 geometrical and nanosecond time scale. However, mechanism of MALDI processes is still debated. This might be caused by several factors including type of matrices, analyte molecules, preparation and experimental conditions, which influence on ionization processes. A number of chemical and physical pathways have been suggested for MALDI ion formation, including gas- phase photoionization, ion–molecule reactions, disproportionation, excited-state proton transfer, energy pooling, thermal ionization, and desorption of preformed ions.133

3.5.2 Mass analyzer

In MALDI TOF MS, time of flight tube works as a mass analyzer. The principle of TOF MS involves the measurement of travelling time of ions from ion source to detector located 1 to 2 m from the source. Various analyzers are orthogonal TOF, FT-ICR, QIT, QTOF, QIT-TOF,134 or orbitrap135 as coupled Ions from with MALDI, but axial Field free drift region ion source TOF including + reflectron TOF, + + TOF/TOF has become the most common type Detector for MALDI Figure 8. Linear mode of ions in linear TOF analyzer. spectrometers. TOF-MS is probably the simplest method of mass measurement, though there are hidden complexities when it comes to higher resolution applications. The first commercial TOF instrument was marketed by the Bendix Corporation in the late 1950's (http://www.chm.bris.ac.uk/ms/theory/tof-massspec.html). The same design was used in the Wiley & MacLaren instrument that was published in 1955.136 Recently, TOF-MS is being an essential instrument for organic, inorganic, polymers, and biological analysis, this is especially the case with the coupling of TOF-MS to MALDI ionization methods and the development of high-resolution and hybrid instruments, for example Q-TOF and TOF-TOF configurations. TOF MS is usually constructed with two mass analyzers: 1. Linear TOF analyzer (Fig. 8) 2. Reflectron TOF analyzer (Fig. 9).

A significant advantage of the reflectron TOF analyzer as compared to linear TOF analyzer is the increase in resolving power. All precursor ions leaving the ion source have approximately the same KE, but because the ions have different m/z values, they have correspondingly different velocities. As the ions transverse in field free region from source to the detector, they separate into groups or packets according to velocity, which is function of their m/z values. There are three factors for low resolving power of linear TOF analyzer: 1. spread in flight times due to differences in kinetic energies of the ions of the same m/z values. 2. spread in flight times due to the different positions from which the ions start when they are pulsed out of the source 3. differences in flight times between two isomass ions starting in the same plane in the source and moving with the same initial velocities, but in opposite Ions from ion source + directions. + + During the longer flight time the + precursor ions might Detector + fragment, resulting in + smaller product ions. Figure 9. Ions of same m/z but different velocity arrived at detector The product ions with same time because travelling time compensated in reflectron mode. almost have the same velocity as their precursors, but differ in their kinetic energies. Since KE is directly proportional to mass, the smaller product ions have less KE than the precursors. The TOF analyzer cannot resolve the ions in linear mode because the fragment ions (product ions) and precursor ions all have the same velocity. However, in reflectron mode, the MALDI TOF instrument takes advantage of the differences in kinetic energies to distinguish between the precursor and fragment ions. Reflectron TOF analyzer is only capable of resolving the KE of ions. For better resolution TOF MS is equipped with mirror in order to reflect ions of same m/z that have different energies. In reflectron TOF MS, ion mirror is constructed in the form of electric field (Fig. 9) that opposes and is of greater magnitude than the electric field in the ion acceleration region. The position of mirror is less than 180˚ in order to avoid reflection of ions

27 directly back into the source. + As3Se3 Three ions of the same m/z, but 100 Linear mode with different KE (i.e. few 50 volts out of several kilovolts) 0 are shown in Figure 9. Ion of 462.5 100 460.5 464.5 highest KE enters first (because Reflectron mode 50 458.6 466.6

it has the highest velocity) in intensity[%] Relative 456.6 0 the opposing electric field and 440 450 460 470 480 penetrates the deepest on the m/z Figure 10. LDI mass spectra of As-Se glass comparing mass resolution mirror until their KE reaches obtained in the linear (upper) and reflecting (lower) modes utilizing continuous ion extraction at laser irradiance 70 a.u., 20 kV accelerating zero. At this point ions begin to potential, 50 laser pulses averaged, linear positive ion mode. be accelerated by the electric field in the opposite direction. Ions of highest initial KE penetrate deeper into the mirror, will acquire more KE during reacceleration than ions of lowest initial KE because the former transverse a longer segment of the electric field in the mirror. Therefore ions leave the mirror with the same distributions of the kinetic energies and velocities as when they entered; however the length of their flight paths now differ because of their differential penetration into the opposing electric field.137 Finally, all ions of same m/z will arrive at detector at same time and will get spectra of higher resolution (Fig. 10). Typical features of TOF analyzer is given in Table 3. Table 3. Typical features of TOF analyzers. - spectrum over the complete mass range can be obtained in microseconds - theoretically, no upper mass limits exist for this type of mass analyzer - high sensitivity is achievable due to the high ion transmission - resolution-improving elements (delayed extraction, reflectron)

3.5.3 Detector

Instrument of MALDI TOF MS is usually equipped with MCP for the detection of particles (ions or electrons).138 MCP has many separate channels for the multiplication of electrons and improves the resolution via secondary emission. A detector positioned at the end of the field-free

28 region determines the flight-time for each m/z. At a fixed KE, small ions travel at higher speed than large ions. The equation for determining the m/z is the following: m t 2  2eU 2 z L where m is mass, z is charge, e is elemental charge, U is the acceleration voltage, t is the time of flight, and L is the length of the drift zone.

3.6 Delayed extraction

Mass resolution and mass accuracy can be substantially improved in linear and reflectron TOF MS by utilizing the technique of DE.139-145 DE means the extraction of ions by some short time delay after the ionization event, i.e. the initial burst of ions and neutrals produced by the laser pulse to equilibrate and dissipate before the ions are accelerated into the flight tube. The fundamental approach originates from the early work of Wiley and McLaren.136 DE is more of a correction to spatial homogeneity which is only critical at the detector.

3.7 Post source decay

The term PSD refers to the fragmentation of molecular ions during their flight time in the MS, subsequent to ionization and acceleration away from the ionization source. PSD fragment ion mass analysis represents an equivalent to conventional tandem MS technique.146-151 Mass spectrometric analysis of product ions from PSD of precursor ions is the powerful method for the primary structure analysis of compounds (organic, inorganic, biomolecule).152-154

3.8 Isotopic envelope

Isotopic envelopes are useful in the interpretation of measured spectra (Fig. 11). The presence of isotopes at their natural abundances makes it essential to define whether an experimental mass value is an "average" value, equivalent to taking the centroid of the complete isotopic envelope. Envelope displays an experimental data superimposed on calculated isotope distributions

(theoretical), and calculates a least-squares goodness of fit between the two (experimental and theoretical m/z) or envelope is a powerful tool for the interactive calculation and visualization of complex Figure 11. Comparison of selected part of experimental mass isotope distributions for Spectra with theoretical ones concerning As2Se3 glass analysis. comparison to experimental Conditions: linear positive ion mode, laser energy 70 a.u. + data. An example of theoretical and experimental isotopic envelopes for As3Se4 cluster is given in Figure 11; there is a good agreement between the theoretical and experimental spectra.

3.9 Cluster ion

An ion formed by the combination of two or more atoms, ions or molecules of a chemical species, often in association with a second species is called cluster ion.

CHAPTER 4. EXPERIMENTAL

4.1 Chemicals & apparatus

4.1.1 Binary As-Se glasses

Elemental arsenic and selenium of purity (5N) were used. The arsenic was obtained from Laboratories de la Vielle (Brussel, Belgium) and stored under an inert argon atmosphere. Selenium was from Aldrich Chemical Company, Inc. (Milwaukee, WI, USA). Red phosphorus was purchased from Riedel de Haën (Hannover, Germany) and was used for mass calibration.54 Water was doubly distilled using a quartz apparatus from Heraeus Quartzschmelze (Hanau, Germany). All other reagents were of analytical grade purity.

4.1.2 Ternary As-S-Se glasses

Elemental arsenic, sulfur, and selenium of purity (5N) were used. The arsenic was from Laboratories de la Vielle (Brussel, Belgium) and was stored under inert Ar atmosphere. Sulfur was ‘pure sulfur’ produced by s.p. LACHEMA (Brno, Czech Republic) purified by multiple chemical distillations at the University of Pardubice. Selenium was from Aldrich Chemical Company Inc. (Milwaukee, WI, USA). Red phosphorus from Riedel de Haën (Hannover, Germany) was used for mass calibration. Water was doubly distilled using a quartz apparatus from Heraeus Quartzschmelze (Hanau, Germany). All other reagents were of analytical grade purity.

4.1.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses

Elemental gallium, germanium, antimony, sulphur (5N) and erbium (3N) were purchased from STREM, Chemicals Inc. Boston, Massachusetts, USA or UMICORE, Brussel, Belgium. Red phosphorus was purchased from Riedel de Haën (Hannover, Germany) and was used for mass calibration. Water was doubly distilled using a quartz apparatus from Heraeus Quartzschmelze

(Hanau, Germany). Acetonitrile and toluene was purchased from Penta Chemicals (Mainaschaff, Germany). All other reagents were of analytical grade purity. Parafilm 'M' was purchased from the Pechiney Plastic Packaging Company (Chicago, IL, USA).

4.1.4 Phosphorus nitride P3N5

Phosphorus nitride, P3N5 (Batch No. 930823), was obtained from Phillips Lighting (Maarheeeze,

The Netherland). DHB, fullerene (C60), CHCA, DIT, HPA, TMN, and ANP were purchased from Sigma‐Aldrich (Taufkirchen, Germany). Elemental sulfur and selenium were purchased from Lachema (Brno, Czech Republic). Acetonitrile was purchased from Penta Chemicals (Mainaschaff, Germany). Red phosphorus fromRiedel de Haën (Hannover, Germany) was used for calibration. Water was doubly distilled using a quartz apparatus from Heraeus Quartzschmelze (Hanau, Germany). All other chemicals were of analytical grade purity. Parafilm 'M' was purchased from the Pechiney Plastic Packaging Company (Chicago, IL, USA).

4.2 Sample preparation

4.2.1 Binary As-Se glasses

Elements (As and Se) were crushed into fine particles and mixed in different molar ratios As:Se

(1:2, 2:3, 4:4, 4:3, and 7:3) which will be represented further as ‘AsSe2 glass’, etc. The mixtures were homogenized in a rocking furnace for 48 h at a temperature of 700 ˚C in vacuum-sealed quartz glass ampoules. The resultant glasses, after cooling down to room temperature in air, were crushed in an agate ball mill. About 0.1 mg of fine powdered glass was suspended in 200 μL of toluene and 1 μL of this suspension was deposited on the target sample plate and dried at RT in a stream of air.

4.2.2 Ternary As-S-Se glasses

Samples of elements (As, S, Se) were crushed into fine particles and were mixed in a specific molar ratio, e.g. 33:33.5:33.5, which will be represented as As33S33.5Se33.5. The mixtures were

32 homogenized in a rocking furnace for 48 h at 700 ˚C in vacuum-sealed quartz glass ampoules. The resultant glasses, after being cooled down to 0 ˚C in an iced-water tank, were crushed in an agate ball mill and ~1mg of fine powdered glass was suspended in 200 μL of toluene and 1 μL of the suspension was deposited on the spot of a MALDI sample plate and dried at RT via a stream of air. Before the deposition of the samples the target plate was cleaned carefully with ethanol and deionized water, and then dried at RT with a stream of air. Each glass sample suspension was deposited on the target, being kept a distance of ~2 cm from the other samples in order to avoid contamination.

4.2.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses

Non-doped Ga-Ge-Sb-S as well as erbium doped (0.05, 0.1, and 0.5 w. % of Er3+)

Ga5Ge20Sb10S65 glasses were prepared by conventional melting and quenching method. Above mentioned, high purity raw materials were used for the glasses synthesis. However, commercial sulphur was further purified by successive distillations to limit carbon and hydrates or sulphide hydride impurities.155 Then, the required amounts of chemical reagents were introduced in silica ampoules; afterwards evacuated and sealed silica tubes were heated at 850 °C for 12 hours in a rocking furnace to ensure the homogenization of the melt. After water quenching, the glass rods were annealed near, but below their glass transition temperatures for 6 hours. For energy- dispersive X-ray analysis, X-ray diffraction measurements and Raman scattering spectroscopy, polished samples of prepared glasses were used. For MS, the glasses were crushed in an agate ball mill. About 1 mg of fine powdered glass was suspended in 1 ml of toluene and 1 μL of this suspension was deposited on the target sample plate and dried at RT in a stream of air. Desiccated red phosphorus (~0.1 mg) was suspended in 100 μL of acetonitrile and 1 μL of the suspension was deposited on the target plate and used for the analysis. Samples were measured from parafilm in order to avoid contamination by other elements coming directly from the target during the plasma processes. Parafilm was used as a support material as it can provide good sensitivity, mass resolution, and ion current stability.156 Before the deposition of the samples, the target plate was always cleaned carefully with ethanol, acetone and deionized water and then dried at RT in a stream of air. The target was covered with a small layer of parafilm (~0.5 × 0.5 cm) and each sample was deposited on the target at a distance of ~2 cm from the other samples

33 in order to avoid contamination.

4.2.4 Phosphorus nitride P3N5

Phosphorus nitride was dried in a desiccator for 72 h before the experiments and ~0.1mg fine powder was weighed and suspended in 200 μL of acetonitrile. One μL of the suspension was deposited on the spot of a MALDI sample plate and dried at RT via a stream of air. Matrices were dissolved in acetonitrile (1mg/mL or saturated solution) and 10 μL of the matrix solution was mixed with the suspension of P3N5 and deposited on the sample plate. The suspension of phosphorus nitride was mixed with matrix solution in the ratio 1:10. Desiccated red phosphorus (~0.1 mg) was suspended in 100 μL of acetonitrile and 1 μL of the suspension was deposited on the target plate and used for the analysis. Samples were measured from parafilm in order to avoid contamination by other elements coming directly from the target during the plasma processes. Parafilm was used as a support material as it can provide good sensitivity, mass resolution, and ion current stability. Before the deposition of the samples, the target plate was always cleaned carefully with ethanol, acetone and deionized water and then dried at RT in a stream of air. The target was covered with a small layer of parafilm (~0.5 × 0.5 cm) and each sample was deposited on the target at a distance of ~2 cm from the other samples in order to avoid contamination.

4.3 Instrumentation

4.3.1 Binary As-Se glasses

The mass spectra of the binary chalcogenide glasses were measured on an Axima-CFR TOF MS supplied by Kratos/Shimadzu (Manchester, UK) using a nitrogen laser of 337 nm. The MS was equipped with linear and reflectron TOF detectors and with DE. The experiments were performed at frequency of 10 Hz and a pulse time width of 3 ns. The laser fluence was 60 mJ per pulse. Full laser power was marked on the instrument as 180 a.u. (6 mW) while the irradiated spot size was approximately 150 mm in diameter. The laser power can be changed in the interval 0–180 a.u. Spectra were accumulated from at least 100 shots. Positive and negative ion mass spectra were recorded in linear and reflectron modes. For external mass calibration of spectra

34 measured, the ions generated from DHB and red phosphorus were used. Raman scattering spectra were recorded at RT with a Bruker IFS 55/FRA106 Fourier transform Raman spectrophotometer (Ettlingen, Germany) using a backscattering method with a Nd:YAG laser (1064 nm, 50 mW) as the excitation source; 200 scans were collected at 2 cm-1 resolution.

4.3.2 Ternary As-S-Se glasses

A bulk sample was used for preparation of the PLD targets (12 x 12 x 4 mm blocks). The thin nano-films were prepared in the Tesla UP 858 vacuum system coupled to a KrF laser (Lambda Physik COMPex 102; Lambda Physik, Göttingen, Germany) operating at λ = 248 nm with constant output energy 300 mJ pulse-1, pulse duration t = 30 ns and repetition frequency f = 20 Hz. The laser incidence angle was 45˚, the target to substrate distance was 5 cm and a plane parallel setup was used. Off-axis geometry with target rotation (10 rpm) and substrate rotation (7.5 rpm) was used to improve the thickness homogeneity. Single crystalline silicon wafers were used as substrates. The mass spectra of the ternary ChG were measured on an Axima-CFR MS from Kratos/Shimadzu (Manchester, UK). The repetition modes of experiments were performed at a frequency of 10 Hz and a pulse time width of 3 ns. The laser power used was higher than the formation/desorption ion threshold and is given in a.u., where the full laser power gain was 180 a.u. (6 mW). The irradiated spot size was approximately 150 μm in diameter. Spectra were accumulated from at least 100 shots. Positive and negative ion mass spectra were recorded in linear and reflectron modes. Clusters of red phosphorus were used for external mass calibration.

4.3.3 Multi-component Erbium doped Ga-Ge-Sb-S glasses

The chemical composition of prepared glasses was measured using SEM (JEOL JSM 6400) connected with EDX. Data of XRD were obtained with a D8-Advance diffractometer (Bruker AXS, Germany) with Bragg–Brentano θ–θ geometry (40 kV, 40 mA) using CuKα radiation with secondary graphite monochromator. The diffraction angles were measured at RT from 5 to 65° (2θ) in 0.02° steps with a counting time of 5 s per step. The optical transmission of the polished samples was measured from 400 nm to 3200 nm using a spectrophotometer (Perkin-Elmer, Lambda 1050, state/city, country).

Raman scattering spectra were collected at RT by double-monochromator Raman spectrophotometer HR800 (Horiba Jobin-Yvon) with 785 nm laser diode as excitation source. Light intensity of laser beam on the sample was kept at low level (to avoid changes of Raman spectra due to thermally-induced structural transformations induced by absorption of high laser power densities) by optical density filters. Mass spectra were acquired on an AXIMA CFR (Kratos Analytical, Manchester, UK) TOF instrument equipped with a nitrogen laser from Laser Science Inc. (Franklin, MA, USA), operated at a wavelength of 337 nm. The repetition modes of experiments were performed at a frequency of 10 Hz and a pulse time width of 3 ns. The laser fluency was 60 mJ per pulse. Full laser power was marked on the instrument as 180 a.u. (20 mW) while the irradiated spot size was approximately 150 μm in diameter. Energy density of laser was 133 J/m3. The laser power can be varied between 0 and 180 a.u. and this relative scale will be used later. Analyses were carried out at a pressure in the TOF analyzer of 10–4 Pa and positive or negative ion spectra were recorded in linear and reflectron modes. Mass spectrometric analysis was performed using at least 100 shots and the obtained data were accumulated. Red phosphorus was used for external calibration in both ionization modes.

4.3.4 Phosphorus nitride P3N5

Mass spectra were acquired on an AXIMA CFR (Kratos Analytical, Manchester, UK) TOF instrument equipped with a nitrogen laser from Laser Science Inc. (Franklin, MA, USA), operated at a wavelength of 337 nm. The repetition modes of experiments were performed at a frequency of 10 Hz and a pulse time width of 3 ns. The laser fluency was 60 mJ per pulse. Full laser power was marked on the instrument as 180 a.u. (20 mW) while the irradiated spot size was approximately 150 μm in diameter. The laser power can be varied between 0 and 180 a.u. and this relative scale will be used later. Analyses were carried out at a pressure in the TOF analyzer of 10–4 Pa and positive or negative ion spectra were recorded in linear and reflectron modes. Mass spectrometric analysis was performed using at least 100 shots and the obtained data were accumulated. Red phosphorus was used for external calibration in both ionization modes. As phosphorus clusters are formed during the ionization of P3N5, internal calibration using these clusters was then also used (phosphorus clusters are marked in the spectra). In this way typical

36 mass precision of ~± (0.03–0.07 m/z units) was achieved.

4.4 Software & computation

Theoretical isotopic envelopes were calculated using Launchpad software (Kompact Version 2.3.4, 2003) from Kratos Analytical Ltd.

CHAPTER 5. RESULTS AND DISCUSSION

5.1 Binary glasses As-Se

Powders of glasses with As:Se molar ratios from 0.5 to 2.3, marked further as ‘AsSe2, As2Se3,

As4Se4, As4Se3, and As7Se3 glasses’ were analyzed. Mass spectra were recorded by TOF MS and singly charged positive and negative AspSeq cluster ions were generated via LA or LDI. The threshold laser energy was found to be almost the same in positive and negative ion mode, at ~60 a.u. Mass spectra with sufficient mass resolution and intensity were observed at laser energies in the interval 60–100 a.u. in both modes. Due to the natural abundances of the selenium isotopes, the clusters containing selenium show several peaks, forming a so-called, ‘isotopic envelope’. The mass spectra were analyzed via comparison of experimental isotopic envelopes with the theoretical models and in this way the stoichiometry of the clusters was determined.

5.1.1 Effect of laser energy

The effect of laser energy on mass spectra of all glasses was followed. As an example, the mass spectra of an As7Se3 glass at different laser energies are shown in Figures 12(A) and 12(B). It + was observed that at the threshold laser energy (60 a.u.) AspSeq clusters were already formed.

Figure 12. (A) Positive ion mode mass spectra of As7Se3 glass for different laser energies. Maximum intensity was set to 200 mV (100 % relative intensity). Selected m/z values given correspond to a maximum peak in the cluster. (B) Negative ion mode mass spectra of As7Se3 glass for different laser energies. ‘Y’ axis is normalized to 40 mV (100% relative intensity).

On increasing the laser energy the intensity of the peak increases but the resolution of the mass spectra falls at laser energies of 90 a.u. and higher. Via analysis of the mass spectra the formation + + of As3Seq (q = 1–5) clusters was demonstrated and the As3Se cluster was found to be the most abundant. In the interval m/z 800–2000, the mass spectra showed features indicating that the glass being analyzed is a type of polymer. The difference in m/z values for the neighboring peak clusters was found to decrease from ~15 to 14 m/z units (Fig. 12(B), inset in the right upper corner) with increasing m/z value. Because the theoretical mass differences between neighboring + + + AspSeq clusters, for example, As2Se2 , As3Se3 , and As4Se4 , are 10, 13, and 18 m/z units (and so on), we propose that the structure of AspSeq glass might represent a polymer made up of AspSeq units with different p and q values. However, it is possible that high laser energies can induce rearrangement or reconstruction of chemical bonds and as a result, new structural species or clusters may be synthesized (Fig. 12(B)).

5.1.2 Positive ion mode

+ Similar results were obtained from glasses of different composition. For all the glasses, As3Seq (q = 1-5) clusters were found to be the main species. Analysis of the isotopic envelopes shows that the ions formed are singly charged. An example of mass spectrum concerning positive ion mode for As2Se3 glass is given in Figure 13(A). It was found that the + series of clusters As3Seq (q = 1-5) is preferably formed in the mass range 300-700 m/z. In spite of supra-pure materials being used to produce the glasses, it is interesting to + note that e.g. As3Se species (with highest Figure 13. As2Se3 glass mass spectra and results of analysis: abundance) is (A) linear positive ion mode and (B) reflectron negative ion mode. accompanied by Conditions: laser energy 90 a.u., maximum intensity was set to 400 mV (A) and 1200 mV (B).

+ + + + hydrogenated (As3SeH2 , As4H , As4H2 , and As4H3 ) clusters. The relative intensity of the + hydrogenated clusters was estimated to ~10-30 % of the intensity corresponding to the As3Se species. We have found that hydrogen is a trace contaminant in metallic arsenic and selenium (in spite of their high purity). Because the detection limit in TOF MS is as low as 10-15 mol, very low level traces of hydrides in ChG can be detected. A small yield of the hydrogenated species + + + + was also found to accompany the As3Se2 , As3Se3 , and As3Se4 clusters. The traces of SeH5 + + and S4H3 and of mixed oxides such as AsSeO were also observed for the As2Se3 glass. The + + cluster of peaks corresponding to As5Se3 species overlaps with some yield of As3Se5 cluster. + Low abundance of As5Se4 was also detected for the As2Se3 glass. + The mass spectra of As4Se3 glass show high abundant As3Seq (q = 2-5) clusters with intensities decreasing for higher q values in contrast to the mass spectra of As2Se3, As4Se4, and

As7Se3 glasses. For glasses with higher arsenic content the group of the peaks around ~ m/z 305 + + + + corresponds to a single As3Se species. The As3Se2 , As3Se4 , and As3Se5 clusters overlap with + a low percentage of some other AspSeq clusters, whose composition was difficult to determine. It was observed that in mass spectra of the samples possessing a high arsenic content, the + + arsenic-rich AspSeq clusters (such as As3Se ) are formed in high abundance and clusters with + higher q values such as As3Seq (q = 2, 5) are also formed but in lower abundance. AsSe and + As2Se clusters were not observed.

5.1.3 Negative ion mode

In comparison with the positive ion mode, ionization in linear and/or reflectron negative ion mode leads to the generation of a slightly higher number of clusters while the spectra in reflectron negative ion mode are better resolved. All the species generated are singly charged as in the positive ion mode. Similar results were observed for all the glasses under study. An example of mass spectra concerning the formation of negatively charged ions is given in Figure - - - 13(B). Three series of clusters, AsSeq (q = 1-3), As2Seq (q = 2-4), and As3Seq (q = 2-5) were - - - identified. The clusters from the As2Seq (q = 2-4) series are accompanied by AsSe3 , As3Se2 , and - - As3Se3 clusters, respectively. The AsSe2 species was found to be dominant in the spectra of all - - - - the glass samples. In addition AsSe3 , As2Se3 , As3Se3 , and As3Se4 clusters were identified. - - - Individual Se and Se2 species were also detected; the Se2 species overlap with the low

40 abundance of AsSe- clusters (5% relative intensity). In contrast to the positive ion mode, the other species were identified in negative - - ion mode: Se , Se2 , AsSe -, AsSe -, 2 3 Figure 14. Comparison of selected parts of experimental mass spectra with - - As2Se2 , As2Se3 , and theoretical ones concerning As4Se3 glass analysis. Conditions: linear negative - ion mode, laser energy 70 a.u. As2Se4 . Examples of - theoretical and experimental isotopic envelopes for high abundance AsSe2 and lower abundant - As3Se4 clusters, respectively, are given in Figure 14; there is good agreement between theoretical and experimental spectra.

Interestingly, peaks corresponding to As3Se4 species were found to be free from interferences from other species, both in the positive (Fig. 15) and in the negative ion modes (Fig. 14). A part of the mass spectrum given as an inset in Figure 16(A), shows the relative abundance of the - AspSeq clusters. It is suggested that the higher selenium content in these species might be caused by (i) the higher electronegativity of selenium which makes ionization easier and (ii) the higher selenium content in this glass. The species detected in reflectron negative and in the linear positive ion modes are slightly different (Fig. 16(A) and 16(B)). The main species identified in positive ion mode + was As3Se , Figure 15. Comparison of the experimental and theoretical isotopic + accompanied by envelopes concerning mass spectra of As2Se3 glass for the As3Se4 cluster. Conditions: linear positive ion mode, laser energy 90 a.u.

+ + + + + + lower abundance AsSe , As2Se , As3Se2 , As3Se3 , As3Se4 , and As5Se3 clusters. The analogues - - Se particles and Se2 clusters were not observed in positive ion mode. It was observed that the + abundance of As3Seq (q = 2-4) clusters increases for higher q value in positive ion mode but in - - negative ion mode the abundances of the As2Se3 and As3Se4 clusters are quite similar as the - - overlap of the As2Se3 cluster with As3Se2 is low (25% relative intensity). A summary of the positively and negatively charged species identified in the mass spectra is given in Table 4.

Figure 16. Comparison of negative and positive ion mode mass spectra concerning AsSe2 glass. Conditions: (A) reflectron negative ion mode, laser energy 90 a.u. and (B) linear positive ion mode, laser energy 100 a.u. Table 4. Summary of cluster ions identified in mass spectra

Number of selenium atoms in the clusters 1 2 3 4 5 Positive ion mode

As = 1 AsSe+ + 2 As2Se + + + + 3 As3Se As3Se2 As3Se3 As3Se4 + + 5 As5Se3 As5Se4 Negative ion mode As = 0 Se- Se - 2 - - - 1 AsSe AsSe2 AsSe3 - - - 2 As2Se2 As2Se3 As2Se4 - - - - 3 As3Se2 As3Se3 As3Se4 As3Se5

5.1.4 Raman scattering of As-Se bulk glasses

On considering the stoichiometry of the clusters given in Table 4, it should be noted that MS cannot determine the structures of these clusters but may provide useful indicators to their structures. Additional information on the structures of bulk As-Se glasses can be obtained by Raman spectroscopy. The dominant feature of the Raman spectra of stoichiometric As2Se3 glass Figure 17. Raman scattering spectra of As-Se glasses normalized is a broad band peaking to the amplitude of the dominant band of stoichiometric As2Se3 glass at ~220–230 cm-1. near 220-230 cm-1 (Fig. 157 17), which is connected with the stretching vibration modes of AsSe3 pyramidal units. This is in agreement with Figure 16(A) where analogous species were detected. In the Raman spectra of -1 selenium-rich samples such as AsSe2, the main band overlaps with the band at ~238 cm , which 158 has been assigned to –Se–Se–Se– chains and/or As–Se vibrations in AsSe3 units. Similarly, an -1 additional shoulder at 257 cm is often suggested to be due to the fractures of Se8 rings.

In the case of arsenic-rich As-Se glasses (As4Se4, As4Se3, and As7Se3 samples), the dominant feature in the spectra is a broad band between 180 and 300 cm-1, which appears to be formed by overlapping of several sharp individual bands with maxima near ~196, 206, 224, 238, 255, and 280 cm-1 (Fig. 17). The Raman band peaking at ~224 cm-1 has been attributed to the stretching vibration modes of AsSe3 pyramids. The other bands can be connected either with vibrations of -1 -1 -1 -1 cage-like molecules of As4Se3 (196 cm – E (ν8); 238 cm – A1 (ν2); 255 cm – E (ν7); 280 cm -1 -1 159,160 – A (ν1), or with vibration of As4Se4 molecules (~190 cm – A1, 206 cm – B2). In accordance with Nagels et al.,161 the Raman band at 206 cm-1 can be also connected to the presence of the As4 arsenic cluster. It follows from Figure 17 that the relative amplitude of the Raman band at 206 cm-1 increases strongly with increasing As content in the glasses; this

43 tendency supports the assignment of that frequency to the vibrations of As clusters such as As4 or

AspSeq ( p/q >4/3) structural units. We assume that in arsenic-rich glasses and films, there is a possibility of the formation of structural units of the type of AspSeq (p = 2-3, q = 1-4), e.g. AsSe3 pyramids in which Se can be substituted by As and As2Se2, As3Se, or even that As4 structural units can be formed. The vibrational frequencies of AsSe3-like structural units, e.g. As2Se2 or

As3Se, which can be formed in As-rich glasses by the substitution of one or two Se atoms in

AsSe3 pyramids by As atoms, may not be very different from the AsSe3 pyramids because the atomic masses of As and Se are similar (74.922 and 78.96), and the energies (and probably force constants) of the As–Se and As–As bonds are similar: E (As-As) = 200 kJ/mol, E (As-Se) = 225 kJ mol-1.162 The group of weaker Raman bands in the low frequency region (100–180 cm-1) of 163 arsenic-rich glasses can be assigned to the presence of As4Se4 and As4Se3 cage-like molecules.

5.1.5 Structure of AspSeq clusters and of As-Se glasses

It has been reported that As-Se glasses are built up from a network of As-As, As-Se, or Se-Se bridges, strongly depending on composition. For example, it was suggested by Popescu et al.164 that the network of As-Se amorphous ChG’s may be made up by cluster polymerization with multiple Se-Se bridges. The same authors also explained anisotropy in a matrix of Se-Se bridged clusters on the basis of the breaking and directionally linking of clusters. The selenium-rich As- Se glasses were supposed to possess a typical cross-linked chain structure165 with As-As homopolar defect bonds and Se-Se-Se fragments.166 Mateleshko et al.167 demonstrated the structure and vibrational spectra of some As-Se clusters using Raman data of As40Se60 glass. The experimental Raman data and DFT were used to make models of clusters such as AsSe, linear 168 + As2Se, triangular As2Se, As3Se, AsSe3 and As4Se (square), for example. The cations As3S4 + + 169 and As3Se4 have been shown to be present in salts, e.g. {(As3Se4 ) (SbF6)}. It was observed + - - - in this work that As3Seq (q = 1-5), AsSeq (q = 1-3), As2Seq (q = 2-4), and As3Seq (q = 2-5) singly charged clusters are present in plasma plume formed from As-Se glasses. 24,169 In analogy with the published structures, we propose structures for the As3Se3 and As3Se4 clusters (Figs. 18(A) and 18(B)) and we suggest they are structural units in the bulk glass. The hypothetical complete structure of As-Se glasses, as proposed in Figure 19, is based on knowledge of the stoichiometry of clusters generated in the plasma plume, on the analysis of

Raman spectra indicating some structural entities such as As2Se2, As3Se, and AsSe3, for example, and also on evidence for the presence of As–As, As–Se, Se–Se, and Se– Se–Se bonds in the studied glasses. The percentage of the bonds and the percentage of the co-ordination numbers for As and Se atoms in a hypothetical As-Se glass structure (Fig. 19) and a comparison with the theoretical calculations by Mauro et Figure 18. Proposed structures concerning As Se and As Se clusters: al. (ab initio and Monte 3 3 3 4 (A) isomers of As3Se3 and (B) isomers of As3Se3. The structures given Carlo simulation)170 are (B) were optimized using the Hyperchem program. given in Table 5. Good agreement with the theoretical calculations by Mauro et al. was found. The percentages of the co-ordination numbers for As and Se atoms are also in acceptable agreement except for the co-ordination number 1 of the selenium atom. Better agreement with Mauro et al. can be obtained by also taking probable interconnecting bonds.

Figure 19. Hypothetical structural motifs in As-Se glasses. Possible laser-caused fragmentation to AspSeq clusters as those detected via TOF MS is marked.

Table 5. Percentage of the bonds and percentage of the co-ordination numbers for As and Se atoms in hypothetical As-Se glass structure (Fig. 19), a comparison to Mauro et al. theoretical (ab initio and Monte Carlo simulation) calculations.170

Percentage of bonds Percentage of co-ordination number As Se Co-ordination numbers References As-As As-Se Se-Se 1 2 3 4 1 2 3 4 This work (Fig. 19)* 16 70 13 - 56 39 22 - 35 56 4 This work (Fig. 19)** 10 90 - - 22 78 17 - 74 13 13

170 3 97 - 2 18 60 20 20 50 25 5 * calculated for the structure (Fig. 19) ** calculated assuming that all 5 heterocyclic structures are the same as the central one (Fig. 19)

5.2 Ternary glasses As-S-Se

Powders of glasses with a different molar ratio of components such as As33S33.5Se33.5,

As33S17Se50, and As33S50Se17 were analyzed. The positively and negatively charged clusters of

AspSqSer were generated via LDI and analyzed using TOF MS. The negative and positive ion mode threshold laser energy during the laser ablation/ionization was estimated at 50 a.u. Mass spectra with sufficient mass resolution and intensity were Figure 20. Effect of laser energy on cluster formation during laser observed at laser energies ablation of As-S-Se glass. Conditions: linear positive ion mode, equal to 60–90 a.u. in laser energy 50, 60, 70 and 80 a.u. Sample: AAs33S17Se50 glass. The m/z values given in the spectrum at the highest laser energy negative and 60–70 a.u. in correspond to the peak maximum of the corresponding cluster isotopic envelope. positive ion modes. An example how the positive ion mode spectra changes with increasing laser energy is given in Figure 20. It follows from this figure, an example of the LDI of a glass with a composition

As33S17Se50, that at the lowest laser energy (50 a.u.) only binary AspSq clusters are formed and no selenium-containing species are observed (as such a laser energy is not sufficient to ionize selenium). At higher laser energies the peak intensities increase and ternary clusters are also formed while at the highest laser energy (80 a.u.) several ternary AspSqSer clusters are formed – as identified later (Fig. 23). Similar changes were observed in negative ion mode.

5.2.1 Negative ion mode

The spectra in negative ion mode show the formation of several singly charged clusters (Figs. 21(A,B,C)). These were analyzed using isotopic envelopes and compared with computer models of theoretical isotopic patterns. Due to the natural abundance of sulfur and selenium isotopes, the

Figure 21. Comparison of mass spectra of As33S17Se50 (A), As33S33.5Se33.5 (B), and As33S50Se17 (C) glasses. Conditions: linear negative ion mode, laser energy 80 a.u. clusters containing sulfur and selenium show several peaks forming a so-called ‘isotopic envelope’. These were compared with computer models of theoretical isotopic patterns calculated using Launchpad software. An example of such a comparison is given in Figure 22.

The spectra of the As33S17Se50 glass mostly show the formation of the same binary and ternary cluster species as observed in the mass spectra generated from As33S33.5Se33.5 and As33S50Se17 glasses (Figs. 21(A,B,C)). However, the As33S50Se17 glass also shows the formation of some binary and ternary cluster species which are not observed in the As33S17Se50 glass; these were:

As2S3ˉ, As2S4ˉ, As3S3ˉ, As3S4ˉ, As3S5ˉ, AsS4Seˉ, As2S3Seˉ, As3S4Seˉ, and As3S3Se2ˉ. The intensities of the clusters in the LDI plume vary according to the composition of the glass. For + + + - example, the intensities of some cluster ions such as As3SSe2 , As3SSe3 , As3Se4 , AsSSe , - - - AsSe2 , AsSSe2 , and As3SSe2 in mass spectra from the ablation of As33S33.5Se33.5, As33S17Se50, and As33S50Se17 glasses increase with increasing content of selenium.

The intensities of AsS3ˉ,

AsS2Seˉ, As2S3ˉ, As2S2Seˉ,

As3S4ˉ, As3S2Seˉ, and As3S3Seˉ clusters decrease with increasing

S:Se ratio in the order As33S50Se17

> As33S33.5Se33.5 > As33S17Se50. Furthermore, the intensities of

AsSe2ˉ, AsSSe2ˉ, As2Se2H3ˉ,

As3SSe3ˉ clusters decreased in the Figure 22. Comparison of experimental and theoretical order As S Se > As S Se 33 50 17 33 33.5 33.5 mass spectra of As33S17Se50 glass. (A) Experimental spectra > As S Se . It was observed (Conditions: reflectron negative ion mode, laser energy 80 a.u.). 33 17 50 (B) Model of the isotopic envelopes assuming the formation of that increased sulfur content in the clusters. glass increases the number of binary and ternary clusters. - - - Hydrogenated As2Se2H3 , As2Se2H5 and As3Se2H3 clusters were observed from the mass spectra of the As33S17Se50 glass (Fig. 21(A)) and some traces of hydrogenated clusters were also observed in the mass spectra of the As33S33.5Se33.5 and As33S17Se50 glasses. Traces of As3S6ˉ,

As3S5ˉ, As3S4Seˉ, As3S5Seˉ clusters were observed in the mass spectra of As33S50Se17 and these species were also observed in the mass spectra of the As33S33.5Se33.5 and As33S17Se50 glasses. The

As3S2Se2ˉ cluster is observed at nearly the same intensity in the mass spectra of the As33S50Se17 and As33S33.5Se33.5 glasses.

The As3S3Se2ˉ cluster was observed in the mass spectra of the As33S50Se17 and As33S33.5Se33.5 glasses but not in the spectra of the As33S17Se50 glass. The As3S5Se2ˉ cluster was observed in the mass spectra of the As33S17Se50 glass but not in the mass spectra of the As33S50Se17 or

As33S33.5Se33.5 glass.

5.2.2 Positive ion mode

A comparison of the mass spectra of As33S17Se50, As33S50Se17, As33S33.5Se33.5 is given in Figure 23. It can be seen that the different molar ratio of elements in the As-S-Se glasses leads to some + + + + + differences in the spectra. Formation of the clusters As3S , As3Se , As3Se2 , As3S2Se , As3S3Se , + + + As3SSe2 , As3S2Se2 , and As3SSe3 was observed. In addition to the AspSq, AspSeq, AspSqSer

49 clusters some hydrogenated clusters were also observed. An overview of all the clusters identified in the positive and negative ion modes from the As33S17Se50, As33S50Se17,

As33S33.5Se33.5 glasses is given in Table 6. More ternary clusters were observed in the negative ion mode than in the positive one.

Figure 23. Comparison of mass spectra of As33S17Se50, As33S50Se17, As33S33.5Se33.5 glasses. Conditions: linear positive ion mode, laser energy 80 a.u.

Table 6. (A) Species common to all chalcogenide glasses generated by laser ablation of ternary chalcogenide As33S33.5Se33.5, As33S17Se50, As33S50Se17 glasses.

Number of Common clusters identified arsenic atoms Binary Ternary

1 AsS2ˉ, AsS3ˉ, AsSe2ˉ, AsSe3ˉ AsSSeˉ, AsS2Seˉ, AsSSe2ˉ,

AsS4Seˉ +/ 2 As2S3ˉ, As2S4ˉ, As2Se3 ˉ As2S2Seˉ, As2SSe2ˉ, As2S2Se2ˉ,

As2S3Seˉ + +/ +/ +/ +/ + 3 As3S , As3S3 ˉ, As3S4 ˉ, As3SSe2 ˉ, As3SSe3 ˉ, As3S2Se , + + +/ +/ As3S5ˉ, As3S6ˉ, As3Se , As3Se2 , As3S2Se2 ˉ, As3S2Se3ˉ, As3S3Se ˉ, + +/ As3Se3 , As3Se4 ˉ As3S3Se2ˉ, As3S4Seˉ, As3S4Se2ˉ,

As3S5Seˉ + 5 As5S3 +/- means the species was observed in both positive and negative ion mode.

(B) Overview of ternary clusters from individual glasses.

As33S33.5Se33.5 As33S50Se17 As33S17Se50 Positive cluster species + + + + + + As3SSe As3SSe2 As3SSe2 As3SSe3 As3SSe As3SSe2 + + + + + + As3S2Se As3S2Se2 As3S2Se As3S2Se2 As3SSe3 As3S2Se + + + + + + As3S3Se As4S2Se As3S3Se As2S6Se2H2 As3S2Se2 As3S3Se + + + + As3SSeH As3SSe3H As3S5Se2 As3SSeH2 + + As4SSe2H3 AsS12SeH2

Negative cluster species

AsSSeˉ AsSSe2ˉ AsSSeˉ AsSSe2ˉ AsS2Seˉ AsSSeˉ AsSSe2ˉ AsS2Seˉ

AsS2Seˉ As2SSe2ˉ AsS4Seˉ As2S2Seˉ As2S2Se2ˉ AsS5Seˉ As2SSe2ˉ As2S2Seˉ

As2S2Seˉ As2S2Se2ˉ As2S3Seˉ As3SSe3ˉ As2S2Se2ˉ As2S5Seˉ

As2S3Seˉ As3SSe2ˉ As3S2Seˉ As3S2Se2ˉ As3SSe2ˉ As3SSe3ˉ

As3SSe3ˉ As3S2Seˉ As3S3Seˉ As3S3Se2ˉ As3SSe4ˉ As3S2Seˉ

As3S2Se2ˉ As3S2Se3ˉ As3S4Seˉ As3S4Se2ˉ As3S2Se2ˉ As3S2Se3ˉ

As3S3Seˉ As3S3Se2ˉ As3S5Seˉ As3SSeH4ˉ As3S3Seˉ As3S5Seˉ

As3S4Seˉ As3SSeH5ˉ As3SSe2H4ˉ As3S5Se2ˉ As4S6SeHˉ

As3SSeH6ˉ As4SSeH3ˉ

A review of all binary and ternary clusters generated via the laser ablation of As33S17Se50, + + + As33S50Se17, As33S33.5Se33.5 glasses is given in Table 7. The species As3SSe , As3Se2 , As3SSe2 , + As3Se4 are present only in traces, as shown in mass spectra of the As33S50Se17 glass (Fig. 24). It should be noted that at very low laser intensities the ionization of some traces of carbon containing compounds was observed which indicates that the ChG’s contained some traces of organic compounds of unknown origin.

Table 7. Review of binary and ternary AsmSnSeo clusters generated via laser ablation of As33S17Se50, As33S50Se17, As33S33.5Se33.5 glasses (films) and mass spectra recorded in the linear, reflectron positive and negative ion modes.

Highest number of sulfur or selenium atoms in the cluster 0 1 2 3 4 5 6 7 + + m = 1 AsH6 AsSH AsS2ˉ AsS3ˉ AsS4Seˉ AsS5Seˉ + AsSeH AsSe2ˉ AsSe3ˉ + AsSeH2 AsS2Seˉ + AsSeH3 AsSSe2ˉ + AsSeH4 AsSSeˉ + 2 As2S2Seˉ As2S3ˉ As2S4ˉ As2S5Seˉ As2S6Se2H2 As2S7ˉ

As2Se2H3ˉ As2S3Seˉ + As2SSe2ˉ As2Se3

+ + +/ +/ +/ 3 As3H As3S As3Se2 ˉ As3S3 ˉ As3S4 ˉ As3S5ˉ As3S6ˉ

+ + +/ + As3H4 As3Se As3Se2H3ˉ As3S3Se ˉ As3S4H6 As3S5Seˉ

+ +/ + + As3SSe As3S2Se ˉ As3S3Se2ˉ As3Se4H As3S5Se2 + +/ +/ As3SSeH As3SSe2 ˉ As3SSe3 ˉ As3S4Seˉ

+ As3SSeH2 As3SSe2H4ˉ As3SSe3Hˉ As3S4Se2ˉ +/ As3SSeH4ˉ As3S2Se2 ˉ

As3SSeH5ˉ As3S2Se3ˉ

As3SSeH6ˉ + + + + 4 As4H As4SH As4S2Se As4S3H5 + + + As4H2 As4SSeH3 As4Se2H3 + + As4H3 As4Se2H4 + As4Se2H5 + As4SSe2H3 + + 5 As5S3 As5S4H2 +/ˉ means the species was observed in positive and negative ion mode.

5.2.3 Analysis of nano-layers

Bulk glasses are used for the PLD manufacturing of nanolayers, important in electronics. The

PLD As33S50Se17 glass nano-layer is used as an example for the comparison of mass spectra for bulk and nano-layer analysis. The mass spectra of the nano-layer were measured in both positive and negative ion modes at different laser energies. The analysis shows that most of the mixed + chalcogenide clusters observed are the same as for the nano-layer (Fig. 24). The species SeS6 , + + As2S5H , and As4S were not observed when ablating nano-layer material. In addition, in negative ion mode similar mass spectra were observed for the nano-layer and the bulk material. From this comparison of mass spectra and the fact that most of the clusters are the same, we can also conclude that the structures of the bulk glasses and of the PLD nanolayer are the same or very similar.

Figure 24. Comparison of mass spectra for laser ablation of bulk material and PLD-produced glass

nano-layer: (A) bulk As33S50Se17 glass and (B) nano-layer of As33S50Se17 glass. Conditions: linear positive ion mode, laser energy 80 a.u.

5.2.4 Structure of binary and ternary clusters

With regard to the stoichiometry of clusters reviewed in Table 7, we should comment that MS cannot determine their structures. It has been reported that the ternary ChG’s (As-S-Se) are built up of a network of S-S, Se-Se or S-Se bridges. Popescu et al.164 suggested that the network of As-Se amorphous ChG’s may be made up by polymerization of the clusters with multiple bonding of Se-Se bridges. The same authors also explained anisotropy in a matrix of Se-Se bridged clusters on the basis of the breaking and directionally linking of clusters. Although a 171 + number of SqSer species has been described in the literature, we have observed only S6Se species. Mateleshko et al.167 demonstrated the structure and vibrational spectra of some As-Se + clusters by analyzing the Raman spectroscopy data of As40Se60 glass. The cations As3S4 ,

+ As3Se4 are present in + salts, e.g. {(As3Se4 ) - 169 (SbF6 )}. We have observed several series of clusters with + the ratio 3:4 (As3S4 , + + As3S3Se , As3S2Se2 , + As3SSe3 , and + + As3Se4 ), 3:3 (As3S3 , + As3S2Se , and + As3SSe2 ), 3:2 + + (As3S2 , As3SSe , + and As3Se2 ), 3:1 + + (As3S , and As3Se ). Figure 25. Proposed structures of clusters with the ratio arsenic:chalcogen = + + + + + + 3:3 (As3S3 , As3S2Se , As3SSe2 ) and 3:4 (As3S4 , As3S3Se , As3S2Se2 , In analogy to the + + As3SSe3 , As3Se4 ). The structures given were optimized using the Hyperchem published structures program. + + of As3S4 , As3Se4 cations we propose the probable structures of clusters (Fig. 25) of the 3:3 series (first three structures) and the 3:4 series (five structures given). There are two structural isomers for each of As3S2Se and As3SSe2 as well as three isomers for each of As3S3Se and

As3SSe3 clusters (isomers are not given). The question remains how such heterocycles are bound in the glasses studied and further research in this field is needed.

5.3 Rare earth (Er3+) doped quaternary glasses Ga-Ge-Sb-S

5.3.1 Chemical composition and amorphous state of prepared samples

EDX analysis shows that the chemical composition of prepared samples is in very good agreement with the nominal composition taking into account error limit of the used EDX method is ±0.5 at.%. However, the concentration of Erbium is too low to be analyzed by EDX but the expected variation of absorption coefficient versus concentration of Erbium was checked by optical transmission measurements. The amorphous state and homogeneity of the samples were

54 confirmed by optical/electron microscopy and XRD patterns.

5.3.2 Glass structure supported by Raman analysis

The physical properties of chalcogenide glasses are generally discussed - without any consideration on the species - considering the average coordination number (m). In the constraints model, a specific value of m = 2.67 was attributed to a structural phase transition

Ge25Ga5S70

Ge15Ga5Sb10S70

Ge20Ga5Sb10S65

Intensity (a.u.) Intensity Ge17Ga8Sb10S65

Ge24Ga1Sb10S65

Ge23Ga5Sb12S60

150 200 250 300 350 400 450 500 Wavenumber(cm-1) Figure 26. Raman spectra and assignment of Raman bands of Ga-Ge-Sb-S bulk glasses to specific vibration modes of structural units of the vitreous network. from a two-dimensional (2D) layered structure to a three-dimensional (3D) network.172 In

Ga5Ge20Sb10S65 glassy alloy and in glasses with close chemical composition, the average coordination number (m) was evaluated using the standard procedure described elsewhere and revealed a structure certainly really close to a model of a three-dimensional amorphous network.173 In Ga-Ge-Sb-S glasses, the gallium coordination number is assumed to be four as it 174,175 has been proposed for Ga-Ge-S or Ga2S3-GeS2- based glasses. For the α-Ga2S3 crystal, the fourth bond is a dative bond with a sulfur lone pair and thus a part of the sulfur atoms should be

55 three-fold coordinated. Nevertheless, such a structural organization could be expected in nonstoichiometric glasses but it has not been experimentally demonstrated up to now. Instead, it is usually proposed the formation of homopolar bonds like (S3-xGex)Ge-Ge(S3-xGex) or involving Ga atoms to overcome the deficit in sulfur.176 The Raman spectrum of (Ga)-Ge-(Sb)-S glasses is dominated by the presence of the band located at 336 cm-1, related to the symmetric stretching 174,177 modes of [GeS4/2] tetrahedral (Fig. 26). The bands related to the three-dimensional structure of the glassy network can also be observed: the band (shoulder) at 370 cm-1 is associated to the c c  1(A 1) "companion" mode of the ν1 mode, linked to the vibrations of tetrahedra bound by their edges. -1 The band located at 433 cm is related to the vibration of S3Ge-S-GeS3 structural units where tetrahedra are connected by their corners. The presence of a weak Raman band observed at 477 -1 cm corresponding to the vibrations of S-S bonds is obviously observed in Ge25Ga5S70 and 178 Ga5Ge15Sb10S70 glass with a clear excess of sulphur for both of them. At the low wavenumbers side, the shoulder at 293 cm-1 is attributed to the symmetric stretching vibration modes of

[SbS3/2] pyramids. Comparing Ga1Ge24Sb10S65 with Ga8Ge17Sb10S65, the location of symmetric -1 stretching modes of [GaS4/2] tetrahedra can be assumed to be around 320 cm . In the case of -1 Ga5Ge20Sb10S65 glass, a weak band around 250-260 cm is assumed to be partially covered by the shoulder previously described. That band is generally associated to the stretching vibration of the Ge-Ge homopolar bonds in the S3Ge-GeS3 units clearly observed on the spectra of 177 Ga3Ge25Sb12S60. Typically, the spectra of S-deficient glasses show some Raman features in the range of 150-250 cm-1, related to the formation of homopolar bonds. Thus, the three bands at -1 163, 208 and 260 cm can be connected with vibrations of Sb-Sb homopolar bonds, (S3-xGex)Ge- 179 Ge(S3-xGex) (x>0) or Ge-Sb dissimilar bond and S3Ge(Ga)-Ge(Ga)S3 units, respectively. In the glassy matrix, the rare earth ions seem to be surrounded by 6 or 7 sulfur atoms,180-182 incorporated on sites close to the gallium atoms (Ga-S-TR) in order to balance the partial 183 negative charge of tetrahedral units [GaS4]. The introduction of Ga in Ge-As-S glasses greatly enhanced rare-earth solubility and dispersal, particularly for Ga : rare earth ratios ≥ 10:1.184 Note that the study on glasses doped with erbium sulphide suggests that the Er3+ ion occupies only one type of site in the glassy matrix.185

5.3.3 Mass spectrometry

Powders of Ga5Ge20Sb10S65 glass doped with different concentrations of erbium (0.05, 0.1, and 0.5 w.%) were analyzed. Mass spectra were recorded by TOF MS and singly charged positive and negative GamGenSboSp cluster ions were generated via LA or LDI. The threshold laser energy was found to be almost the same in positive and negative ion mode, at ~50 a.u. Mass spectra with sufficient mass resolution (>250) and intensity (>25 mV) were observed at laser energies in the interval 70-130 a.u. in positive and 70-160 a.u. in negative ion modes. Clusters were generated in plasma plume and identified in measured mass spectra up to m/z 1050 in positive ion mode and m/z 1010 in negative ion mode. The mass spectra were analyzed via comparison of experimental isotopic envelopes with the theoretical models and in this way the stoichiometry of the clusters was determined. Even the cluster ions were generated up to m/z +/– ~1000 in both ion modes but surprisingly, we have identified only four species Sb3S4 , +/– +/– GamSboSp (m =1, o = 2, p = 4,5), Ga3Sb2S7 common to both ion modes. In comparison with +/– GamSboSp species, formation of germanium containing species was lower in number, overall survey from mass spectra in negative ion mode show higher species (15) than positive ion mode + + (5). Few species (GaSb2SEr and GaS6Er2 ) of erbium were identified in positive ion mode only.

5.3.4 Effect of laser energy

The effect of laser energy on mass spectra of all glasses was followed. As an example, the mass spectra of Ga5Ge20Sb10S65 glass doped with erbium (0.05 w.%) at different laser energies are shown in Figure 27(A). It was observed that at lower laser energy (90 a.u.) GamGenSboSp clusters were embarking on. On increasing the laser energy, the intensity of the peak increases but the resolution of the mass spectra falls at laser energies of 150 a.u. and higher. Enlarged part of spectra (Fig. 27(A)) for higher mass range from 500-1200 is shown in Figure 27(B). Laser 3+ energy effect for highest concentration (0.5 w.%) of Er ion in Ga5Ge20Sb10S65 glass is shown in Figure 27(C). Formation and proliferation of peaks are observed from lower to higher mass range with increasing laser energy. Several peaks were observed at laser energy 90 a.u. in higher mass range 800-1000 a.u. whilst such peaks were not observed at lower laser energy. In Ga-Ge- Sb-S glass, species in lower mass range 1-800 ionize at laser energy 50-70 a.u., however it was observed that same sample with same experimental conditions starts ionization of species in

57 higher mass range 800-1050 at laser energy 80 a.u. and sufficient peaks intensity and resolution were achieved at 90 a.u. ‘Sweet spot’ of deposited sample on the target plate could be the reason for such ionization.

Figure 27. Full scale mass spectra show laser energy

effect for Ga5Ge20Sb10S65 glass doped with erbium (0.05 w.%). (A) Mass range 0-1200; the spectra were normalized to 1182 mV. (B) Mass range 500-1200; the spectra were normalized to 200 mV. Conditions: linear positive ion mode, laser energy 90, 110, 130, and 150 a.u. (C) Full scale mass spectra show laser energy

effect for Ga5Ge20Sb10S65 glass doped with erbium (0.5 w.%) Conditions: linear positive ion mode, laser energy 50, 60, 70, 80, 90 a.u., mass spectra were normalized to 2100 mV.

5.3.5 Positive ion mode

Mass spectra were measured and analysed for three samples (erbium doping 0.05, 0.1, 0.5 w.%) of Ga5Ge20Sb10S65 glass. Analyses of all glass samples show similar species generated in the plasma plume via the action of laser. Similar results were identified to all samples of glass. An example of mass spectra for 0.1 w.% Er3+ at laser energy 90 a.u. is shown in Figure 28(A,B). The +/– stoichiometry of the GamGenSboSp clusters was determined via isotopic envelope analysis and + + computer modeling. For all glasses Ga cluster was found to be the main species and Sb3S was + + found to be the second main species. Nine series of clusters, GamSp (m = 3, p = 1,2), SboSp (o = + + + 3, p = 1-4), GamSboSp (m = 1, o = 1, p = 1,2), GamSboSp (m = 2, o = 1, p = 2,3), GamSboSp (m + + = 1, o = 2, p = 2-5), GamSboSp (m = 2, o = 2, p = 4,5), GamSboSp (m = 3, o = 2, p = 4-7), + + GamSboSp (m = 2, o = 3, p = 5-7), and GamSboSp (m = 1, o = 4, p = 6,7) were identified. Six

+ + + species were identified with overlap of another species: GaSbS [GaSbSH (~30%)], GaSb2S5 + + + + + [Ga5S4 (~80%)], ErGaSb2S [Ga2Sb2S4 (~20%)], Ga4SbS5 [GaSb3S4 (~60%)], + + + + + Ga3GeSb2S3H4 [Ga8GeH3 (~20%)], Ga2Sb3S5 [Ga5SbS6H (~40%)]. Up to m/z 300 unary Ga + and binary GamSp (m = 3, p = 1, 2) species were found to be the most abundant. Two species + SboSp with 1:1 elemental proportions and their peaks intensity were found to be almost similar. + Binary SboSp (o = 3, p = 1-4) species with odd number of sulfur atoms were identified with prolific peaks than the peaks of even number of sulfur atoms. At lower mass, two species GaSbS+, GaGeS+ were detected with 1:1:1 elemental proportion and almost with similar peaks intensity. Species of GaSbS+ series with high number of sulfur atoms was found with decreasing + peak intensity. Species of GamSboSp (m = 2, o = 1, p = 2-4) series with two and three atoms of + sulfur were identified with increasing intensity of peaks. Series GamSboSp (m = 1, o = 2, p = 2-5) was identified with increasing peaks intensity of even number of sulfur atoms (2,4) and decreasing peaks intensity with odd number of sulfur atoms (3,5).

Figure 28. Full scale mass spectra show results of analysis for Ga5Ge20Sb10S65 glass doped with erbium (0.1 w.%). (A) Mass range up to m/z 500. (B) Mass range over m/z 500. Conditions: linear positive ion mode, laser energy 90 a.u.

+ In comparison with odd number of sulfur containing species in GamSboSp (m = 1, o = 2, p = 2- 5) series, even number of sulfur containing species was found with prolific peak intensity. At + higher mass range 500-1050, the main species Ga3Sb2S7 was detected (Fig. 28(B)). + Interestingly, two clusters with difference of one sulfur atom GamSboSp (m = 3, o = 2, p = 4,5) were identified with similar peak intensities. In addition, similar peak intensities were identified + for second main species GamSboSp (m = 2, o = 3, p = 6,7) with difference of one sulfur atom. + + Two species Ga3GeS15 and Ga4SbS16 are rich in sulfur and detected at higher m/z. Selected

59 parts of experimental mass spectra measured for Ga5Ge20Sb10S65 glass doped with erbium (0.1 w.%) and comparison with Experiments 100 theoretical model at lower, middle, and higher m/z values among whole 50 range of experimental mass spectra are shown in Figure 29. 0 + + + GaGeS Sb3S4 Ga4Sb3S9 174.8 492.6 Models 932.2 Interestingly, Ga5Ge20Sb10S65 glass 100 was doped with different 50

concentration (0.05, 0.1, 0.5 w.%) Relative intensity [%] of Er3+ ion but analyses of all 0 171 175 179 490 494 498 926 932 938 samples show only two species m/z + + Figure 29. Selected parts of experimental mass spectra GaSb2SEr , GaS6Er2 of erbium at measured for Ga5Ge20Sb10S65 glass doped with erbium high mass range. There was no (0.1 w.%) and comparison with theoretical model at lower, middle, and higher m/z values among whole range erbium species at low mass range. of experimental mass spectra. Conditions: linear positive + Ternary GamSboSp species of Ga, ion mode, laser energy 90 a.u. + + Sb, and S were found with higher in number. Only two species Ga2GeSbS6 and Ga3GeSb2S3H4 with all elements of Ga5Ge20Sb10S65 glass were identified. Some un-interpreted peaks at lower and higher mass range 80-160, 800-1050, respectively, were complicated to resolve. Summary of +/– positively and negatively charged GamGenSboSp species identified in mass spectra of

Ga5Ge20Sb10S65 glass doped with erbium- 0.05, 0.1, 0.5 w.% is given in Table 8.

+/– Table 8. Summary of positively and negatively charged GamGenSboSp species identified in mass spectra of Ga5Ge20Sb10S65 glass doped with erbium- 0.05, 0.1, 0.05 M.

Number of sulphur atoms in cluster 0 1 2 3 4 5 6 7 9 10 15 16 + + + +/– m = 0 Sb3S Sb2S2 Sb3S3 Sb3S4 + Sb3S2 + + + + +/– +/– + + 1 Ga GaGeS GaSbS2 GaSb2S3 GaSb2S4 GaSb2S5 GaS6Er2 GaSb4S7 + + + + + GaSbS GaSb2S2 GaSb3S3H GaSb3S4 GaSb4S6 + + GaSbSH GaSb2SEr + + + + + + 2 Ga2S2 Ga2SbS3 Ga2Sb2S4 Ga2Sb2S5 Ga2GeSbS6 Ga2Sb3S7 + + + Ga2SbS2 Ga2Sb3S5 Ga2Sb3S6 + + + + + + +/– + + 3 Ga3S Ga3S2 Ga3GeSb2S3H4 Ga3S4H4 Ga3Sb2S5 Ga3Sb2S6 Ga3Sb2S7 Ga3Sb4S9 Ga3GeS15 + Ga3Sb2S4 + + + + + + 4 Ga4Sb5S4 Ga4SbS5 Ga4Sb2S6H Ga4Sb3S9 Ga4Sb3S10 Ga4SbS16 + + + 5 Ga5S4 Ga5SbS6H Ga5Sb2S9 + 6 Ga6Sb2S9 + 8 Ga8GeH3 m indicate number of gallium atoms in GamGenSboSp clusters. +/– indicate that the species was observed in both positive and negative ion mode.

5.3.6 Negative ion mode

In comparison with the positive ion mode, ionization in linear and/or reflectron negative ion mode leads to the generation of a slightly higher number of clusters. Mass spectra were measured and analysed for three samples (erbium doping 0.05, 0.1, 0.5 w.%) of Ga5Ge20Sb10S65 glass. Similar results were identified to all samples of glass. An example of mass spectra for 0.5 w.% of 3+ +/– Er at laser energy 80 a.u. is shown in Figure 30(A,B). The stoichiometry of the GamGenSboSp clusters was determined via isotopic envelope analysis and computer modeling. For all glasses, ‾ ‾ GaS2 species was found to be the main species and SbS2 was found to be the second main ‾ ‾ ‾ species. Fifteen series of clusters, Sp (p = 1-3), SboSp (o = 1, p = 1-3), GamSp (m = 1, p = 2,3), ‾ ‾ ‾ ‾ GamSp (m = 3, p = 3,4), GamSbo (m = 4,5, o = 2), GenSboSp (n = 1, o = 1, p = 3-5), GamGenSp (m ‾ ‾ = 1, n = 1, p = 3-5), SboSp (o = 3, p = 4,5), GamGenSboSp (m = 1, n = 1, o = 1, p = 4,5), ‾ ‾ ‾ GamSboSp (m = 2, o = 1, p = 4,5), GamSboSp (m = 1, o = 2, p = 4-6), GamSboSp (m = 1, o = 3, p = ‾ ‾ ‾ 5,6), GamSboSp (m = 2, o = 2, p = 6,7), GamSboSp (m = 3, o = 2, p = 7,8), and GamSboSp (m = 5,

61 o= 2, p = 10,11) were identified. GaS ‾ SbS ‾

Fourteen of the species 100 2 2 A

‾

3 3 4 4

identified were found to overlap ‾

‾ 2 2 SbS ‾ 4

‾ 3

‾

GaSbS

GaGeS ‾

with another one: GaSbS3 with 5

4 4

‾

GeSH

5 5 GeSbS

‾ ‾ SbS

SbS ‾

GeSbS (~60%), Ga S with 2

3 3 3 Ga

3 3

‾

‾ 5 5

50 4

‾

S Ga

‾ ‾

2 2

‾ ‾ Sb GeSbS

S 3 ‾

GaGeS5 (~70%), Sb2S3 with 3 2

3 3

‾

SbS SbS

GaGeS

‾

GaS

2 3

‾ ‾ 3

GaSb

4 4

‾

Sb S

Ga3S4 (~90%), Ga2SbS4 with 3

Ge Ga ¯ 3 S2

‾ Ga Relative intensity [%] ‾ ‾ Sb

S SbS GaGeSbS4 (~50%), Ga2SbS5 ‾ 0 with GaGeSbS5 (~40%), 100 200 300 400 500 m/z ‾ ‾ Ga GeS with GaSb S (~40%), ‾ 3 7 2 6 Ga3Sb2S8 ‾ ‾ Ga Sb S ‾ B Sb S with Ga Sb (~30%), 6 2 ‾

3 5 4 2 100 ‾

11 11

‾

S S

8 8

‾

5 5

‾

7 7 S

‾ ‾ S

2 2

3 S

Ga GeSb with Ga Sb S Sb

2 3 2 2 6 2

‾ 4 ‾

‾ Ga Sb S

SbS

5 5 GeSb

Sb 5 2 11

‾

3 3

‾ ‾ Ga

‾

3 3

GaSb

‾ 4

(~75%), GaSb S with Ga Sb S ‾

3 5 5 2 ‾

‾

4 4

‾

5 5

3 3

5 5

11 11

GaGeSb

16 16

‾

S S

‾

Sb 7 7

‾ ‾ S

‾

2 2

5 2

(~90%), Sb with Ga Sb S 8

5 2 2 7 ‾

SbS

2 2

3 GeSb

50 Ga

GeS

GaSb ‾

‾ 3

SbS

2 2

‾

SbS Ga

(~20%), Ga Sb SH with 9 Sb

5 2 2 S

3 3

4 4

10 10

‾

10 10 Ge

‾ ‾ Sb

Sb Sb

GaSb S (~70%), Ga S with 4 4

3 6 5 16 Sb

Relative intensity [%]

Ga Ga ‾ ‾ Ga

Ga4Sb4S3 (~50%), GaGeS4 with Ga GaSb ‾ ‾ 0 Ga2S4H (~40%), and Ge3SH3 600 700 800 900 1000 ‾ ‾ m/z with SbS4H4 (~25%). In GamSp Figure 30. Full scale mass spectra show results of analysis for Ga Ge Sb S glass doped with erbium (0.5 w.%). (m = 1, p = 2,3) and Sb S ‾ (o = 5 20 10 65 o p (A) Mass range up to m/z 500. (B) Mass range over m/z 500. ‾ 1, p = 1-3) series, species GaS2 Conditions: linear negative ion mode, laser energy 80 a.u. ‾ and SbS2 have 1:2 (Ga:S or Sb:S) stoichiometries, mass spectra of such species show highest peak intensity. Such species could be considered as stable units in the structural network of Ga- Ge-Sb-S glasses and it might reveal information about Ga-Ge-Sb-S glass built by binary-stable structural motifs (GaS2 and SbS2) as a dominant unit. However, puzzle still unsolved and raise the question ‘are such motifs being a part of original structure of glass or are they synthesized in ‾ ‾ plasma reactions? In GamGenSp (m = 1, n = 1, p = 3-5) series, species GaGeS4 is the main one. ‾ Only one series GamGenSboSp (m = 1, n = 1, o = 1, p = 4,5) was identified with all elements of ‾ Ga-Ge-Sb-S glass. At lower mass (up to 500), one species (Ge3SbS ) with maximum three atoms ‾ of germanium, one species (Sb3S4 ) with maximum three atoms of antimony, one species ‾ ‾ ‾ (Ga5GeSH2 ) with maximum five atoms of gallium, and three species (GeSbS5 , Ga2SbS5 ,

‾ GaSb2S5 ) with maximum five atoms of sulfur were identified. Up to mass 220 unary (only sulfur) and binary (only gallium and antimony) clusters were identified, there was no identification of germanium clusters at low m/z. We can suggest that ionization and formation of unary clusters of sulfur and binary clusters of gallium-sulfur and antimony-sulfur at lower m/z were seems to be most favourable (up to m/z 220) than the formation of ternary clusters whilst the formation of few binary gallium-sulfur and antimony-sulfur clusters was also possible/generated at higher m/z. Spectrum of higher mass range 500-1010 is presented in Figure ‾ ‾ 30(B). Species Ga3Sb2S8 was detected as a main species and Ga6Sb2S as a second main species ‾ ‾ ‾ in higher mass range of spectra. Three species Ga3GeSb2S , Ga2Ge3SbS2 , and GaGeSb2S8 were ‾ identified with all elements of Ga-Ge-Sb-S glass. Among them two species Ga3GeSb2S , ‾ Ga2Ge3SbS2 show exchange of stoichiometry 1, 2, 3 for atoms Ga, Ge, Sb, and S as well. Unary ‾ ‾ ‾ Sb5 was found to be the third main species. Binary four species GamSbo (m = 4,5, o = 2), Sb3S5 , ‾ and Ga5S16 were identified. Interestingly, all series show sulfur atoms, however binary one ‾ series show only gallium-antimony GamSbo (m = 4,5, o = 2) atoms. One ternary species ‾ Ga2Ge2Sb of 2:2:1 stoichiometry was identified with absent of sulfur atoms. Five germanium ‾ containing species were identified. At higher mass (up to 1010), one species (Ga8SbS2 ) with ‾ maximum eight atoms of gallium, one species (Ga2Ge3SbS2 ) with maximum three atoms of germanium, one species ‾ (GaSb6S4 ) with maximum six atoms of antimony, and one ‾ species (Ga5S16 ) with maximum sixteen atoms of sulfur were identified. Selected parts of experimental mass spectra measured for

Ga5Ge20Sb10S65 glass doped with erbium (0.5 w.%) and comparison with theoretical model at lower, middle, and higher m/z values among whole range of experimental

‾ mass spectra are shown in Figure 31. Summary of negatively charged GamGenSboSp species identified in mass spectra of Ga5Ge20Sb10S65 glass doped with erbium- 0.05, 0.1, 0.5 w.% is given in Table 9.

‾ Table 9. Summary of negatively charged GamGenSboSp species identified in mass spectra of

Ga5Ge20Sb10S65 glass doped with erbium- 0.05, 0.1, 0.5 M.

Number of sulphur atoms in cluster 0 1 2 3 4 5 6 7 8 9 10 11 16 ‾ ‾ ‾ ‾ ‾ ‾ m = 0 Sb5 S S2 S3 Sb3S4 Sb3S5 ‾ ‾ ‾ ‾ ‾ SbS SbS2 SbS3 GeSbS4 GeSbS5 ‾ ‾ Ge3SbS Sb2S3 ‾ GeSbS3 ‾ ‾ ‾ ‾ ‾ ‾ ‾ 1 GaS2 GaS3 GaGeS4 GaGeS5 GaSb2S6 GaGeSb2S8 GaSb4S10 ‾ ‾ ‾ ‾ GaGeS3 GaGeSbS4 GaGeSbS5 GaSb3S6 ‾ ‾ ‾ GaSbS3 GaSb2S4 GaSb2S5 ‾ ‾ GaSb6S4 GaSb3S5 ‾ ‾ ‾ ‾ ‾ ‾ ‾ ‾ 2 Ga2Ge2Sb Ga2Ge3SbS2 Ga2Sb5S3 Ga2SbS4 Ga2SbS5 Ga2Sb2S6 Ga2Sb2S7 Ga2Sb3S8 ‾ ‾ ‾ ‾ ‾ ‾ 3 Ga3GeSb2S Ga3S3 Ga3S4 Ga3Sb5S5 Ga3GeS7 Ga3Sb2S8 ‾ ‾ ‾ Ga3Sb4S Ga3Sb4S3 Ga3Sb2S7 ‾ ‾ ‾ ‾ ‾ 4 Ga4Sb2 Ga4Sb4S3 Ga4SbS8 Ga4Sb2S9 Ga4Sb3S11 ‾ ‾ ‾ ‾ ‾ ‾ ‾ 5 Ga5Sb2 Ga5GeSH2 Ga5Sb3S2 Ga5Sb4S4 Ga5Sb2S10 Ga5Sb2S11 Ga5S16 ‾ Ga5Sb2SH2 ‾ ‾ 6 Ga6Sb2S Ga6SbS11 ‾ 7 Ga7Sb2S3 ‾ 8 Ga8SbS2 m indicate number of gallium atoms in GamGenSboSp clusters.

Unary 4, binary 13 species in negative ion mode and unary 1, binary 9 species in positive ion mode were detected. There was no sulfur (unary) species identified in positive ion mode. Three ‾ ‾ sulfur species So (o = 1-3) and one antimony species Sb5 were identified in negative ion mode. Review of coincided species in mass spectra of positive and negative ion modes is given in Table 10. It seems that overlapping of ternary-ternary species was found with slightly higher numbers in both ion modes. However, overlapped species with unary-ternary and binary-ternary combination were unidentified in positive ion mode. Species of erbium were not identified in + negative ion mode. Ternary GamSboSp species composed of Ga, Sb, and S elements were found

64 with higher in number for both ion modes.

Table 10. Survey and classification of unary, binary, ternary, and quaternary clusters with the contribution of another cluster ions identified in negative and positive ion modes.

Unary-ternary Binary-binary Binary-ternary Ternary-ternary Ternary-quaternary Negative ion mode ‾ ‾ ‾ ‾ ‾ Sb5 Sb2S3 Ga3S3 GaSbS3 Ga2SbS4 ‾ ‾ ‾ ‾ ‾ [Ga2Sb2S7 20 %] [Ga3S4 90 %] [GaGeS5 70 %] [GeSbS3 (60 %)] [GaGeSbS4 50 %] ‾ ‾ ‾ ‾ Sb3S5 GaSb3S5 Ga3GeS7 Ga2SbS5 ‾ ‾ ‾ ‾ [Ga4Sb2 30 %] [Ga5Sb2 90 %] [GaSb2S6 40 %] [GaGeSbS5 40 %] ‾ ‾ Ga5S16 Ga2GeSb3 ‾ ‾ [Ga4Sb4S3 50 %] [Ga2Sb2S6 75 %] ‾ Ga5Sb2SH2 ‾ [GaSb3S6 70 %] Positive ion mode + + + GaSb2S5 GaSbS ErGaSb2S + + + [Ga5S4 (80 %)] [GaSbSH (30 %)] [Ga2Sb2S4 (20 %)] + + Ga4SbS5 Ga3GeSb2S3H4 + + [GaSb3S4 (60 %)] [Ga8GeH3 (20 %)] + Ga2Sb3S5 + [Ga5SbS6H (40 %)] Species given in parenthesis show contribution with main species.

5.3.7 Germanium species and relationship with hydrogen

Germanium containing species in both ion modes were detected with less in number, the reason is that introduction of antimony in the glass system decreases the number of GeS4 units in the glass network.186 Five species of germanium content were identified in positive ion mode; two of + + them were found with traces of hydrogen. Four species of GamSboSp and one species of GamSp ‾ ‾ were identified with traces of hydrogen in positive ion mode. Three species Ge3SH3 , Ga5GeSH2 ‾ and Ga5Sb2SH2 were identified with traces of hydrogen in negative ion mode. Hydrogen is one of the main impurities in ChG’s.187,188 Although using 5N purity of elements Ga/Ge/Sb/S, strength of hydrogen bonds (covalent) with sulfur (S-H) and germanium (Ge-H) is stronger.

Usually for 2S2G glasses, the purification to vanish Hydrogen is mainly focused on sulphur purification. The proportion of [S-H] can vary from 15 ppm to 700 ppm depending on the level of purification which can affect strongly the luminescence efficiency of erbium ions.189 Detection limit of TOF MS is upto femtomole, thereby traces of hydrogen containing species were detected. It should be noted that spectra were also acquired in reflectron mode but no significant differences were observed in both ion mode.

5.3.8 Comparison of glass structure and identified clusters

Structural studies of Ga-(Ge)- (Sb)-S glasses using mainly Raman spectroscopy, EXAFS, EPR have been reported in the literature as we discussed previously. However, mass spectrometric study of Ga-Ge-Sb- S glasses is not reported yet. It was observed in this work that many singly charged clusters are present in plasma plume generated from Ga5Ge20Sb10S65 glasses doped with Er3+ ion. Several identified clusters are in agreement with structural network of concerning glass, though rest of all identified clusters and their series reveal the attention for further study. Several papers190,186 show glasses network contain

GaS4, GeS4, and SbS3 units, our

66 results are evidence for such units, we have identified only four species common to both ion

+/– modes for studied glasses. Among them Sb2[GaS4] species show GaS4 tetrahedral group, also ‾ ‾ the species Ga[GeS4] and SbS3 detected in negative ion mode are congruent to GeS4 tetrahedral ‾ ‾ ‾ group and SbS3 units, respectively. We suggest that sulphur species, for example S , S2 , S3 , ‾ ‾ GaS2 , and GaS3 generated in plasma plume can be the part of sulphur bridges in a glass ‾ structure. The most intensive peak for GaS2 species was identified. It means material (glass) is ‾ mostly decomposed to GaS2 species and such species S-Ga-S could be the main constituent of ‾ the glass. Similarly, third most intensive peak for SbS3 species, fourth intensive peak for ‾ ‾ [GaS4]Ge species in lower mass range (50-250), and third intensive peak for Ga2Sb2S7 species in higher mass range (500-700) were identified. This is in agreement with paper of Feng et al.190 ‾ that the main structural network in Ga-Sb-S glass is [Ga2Sb2S7]S2 formed by the combination of two GaS4 tetrahedra and two SbS3 trigonal pyramids, in which there exist two Ga2SbS3 rings. + ‾ ‾ ‾ Species Ga2SbS3 was also detected in positive ion mode. Species Ga2Sb2S7 , SbS3 , [GaS4]Ge , + and Ga2SbS3 identified in mass spectra are in agreement with the results of Feng et al., all such species are marked in tetranuclear heterometallic cluster of Ga2Sb2S9 (Fig. 32(A)). Clusters identified in the mass spectra of erbium doped Ga5Ge20Sb10S65 glasses, some of them are marked 2- in the 2D anionic network of [Ga2Sb2S7] , demonstrated by Feng et al. (Fig. 32(B)). Literature doesn’t show much information about the contribution of erbium in glass structure. To date, few papers describe the erbium could modify the structure of glass.191 We have identified only two + species of erbium, among them Er2GaS6 species are in better agreement with the species 191 3+ Er[Er2GaS6] explained in literature. The structural units Er[Er2GaS6] contain Er clusters which demonstrate strong quenching of Er3+ luminescence. In conclusion some of the species identified in the plasma could originate from part of the Ga5Ge20Sb10S65 structure (Fig. 32(B)). However, several of the species are not part of the structure and we suggest that they were generated by the action of the laser and synthesized in the plasma plume.

5.4 Phosphorus nitride P3N5

The first part of this section focuses on the mass spectrometric analysis of the products formed during the LDI of P3N5 and the second part shows the results of P3N5 ionization using matrices. The threshold laser energy in positive and negative ion modes during the laser

67 ablation/ionization was estimated to be ~120 a.u. Mass spectra with sufficient mass resolution (>400) and intensity (>2 mV) were observed at a laser energy of 120–150 a.u. in both ionization modes. During the evaluation of the spectra we concentrated only on those signals which could be identified as containing both phosphorus and nitrogen, i.e. PmNn clusters. The others, uninterpreted peaks, were considered as impurities, fragments of matrices or other and thus were not considered. In all the Figures (except Fig. 37) a relative intensity scale was used where 100% corresponds to the maximum intensity observed in the m/z range studied.

5.4.1 LDI of P3N5

An example of the spectra measured in positive ion mode is given in Figure 33(A). +

+ + + 2

A P4N P6N2 2 B 353.7 Experiment

+ + 5 H

+ + 100 +

P N 3

5 4 18

18 16

3 4 4 7 16 379.7

100 N

N +

P P

P 407.6

5 2 + 646.9

+ 441.4

+ +

+ 50

6 2

+ 0

N 4

P + 2 353.9

+ 10 2 Model

50 + 4 P 100

6 379.7

N P

6 N

+ +

7 6

5 407.8 6

N PN

P 44

P P

N 50 441.7

+ N

6 + 647.1

Relative intensity [%]

N N

2 N

P P 0 0 P 100 150 200 250 300 350 350 400 450 500 550 600 650 700 m/z m/z Experiment Figure 33. LDI TOF mass spectra of P N in C 137.8 3 5 100 positive ion mode are showing the formation of + 50 136.9 PmNn clusters. (A) Spectra in a low range of up to 138.8 135.9 139.9 m/z 350. Conditions: reflectron positive ion mode, 0 laser energy 150 a.u., maximum intensity equal to + P4N Model + 137.9 ~30 mV; peaks marked by numbers are Pm clusters 100 + (only m is given). (B) Identification of PmNn + P3N3H2 clusters at a higher mass range and comparison with 50 + 136.9 P NH+ P3N3H 4 Relative intensity [%] + 135.9 138.9 P4NH2 theoretical model. Experimental S/N values for 139.9 + + + + + 0 peaks P6N12 , P10N5 , P10N7 , P12N5 , and PN44 135 136 137 138 139 140 141 m/z were 52, 14, 5, 8, and 10, respectively. + + Identification of PmNn clusters around P4N and comparison with theoretical model. The signal/noise (S/N) ratios were ≤700 but only peaks with S/N ratio >5 were evaluated. The stoichiometry of the PmNn clusters was determined via isotopic envelope analysis and computer modelling. It is evident that P3N5 yields a high number of singly charged ions of a general

+ + + + formula PmNn , Pm , or Nn (Fig. 33(A)). Ten series of PmNn clusters were identified in positive + + + + + + + + + ion mode: PN8-11 , P4N3- 4 , P5N1-5 , P6N1-3 , P6N5-8 , P2-7N , P5-10N2 , P4-6N3 , P4-5N4 , and P5- + + + + 6N5 . Phosphorus clusters P2-7 and nitrogen clusters N9-10 and N12-13 were also identified. The phosphorus clusters were much more intense than the nitrogen clusters. The phosphorus clusters 54 formed from red phosphorus usually show odd number species, but we observed that in the P2- + 7N series the ions with an even number of phosphorus atoms are of higher intensity than those containing an odd number of 121.0 138.0 155.0 phosphorus atoms; the P N+ ion is, 100 Experiment 6 108.0 + however, an exception. In the P5N1-5 92.9

50 96.1

154.1

169.1 153.0

+ 84.1

168.0

137.0 124.0 series, P5N3 was of higher intensity 140.1 182.1 214.0 217.0 239.1 255.1 than P N + and P N +. The P N + 0 5 1-2 5 4 6 1-3 P3N2 ¯ P4N ¯ 154.8

120.9 137.9 ¯ 5

100 P NH ¯ Model ¯

3 ¯

H H ¯

clusters (low nitrogen content) show ¯ 2

107.9 2

N N

11 11

12 12

N P

P H ¯ P N

peaks of similar intensities and the 3 3 ¯

0 1

92.9 P

154.0 168.0 + 50 168.8 N ¯ 152.9 Relative intensity [%] intensity Relative 6 P6N5-8 clusters (with higher nitrogen 123.8 P N ¯ ¯ 84.0 3 6 2 P5N6

4 136.9 N13 ¯ 213.8 216.8 238.8 PN16 ¯ 140.0 content) show the highest intensity 95.9 182.0 7 255.0 0 + 100 150 200 250 peaks. Interestingly, in the P6N5-8 m/z + Figure 34. LDI TOF mass spectra of P N in negative series the most intense ion was P N . 3 5 6 6 ion mode and theoretical model. Conditions: linear + The intensity of the P6N7-8 (higher negative ion mode, laser energy 125 a.u., maximum + intensity set to 10 mV. Peaks marked by numbers are nitrogen content than in P6N6 ) + Pm clusters (only m is given). cluster was about half that of the high + + + intensity P6N6 ion. PN4,8 and P2N10 ions with one or two phosphorus atoms and a high number + of nitrogens were also observed. A few PmNn clusters were identified in the range m/z 350–800 but with low peak intensity (2.5 mV), as shown in Figure 33(B). Interestingly, the high nitrogen + content cluster PN44 was also identified (S/N ratio ~10). Comparison of the spectra with computer models shows quite good agreement. An example of a mass spectrum in negative ion mode is shown in Figure 34. The S/N values – – for the peaks identified were in the range from 430 (P4N ) to ~20 (N11 ). The stoichiometry of the PmNn clusters was determined via isotopic envelope analysis and computer modelling. – – – The spectra show a lower number of PmNn , Pm , or Nn species than are detected in positive ion – – – – mode. Only P4,5N and unary Nn (n = 6,10-13), P3-5 , and P7 clusters were identified. Intense – – – P4N and P3N2 ions were also observed. The P5N ion was of low intensity. It appears that P:N

69 stoichiometric ratios of 3:2 and 4:1 are favoured; that the species with a higher number of – phosphorus atoms (>4) show low intensity; and that the nitrogen‐rich PmNn species are of higher – – +/– intensity. The N6 cluster had a higher intensity than N13 . P4N was found to the most intense

(27.1 mV) of the PmNn clusters in the positive ion mode and it also showed the highest intensity

(13.2 mV) of all the clusters in the negative ion mode. Phosphorus clusters P3-5 and P7 were observed with a high intensity in the positive ion mode, but in the negative ion mode they were of low intensity, except for P5. A few hydrogenated PmNn species were identified in both ionization modes and repeated drying did not eliminate the presence of these hydrogenated species. It seems probable that traces of hydrogen arise from the synthesis of P3N5. An overview of the clusters obtained by LDI in both ionization modes is given in Table 11. It should be noted that spectra were also acquired in reflectron mode but no significant differences were observed.

Table 11. Review of positively and negatively charged PmNn species identified from P3N5 mass spectra.

m 0 1 2 3 4 5 6 7 10 12 + +/– +/– +/– + +/– + n = 0 P2 P3 , P4 P5 P6 P7 P10 – P3H3 + + +/– +/– + + 1 P2N P3N P4N P5N P6N P7N – P3NH +/– – + +/– + + 2 P3N2 P4N2H P5N2 P6N2 P7N2 P10N2 – + + + 3 P3N3H2 P4N3 P5N3 P6N3 + + + + 4 PN4 P2N4 P4N4 P5N4 + + + + + 5 N5 P5N5 P6N5 P10N5 P12N5 – – + 6 N6 P5N6 P6N6 + + + 7 P2N7 P6N7 P10N7 + + 8 PN8 P6N8 + + 9 N9 PN9 +/– + + + + 10 N10 PN10 P2N10 P3N10 P6N10 – + 11 N11 PN11 +/– + + 12 N12 P2N12 P6N12 +/– 13 N13 + 14 PN14 – + 16 PN16 P2N16 , + P2N16H2 + 18 P2N18 , + P2N18H2 + 20 P2N20H2 + 44 PN44

5.4.2 MALDI of P3N5

The use of MALDI with ANP, C60, CHC, DHB, DIT, HPA, TMN, elemental sulfur and selenium as matrices with the aim of generating high‐mass clusters from P3N5 was also examined. These matrices included those most often used in MALDI analysis, but some unusual matrices such as fullerene or elemental sulfur and selenium were also selected in order to map the possible effect

71 of matrix properties. It is important, however, that peaks arising from the 100 *

matrices are identified in order to 7 * +

eliminate possible misidentification. 20

8 6

The stoichiometry of the PmNn N

2 P

50 N +

4 *

+ 3

clusters was determined via isotopic 2

6 P

3 4 N

10 6

9 6 11 envelope analysis and computer P

P 11 P

5 P

8 Relative intensity [%] modelling. An example of the mass P 0 spectra obtained when HPA matrix 100 150 200 250 300 350 400 was used shows, in addition to matrix m/z Figure 35. MALDI TOF mass spectra of P3N5 using peaks, the formation of various PmNn HPA as a matrix. Conditions: linear positive ion mode, laser energy 120 a.u., maximum intensity set to 17 clusters in positive ion mode (Fig. 35). + mV. Peaks marked by numbers are Pm clusters (only + The P6N5,6 series of clusters was m is given). The values of S/N for identified species + were found to be in the range 70–400. identified here, with P6N6 having a + + higher intensity than P6N5 . The P6N8 ion was also identified and shown to have a lower + + intensity than P6N6 . Using HPA as a matrix, PmNn species with a higher nitrogen content were + + + + + + + + + + identified: P2N4 , P3N5 , P4N8 , P3N14 , P6N8 , PN20 , P6N10 , P3N16 , P6N12 , and P8N11 . Peaks + + corresponding to the P3N14 and P3N16 clusters coincide with peaks arising from the matrix and, + thus, their value is doubtful. Interestingly, nitrogen‐rich PmNn species with an odd number of phosphorus atoms and an even number of nitrogen atoms seem to be favoured. For example, the + + + + nitrogen‐rich PN20 cluster is of high intensity, whereas P6N10 , P10N2 , and P6N12 , observed in the m/z 320–360 range with an even number of phosphorus and nitrogen atoms, are of much + + + lower intensity. At the lower mass range of m/z 230–300 the P4N8 , P6N6 , and P6N8 ions, also with an even number of phosphorus and nitrogen atoms, are quite intense. In the same spectra + + Pm (m = 3-5, 7, 9) clusters and the C6H4NO3H ion from HPA were identified. In negative ion – – – mode P3N2 and P3N6 species and Pm (m = 5-11, 13, 15, 17) clusters were identified. When – – using TMN as a matrix, several PmNn clusters were generated (Fig. 36) with P3N8 being the most intense.

The stoichiometry of the PmNn clusters was determined via isotopic envelope analysis and – – – computer modelling. In the P3N2 , P3N5 , and P3N8 series we suggest that the same P3 entity is – coordinated with N2, N5, and N8 clusters. The P4N species with the more common P4 entity is of – high intensity while PN8 is of low intensity. Thus, when using the TMN matrix, the low – phosphorus content (one or two atoms) PmNn clusters were of low intensity whiles the – – P3‐containing clusters were of high intensity. Unary Pm (m = 4-7) and Nn (n = 10, 12-15) – clusters were also observed. In addition, Pm (m = 8-11, 13, 15, 17, 19, 23) clusters and traces of – + + PN16 were detected at high mass (spectra are not shown). In positive ion mode P4N4 and P3N9 + +/– species with unary Pm (m = 5, 7, 13) clusters were identified. The P3N5 species identified in the mass spectra when using HPA and TMN matrices were not observed in the LDI mass spectra of P3N5. The mass spectra obtained when using DHB as matrix show the formation of P3N6,

PN21, and PN23 species in negative ion TMN Fragments P N ¯

100 3 8 ¯

mode. When using the C60 matrix, N6 2 5

H 9 and Pm (m = 5, 7‐10, 13, 15, 17) N

¯ 7

5 5

N N

¯ P

clusters are observed in negative ion 3

2 2

¯ 3

50 ¯ 7 7

mode (spectra for DHB and C60 are P

H H

7 2

¯ 6

8 8

P N

¯ 3

not shown). When using DHB and C 10 ¯

60 ¯

13 13

15 15

12 12 14

4 N N

matrices in positive ion mode, no N Relative intensity [%]

PmNn, Pm, and Nn clusters were 0 120 140 160 180 200 220 observed. When using sulfur as a m/z Figure 36. MALDI TOF mass spectra of P N in matrix, several PmNn clusters were 3 5 negative ion mode using the TMN matrix. Conditions: identified. The mass spectra obtained linear negative ion mode, laser energy 120 a.u., when using TMN as a matrix show maximum intensity was set to 15 mV. Peaks marked by numbers are clusters Pm+ (only m is given). The traces of a few hydrogenated species. values of S/N for identified species were found to be ‐ A summary of the positively and in the range ~20 (for N14 ) up to 850 (P3N8). negatively charged PmNn, Pm, and Nn cluster ions using various matrices is given in Table 12.

When using ANP, CHC, DIT, and elemental selenium as matrices, no PmNn clusters were observed in the mass spectra.

Table 12. Summary of positively and negatively charged PmNn species identified in P3N5 mass spectra using various matrices.

m 0 1 2 3 4 5 6 7 8 9 10 HPA matrix – + n = 2 P3N2 P10N2 + 4 P2N4 + + 5 P3N5 P6N5 – + 6 P3N6 P6N6 + + 8 P4N8 P6N8 + 10 P6N10 + 11 P8N11 + 12 P6N12 + 20 PN20 TMN matrix – 1 P4N – 2 P3N2 + 4 P4N4 – 5 P3N5 – – 7 P2N7 P3N7H – – 8 PN8 P3N8 – + 9 N9H2 P3N9 – 10 N10 – 12 N12 – 13 N13 – 14 N14 – 15 N15 – 16 PN16 DHB matrix – 6 P3N6 – 21 PN21 – 23 PN23 C60 matrix – 6 N6 Sulfur as a matrix + 1 P8N + 2 P3N2 + – + 3 P2N3 P5N3 P6N3 + 4 P3N4 + 5 P2N5 – – 6 N6 P3N6 – – 7 P2N7 P5N7 + + – – 8 N8 PN8 P3N8 P5N8 *m and n indicate number of phosphorus and nitrogen atoms

5.4.3 Stoichiometry of clusters and the crystal structure of solid P3N5

An overview of all the PmNn clusters generated via the laser N N N N N ablation of phosphorus nitride is

P P P P P given in Tables 11 and 12. N N N N N N N N N N N Clusters of PmNn identified from

P P P P P P the mass spectra might be either a N N N N N N N N N N part of the α‐P N structure or are P N 3 5 P P P P formed in the plasma plume. The Figure 37. Hypothetical laser‐derived fragmentation of phosphorus nitride marking the possible way of formation structure of P3N5 as proposed in of some P N +/– clusters demonstrated on a simplified 30 m n the literature is given in Figure structure of α‐P3N5. Single layer of three‐dimensional structure of P3N5 is shown only. 37 and some of the clusters observed which might have formed by fragmentation from the α‐P3N5 crystal structure are marked. In conclusion some of the species identified in the plasma could originate from part of the P3N5 structure (Fig. 37). However, several of the species are not part of the structure and we suggest that they were generated by the action of the laser and synthesized in the plasma plume.

5.5 Publications

1. Sachinkumar Dagurao Pangavhane, Lucie Hebedová, Milan Alberti, Josef Havel. Laser ablation synthesis of new phosphorus nitride clusters from α-P3N5 via laser desorption ionization and matrix assisted laser desorption ionization time of flight mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25: 917.

2. Sachinkumar Dagurao Pangavhane, Petr Němec, Tomáš Wágner, Jan Janča, Josef Havel. Laser desorption ionization time-of-flight mass spectrometric study of binary As-Se glasses. Rapid Commun. Mass Spectrom. 2010, 24: 2000.

3. Sachinkumar Dagurao Pangavhane, Jan Houška, Tomáš Wágner, Martin Pavlišta, Jan Janča, Josef Havel. Laser ablation of ternary As-S-Se glasses and time of flight mass spectrometric study. Rapid Commun. Mass Spectrom. 2010, 24: 95.

4. Guillermo Ramírez-Galicia, Eladia María Peña-Méndez, Sachinkumar Dagurao + Pangavhane, Milan Alberti, Josef Havel. Ab initio structure modeling of AsSn (n = 1-7) cluster ions. Polyhedron 2010, 29: 1567.

100

101

102

103

104

105

106

107

108

109

CHAPTER 6. CONCLUSIONS

6.1 As-Se glasses

The spectra generated from the binary glasses with the ratio As:Se = 1:2, 2:3, 4:4, 4:3, and 7:3 + - - - show the formation of As3Seq (q = 1-5), AsSeq (q = 1-3), As2Seq (q = 2-4), and As3Seq (q = 2- 5) singly charged clusters both in positive or negative ion modes, while a slightly higher number of different clusters was detected in the negative ion mode. The clusters AsSe, As3Se2, As3Se3 and As3Se4 were found to be common to all the glasses studied and were detected in positive and negative ion modes, respectively. One series of clusters with the ratio arsenic: selenium AspSeq (p = 3, q = 1-5) in positive ion mode and three series (i) p = 1, q = 1-3, (ii) p = 2, q = 2-4, and (iii) p = 3, q = 1-5 in negative ion mode were identified. In addition to the previously proposed 167 + chains, the structure of As3Se4 cation and also in analogy with the previously proposed As-S- 163 Se heterocyclic entities in the ternary As-S-Se glasses, the heterocyclic structures for As3Se4 cluster are proposed. With respect to the number of possible isomers, the chemical nature of plasma generated particles/clusters by the LA of AspSeq glasses appears to be quite complex. The clusters AspSeq found as structural fragments are probably interconnected via As-As, As-Se, or Se-Se bridges into a complex polymer structure (Fig. 19). There is a variety of different species observed in plasma plume. In agreement with the literature, we suggest that most of them are fragments of the solid phase structures 24,25 as marked for studied As-Se glasses in Figure 19, In addition, during PLD, due to high kinetic energy of plasma particles, their possible recombination, and fast plasma quenching on substrates, a different structure than that one of the parent glasses can also be expected. This hypothesis was partly verified by the Raman study of arsenic rich As-Se PLD films; arsenic-based pyramidal units (As2Se2, As3Se or As4) were suggested to form the main structural motifs in contrast with the main structural motifs of parent 163 glasses (As4Se4, As4Se3, AsSe3). This is in agreement with TOF mass spectrometry results. In conclusion, the structures of As-Se glasses depend strongly on glass composition, but TOF MS and Raman scattering spectroscopy would suggest that the structure shows the structural motifs, as demonstrated in Figure 19, where different heterocyclic units might be interconnected into a three dimensional structure via As-Se, Se-As, Se-Se and As-As bridges.

110

6.2 As-S-Se glasses

The spectra of As33S33.5Se33.5, As33S17Se50, and As33S50Se17 chalcogenide glasses show the formation of 36 binary and 35 ternary singly charged clusters in positive or negative ion modes while a higher number of various clusters was detected in negative ion mode. The cluster species

AsS2ˉ, AsS3ˉ, AsSe2ˉ, AsSSeˉ, AsS2Seˉ, As3S2Seˉ, As3S3Seˉ, As3S2Se2ˉ, As3SSe3ˉ, AsSSe2ˉ, and

As2S2Seˉ were found common to all the glasses studied. Four series of clusters with the ratio arsenic: chalcogen = 3:4, 3:3, 3:2, and 3:1 were described. Almost the same clusters as observed from bulk material were generated from glass nano-layer prepared by PLD. In addition to chain structures proposed in the literature,167 it is suggested that the structure of some binary and ternary AspSqSer clusters is probably heterocyclic and with respect to the number of possible isomers, the chemical structure of AspSqSer glass appear to quite complex.

6.3 Ga-Ge-Sb-S glasses

Up to now Time-of-flight mass spectrometry is applied for the analysis of solid materials of binary and ternary elemental composition and has been proved successfully, this work is proving that this technique is also suitable to analyze multi-component solid materials. The mass spectra measured from Ga5Ge20Sb10S65 glass doped with erbium (0.05, 0.1, and 0.5 w.% ) show the formation of nine series of clusters in positive ion mode and fifteen series of clusters in negative ion mode. Slightly higher number of different clusters was detected in the negative ion mode. + + For all studied glasses Ga cluster was found to be the main species and Sb3S was found to be ‾ the second main species in positive ion mode. The GaS2 species was also found to be the main ‾ +/– species and SbS2 the second main species in negative ion mode. Only four species Sb3S4 , +/– +/– GamSboSp (m =1, o = 2, p = 4,5), and Ga3Sb2S7 were detected/common in positive and negative ion modes. Six species in positive ion mode and twelve species in negative ion mode + were identified with overlap of another species (Table 10). Only two species GaSb2SEr and + GaS6Er2 containing erbium were detected first time (in high mass range) in the positive ion +/– mode; while species with erbium were not detected in negative ion mode. Ternary GamSboSp species was found with higher intensity in number. Several clusters can be identified in the 2D

111

2- anionic network of [Ga2Sb2S7] structure. They could thus be considered to be structural fragments of the Ga5Ge20Sb10S65 glass structure. However, some clusters formed in the plasma plume are not part of the Ga5Ge20Sb10S65 glass structure and are evidently formed via the interaction of the species in the plasma plume.

6.4 Phosphorus nitride (P3N5)

+/– TOF MS was found useful in the study of PmNn cluster formation from α-P3N5 either in LDI or

MALDI modes. Many new Nn and binary PmNn cluster ions were identified in positive and negative ion modes. Mass spectra obtained from the LDI of P3N5 show formation of several new + PmNn clusters not described before. It was found that HPA is the most suitable matrix to + generate nitrogen rich PmNn clusters in positive ion mode. Using the HPA matrix, eight PmNn – species with highest nitrogen content (n = 20) were identified. A single cluster of N6 was 23,24,25 detected using the C60 matrix via LDI of P3N5. Even if it was shown in several papers that LDI TOF MS can give the structural information about solid materials, it is demonstrated here that, in this case, it is not so straightforward or easy. It seems that here the reactions in the plasma plume prevail. On the other hand, high nitrogen clusters observed (N13, N15) and the formation of those containing 1-2 atoms of phosphorus and rich in nitrogen (PN14, PN16, PN44,

P2N18, etc) might mean that P3N5 solid could perhaps be a suitable precursor for laser ablation synthesis of high nitrogen clusters. Further extensive work is however needed. Several of the identified clusters are common in the LDI and MALDI of P3N5 and can be identified in the crystalline structure of P3N5 and thus could be considered to be structural fragments of the α-

P3N5 structure. However, some clusters formed in the plasma plume are not part of the solid P3N5 crystal structure and are evidently formed via the interaction of the species in the plasma plume.

6.5 General conclusions

The results obtained prove that the use of MALDI TOF MS instrumentation is not limited only for protein/peptide analyses, but the technique is found i) to be powerful tool also for inorganic materials analyses, ii) reveals important information about material structure and, iii) about plasma processes during laser desorption ionisation.

112

The results concerning the LDI study of glasses enhance the information and understanding of ionized structural fragments formed in the plasma plume and, thus might also be important to elucidate the processes during PLD of target materials. The knowledge gained contributes to better understanding of the properties of solid materials and their thin films fabricated by plasma deposition techniques, especially PLD. Determined clusters stoichiometry brings information about possible structural units or motifs of solids, such information can be useful for proposal of new materials manufacturing.

If we raise the question ‘Do mass spectra reflect condensed phase chemistry of solids?’ then this dissertation answers: ‘‘Yes, mass spectra do reflect to some extend condensed phase chemistry of solids and might reveal impact information about the structure of inorganic materials.’’

With respect to the AIMS of Thesis, it can be concluded that the aims as they have been formulated (Page 16) has been fulfilled.

6.6 Obecné závěry (in Czech)

Výsledky prokazují že aplikace MALDI TOF MS instrumentace nejsou ani zdaleka limitovány proteomikou, tj. analysou peptidů a proteinů ale tato technika je také i) silným a užitečným prostředkem rovněž pro analysu anorganických materiálů ii) poskytuje důležité infromace o struktuře pevných materiálů a o iii) procesech probíhajících v plasmatu během ionizace laserovou desorpcí (LDI). Získané výsledky např. LDI studia skel poskytují cenné informace a pomáhají tak pochopit strukturu ionisovaných strukturních fragmentů v plasmovém oblaku a mohou být tak důležité pro osvětlení procesů pulsní laserové deposice vrstev studovaných materiálů. Získané znalosti přispívají k lepšímu pochopení vlastností látek v pevném stavu a jejich tenkých vrstev připravovaných plasmovými deposičními technikami, zejména PLD. Určení stechiometrie klastrů vznikajících v plasmatu přináší informace o možných strukturních jednotkách a strukturních motivech pevných látek. Tato informace může být užitečná pro návrh přípravy nových materiálů.

113

Pokud vzneseme otázku zda “Jsou hmotnostní spektra pevných látek obrazem jejich struktury?” pak disertace odpovídá, ”Ano, hmotnostní spectra do jisté miry reflektují chemickou strukturu pevné fáze a mohou poskytovat impaktní inforomace o struktuře anorganických materiálů.” Lze konstatovat a shrnout, že cíle Disertace jak byly formulovány (Strana 16) byly splněny.

114

References

1. R. Frerichs. New optical glasses with good transparency in the infrared. J. Opt. Soc. Am. 1953, 43, 1153. 2. F. Werner. A second note on the term ‘chalcogen’. J. Chem. Educ. 2001, 78, 1333. 3. F. Devillanova (editor) Handbook of chalcogen chemistry - new perspectives in sulfur, selenium and tellurium. Royal Society of Chemistry 2007, ISBN 0854043667. 4. IUPAC goldbook amides. Chalcogen replacement analogues (of amides) are called thio-, seleno-, and telluro-amides. 5. O. Takahisa. Passivation of GaAs(001) surfaces by chalcogen atoms (S, Se and Te). Surf. Sci. 1991, 255, 229. 6. B.J. Eggleton, B. Luther-Davies, K. Richardson. Chalcogenide photonics. Nat. Photonics 2011, 5, 141. 7. S. Dekker, C. Xiong, E. Magi, A.C. Judge, J.S. Sanghera, L.B. Shaw, I.D. Aggarwal, D.J. Moss, B.J. Eggleton. Broadband low power super-continuum generation in As2S3 chalcogenide glass fiber nanotapers. CLEO & QELS, conference 2010. 8. C. Florea, M. Bashkansky, J. Sanghera, I. Aggarwal, Z. Dutton. Slow-light generation through Brillouin scattering in As2S3 fibers. Opt. Mater. 2009, 32, 358. 9. S. Maurugeon, B. Bureau, C. Boussard-Pledel, A.J. Faber, P. Lucas, X.H. Zhang, J. Lucas. Selenium modified GeTe4 based glasses optical fibers for far-infrared sensing. Opt. Mater. 2011, 33, 660. 10. Q. Coulombier, L. Brilland, P. Houizot, T.N. N'Guyen, T. Chartier, G. Renversez, A. Monteville, J. Fatome, F. Smektala, T. Pain, H. Orain, J.C. Sangleboeuf, J. Troles. Fabrication of low losses chalcogenide photonic crystal fibers by molding process, in: S. Jiang, M.J.F. Digonnet, J.W. Glesener, J.C. Dries (eds.) Optical Components and Materials Vii, 2010. 11. A.B. Seddon. A prospective for new mid-infrared medical endoscopy using chalcogenide glasses. Int. J. Appl. Glass Sci. 2011, 2, 177. 12. Z.L. Samson, K.F. MacDonald, F. De Angelis, B. Gholipour, K. Knight, C.C. Huang, E. Di Fabrizio, D.W. Hewak, N.I. Zheludev. Metamaterial electro-optic switch of nanoscale thickness. Appl. Phys. Lett. 2010, 96, 143105. 13. S.J. Madden, D.Y. Choi, M.R. Lamont, V.G. Ta’eed, N.J. Baker, M.D. Pelusi, B. Luther- Davies, B.J. Eggleton. Chalcogenide glass photonic chips. Optics and Photonics News 2008, 19, 18. 14. M. El-Amraoui, G. Gadret, J.C. Jules, J. Fatome, C. Fortier, F. Désévédavy, I. Skripatchev, Y. Messaddeq, J. Troles, L. Brilland, W. Gao, T. Suzuki, Y. Ohishi, F. Smektala. Microstructured chalcogenide optical fibers from As2S3 glass: towards new IR broadband sources. Opt. Express 2010, 18, 26655. 15. I. Abdulhalim, C.N. Pannell, R.S. Deol, D.W. Hewak, G. Wylangowski, D.N. Payne. High performance acousto-optic chalcogenide glass based on Ga2S3-La2S3 systems. J. Non-Cryst. Solids 1993, 2, 1251. 16. V. Balan, C. Vigreux, A. Pradel. Chalcogenide thin films deposited by radio-frequency sputtering. J. Optoelectron Adv. M. 2004, 6, 875. 17. T. Schweizer, D.W. Hewak, B.N. Samson, D.N. Payne. Spectroscopy of potential mid- infrared laser transitions in gallium lanthanum sulphide glass. J. Lumin. 1997, 72, 419. 18. L. Ying, C. Lin, Y. Xu, Q. Nie, F. Chen, S. Dai. Glass formation and properties of novel GeS2–Sb2S3–In2S3 chalcogenide glasses. Opt. Mater. 2011, 33, 1775.

115

19. J.S. Sanghera, L.B. Shaw, L.E. Busse, V.Q. Nguyen, P.C. Pureza, B.C. Cole, B.B. Harbison, I.D. Aggarwal, R. Mossadegh, F. Kung, D. Talley, D. Roselle, R. Miklos. Development and infrared applications of chalcogenide glass optical fibers. Fiber Integr. Opt. 2000, 19, 251. 20. J. Zhang, J. Heo, X. Zhao, W. Chung. Compositional dependences on the mechanism of upconversion in Nd3+/Tm3+ co-doped chalcohalide glasses. J. Non-Cryst. Solids 2011, 357, 2421. 21. M. Virginie, N. Virginie, T. Johann, H. Patrick, A. Jean-Luc, D. Jean-Louis, M. Richard, S. Frédéric, G. Gregory, P. Stéphane, C. Guillaume. Er3+-doped GeGaSbS glasses for mid-IR fibre laser application: synthesis and rare earth spectroscopy. Opt. Mater. 2008, 31, 39. 22. J. Heo, J. M. Yoon, S.Y. Ryou. Raman spectroscopic analysis on the solubility mechanism of 3+ La in GeS2-Ga2S3 glasses. J. Non-Cryst. Solids 1998, 238, 115. 23. S.D. Pangavhane, P. Němec, T. Wágner, J. Janča, J. Havel. Laser desorption ionization time- of-flight mass spectrometric study of binary As-Se glasses. Rapid Commun. Mass Spectrom. 2010, 24, 2000. 24. S.D. Pangavhane, J. Houška, T. Wágner, M. Pavlišta, J. Janča, J. Havel. Laser ablation of ternary As-S-Se glasses and time-of-flight mass spectrometric study. Rapid Commun. Mass Spectrom. 2010, 24, 95. 25. J.Houška, E.M. Peña-Méndez, J. Kolář, M. Frumar, T. Wágner, J. Havel. Laser ablation of AgSbS2 and cluster analysis by time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2009, 23, 1715. 26. Z. Špalt, M. Alberti, E. Peña-Méndez, J. Havel. Laser ablation generation of arsenic and arsenic sulfide clusters. Polyhedron 2005, 24, 1417. 27. J. Houška, M. Alberti, J. Havel. Laser ablation synthesis of phosphorus sulphides, selenides and ternary PpSqSer clusters from various precursors. Rapid Commun. Mass Spectrom. 2008, 22, 417. 28. G. Ramírez-Galicia, E.M. Peña-Méndez, S.D. Pangavhane, M. Alberti, J. Havel. Mass + spectrometry and ab initio calculation of AsSn (n = 1-7) ion structures. Polyhedron 2010, 29, 1567. 29. E.V. Borisov, E.E. Nifant'ev. Phosphorus nitrides. Russ. Chem. Rev. 1977, 46, 842. 30. W. Schnick, J. Lücke, F. Krumeich. Phosphorus nitride P3N5: synthesis, spectroscopic, and electron microscopic investigations. Chem. Mater. 1996, 8, 281. 31. H.P. Baldus, W. Schnick, J. Lücke, U. Wannagat, G. Bogedain. Silicon phosphorus nitride, the first ternary compound in the silicon-phosphorus-nitrogen system. Chem. Mater. 1993, 5, 845. 32. W. Schnick. Solid state chemistry with non-metal nitrides. Angew. Chem., Int. Ed. Engl. 1993, 32, 806. 33. W. Schnick, J. Lücke. Lithium ion conductivity of LiPN2 and Li7PN4. Solid State Ionics 1990, 38, 271. 34. W. Schnick. Nitrido zeolites - a novel and promising class of compounds. Stud. Surf. Sci. Catal. 1994, 84, 2221. 35. Y. Hirota, T. Kobayashi. Chemical vapor deposition and characterization of phosphorus nitride (P3N5) gate insulators for InP metal-insulator-semiconductor devices. J. Appl. Phys. 1982, 53, 5037. 36. L. Chen, Y. Gu, L. Shi, Z. Yang, J. Ma, Y. Qian. Room temperature route to phosphorous nitride hollow spheres. Inorg. Chem. Commun. 2004, 7, 643. 37. S. Vepřek, Z. Iqbal, J. Brunner, M. Schärli. Preparation and properties of amorphous phosphorus nitride prepared in a low-pressure plasma. Philos. Mag. B 1981, 43, 527.

116

38. K. Landskron, H. Huppertz, J. Senker, W. Schnick. High-pressure synthesis of γ-P3N5 at 11 GPa and 1500 °C in a multianvil assembly: a binary phosphorus (V) nitride with a three- dimensional network structure from PN4 tetrahedra and tetragonal PN5 pyramids. Angew. Chem., Int. Ed. 2001, 40 No. 14, 2643. 39. P. Kroll, W. Schnick. A density functional study of phosphorus nitride P3N5: refined geometries, properties, and relative stability of α-P3N5 and γ-P3N5 and a further possible high- pressure phase δ-P3N5 with kyanite-type structure. Chem. Eur. J. 2002, 8, 3530. 40. J.A. Tossel. Second-nearest-neighbor effects on the NMR shielding of N in P3N5 and hexagonal BN. Chem. Phys. Lett. 1999, 303, 435. 41. S. Horstmann, M.E. Irran, W. Schnick. Phosphorus(V) nitride imide HP4N7: synthesis from a molecular precursor and structure determination with synchrotron powder diffraction. Angew. Chem. 1997, 36, 1992. 42. J. Ronis, B. Bondars, A. Vitola, T. Millers. Crystal structure of the phosphorous oxynitride P4ON6. J. Solid State Chem. 1995, 115, 265. 43. K. Landskron, E. Irran, W. Schnick. High-temperature high-pressure synthesis of the highly condensed nitridophosphates NaP4N7, KP4N7, RbP4N7, and CsP4N7 and their crystal-structure determinations by X-ray powder diffraction. Chem. Eur. J. 1999, 5, 2548. 44. K. Landskron, W. Schnick. Rb3P6N11 and Cs3P6N11-new highly condensed nitridophosphates by high-pressure high-temperature synthesis. J. Solid State Chem. 2001, 156, 390. 45. L. Wang, P.L. Warburton, P.G. Mezey. A theoretical study of nitrogen-rich phosphorus nitrides P(Nn)m. Chem. Phys. Lett. 2005, 109, 1125. 46. A.V. Bulgakov, O.F. Bobrenok, V.I. Kosyakov. Laser ablation synthesis of phosphorus clusters. Chem. Phys. Lett. 2000, 320, 19. 47. A.K. Hearley, B.G. Johnson, J.S. McIndoe, D.G. Tuck. Mass spectrometric identification of singly-charged anionic and cationic sulfur, selenium, tellurium and phosphorus species produced by laser ablation. Inorg. Chim. Acta 2002, 334, 105. 48. R. Huang, H. Li, Z. Lin, S. Yang. Experimental and theoretical studies of small homoatomic phosphorus clusters. J. Phys. Chem. 1995, 99, 1418. 49. M.D. Chen, J.T. Li, R.B. Huang, L.S. Zheng, C.T. Au. Structure prediction of large cationic phosphorus clusters. Chem. Phys. Lett. 1999, 305, 439. 50. M.D. Chen, R.B. Huang, L.S. Zheng, Q.E. Zhang, C.T. Au. A theoretical study for the isomers of neutral, cationic and anionic phosphorus clusters P5, P7, P9. Chem. Phys. Lett. 2000, 325, 22. 51. M.D. Chen, R.B. Huang, L.S. Zheng, C.T. Au. The prediction of isomers for phosphorus + clusters P8 and P9 . J. Mol. Struct. 2000, 499, 195. 52. B. Song, P. Cao. Structures of P20 cluster using FP-LMTO MD method. Phys. Lett. A 2001, 291, 343. 53. O. Šedo, Z. Voráč, M. Alberti, J. Havel. Laser ablation synthesis of new phosphorus and phosphorus-sulfur clusters and their TOF mass spectrometric identification. Polyhedron 2004, 23, 1199. 54. K. Sládková, J. Houška, J. Havel. Laser desorption ionization of red phosphorus clusters and their use for mass calibration in time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2009, 23, 3114. 55. M.R. Manaa. Toward new energy-rich molecular systems: from N10 to N60. Chem. Phys. Lett. 2000, 331, 262. 56. J.E. Huheey, E.A. Keiter, R.L. Keiter, in Inorganic chemistry 4th ed., Harper-Collins College

117

Publishers, New York, 1993, A31. 57. H.W. Kroto, J.R. Heath, S.C. O'brien, R.F. Curl, R.E. Smalley. C60: Buckminsterfullerene. Nature 1985, 318, 162. 58. Z.C. Ying, R.L. Hettich, R.N. Compton, R.E. Haufler. Synthesis of nitrogen-doped fullerenes by laser ablation. J. Phys. B: At. Mol. Opt. Phys. 1996, 29, 4935. 59. W. Henkes, G. Isenberg. Multiply charged cluster ions of nitrogen. Int. J. Mass Spectrom. Ion Phys. 1970, 5, 249. 60. L.J. Wang, M.Z. Zgierski. Super-high energy-rich nitrogen cluster N60. Chem. Phys. Lett. 2003, 376, 698. 61. S. Evangelisti, T. Leininger. Ionic nitrogen clusters. J. Mol. Struct. (Theochem) 2003, 621, 43. + - 62. C.K. Law, W.K. Li, X. Wang, A. Tian, N.B. Wong. A Gaussian-3 study of N7 and N7 isomers. J. Mol. Struct. (Theochem) 2002, 617, 121. 63. C. Chen, K. Sun, S. Shyu. Theoretical study of various N10 structures. J. Mol. Struct. (Theochem) 1999, 459, 113. 64. J.D. Gu, K.X. Chen, H.L. Jiang, J.Z. Chen, R.Y. Ji, Y. Ren, A.M. Tian. N18: a computational investigation. J. Mol. Struct. (Theochem) 1998, 428, 183. 65. K. Balasubramanian. Nuclear spin multiplets for the dodectahedral N20 clusters. Chem. Phys. Lett. 1993, 202, 271. 66. L. Gagliardi, S. Evangelisti, V. Barone, B.O. Roos. On the dissociation of N6 into 3 N2 molecules. Chem. Phys. Lett. 2000, 320, 518. 67. Q.S. Li, Y.D. Liu. Structures and stability of N11 cluster. Chem. Phys. Lett. 2002, 353, 204. + − 68. Y.D. Liu, P.G. Yiu, J. Guan, Q.S. Li. Structures and stability of N11 and N11 clusters. J. Mol. Struct. (Theochem) 2002, 588, 37. + – 69. P.G. Yin, Q.S. Li. Structures and stability of N13 and N13 clusters. J. Mol. Model 2004, 10, 13. 70. L.J. Wang, P.G. Mezey, M.Z. Zgierski. Stability and the structures of nitrogen clusters N10. Chem. Phys. Lett. 2004, 391, 338. 71. Y.C. Mayur, M. Gouda, V.S. Gopinath, S.M. ShantaKumar, V.S. Rajendra Prasad. Synthesis and chemical characterization of N10-substituted acridones as reversers of multidrug resistance in cancer cells. Lett. Drug Des. Discovery 2007, 4, 327. 72. Y. Baba, T. Sekiguchi, I. Shimoyama. Desorption of cluster ions from frozen gases following high-density electronic excitation. Surf. Sci. 2005, 593, 324. 73. M.L. Delitsky, R.P. Turco, M.Z. Jacobson. Nitrogen ion clusters in Triton’s atmosphere. Geophys. Res. Lett. 1990, 17, 1725. 74. J. Griffiths. A brief history of mass spectrometry. Anal. Chem. 2008, 80, 5678. 75. M.A. Grayson. Measuring mass: from positive rays to proteins. Philadelphia: Chemical Heritage 2002, ISBN 0-941901-31-9, p4. 76. Thomson on the cathode rays. Proceedings of the Cambridge Philosophical Society 1897, 9, 243. 77. J.J. Thomson. (1856-1940) Rays of positive electricity. Classic Chemistry. Retrieved 2009- 12-01. 78. J.J. Thomson. On the masses of the ions in gases at low pressures. Phil. Mag. 1899, 48, 547. 79. S. Gordon. Francis Aston and the mass spectrograph. Dalton Trans. 1998, 23, 3893. 80. F.W. Aston. Isotopes. London: E. Arnold. 1922, p152. 81. F.W. Aston. Mass-spectra and isotopes. London: Edward Arnold. 1933, p91.

118

82. O.L. Ernest. Initial performance of the 60-inch cyclotron of the William H. Crocker Radiation Laboratory. Phys. Rev. 1939, 56, 124. 83. W.E. Parkins. The uranium bomb, the calutron, and the space-charge problem. Phys. Today 2005, p51. 84. J. Mattauch, R. Herzog. Annalen der Physik 1934, 19, 345. 85. J. Mattauch, R. Herzog. Zeitschrift für Physik 1934, 89, 789. 86. A.J. Dempster. A new method of positive ray analysis. Phys. Rev. 1918, 11, 316. 87. W.E. Stephens. A pulsed mass spectrometer with time dispersion. Phys. Rev. 1946, 69, 691. 88. F.W. McLafferty. Mass spectrometric analysis. Molecular rearrangements. Anal. Chem. 1959, 31, 82. 89. M.L. Gross. Focus in honor of Fred McLafferty, 2003 Distinguished Contribution awardee, for the discovery of the McLafferty rearrangement. J. Am. Soc. Mass Spectrom. 2004, 15, 951. 90. N.M. Nibbering. The McLafferty rearrangement: a personal recollection. J. Am. Soc. Mass Spectrom. 2004, 15, 956. 91. M.S.B. Munson, F.H. Field. Chemical ionization mass spectrometry. I. General introduction. J. Am. Chem. Soc. 1966, 88, 2621. 92. H.M. Fales, G.W. Milne, J.J. Pisano, H.B. Brewer, M.S. Blum, J.G. MacConnell, J. Brand, N. Law. Biological applications of electron ionization and chemical ionization mass spectrometry. Recent Prog. Horm. Res. 1972, 28, 591. 93. R.C. Dougherty. Negative chemical ionization mass spectrometry: applications in environmental analytical chemistry. Biomed. Mass Spectrom. 1981, 8, 283. 94. M. Dole, L.L. Mach, R.L. Hines, R.C. Mobley, L.P. Ferguson, M.B. Alice. Molecular beams of macroions. J. Chem. Phys 1968, 49, 2240. 95. J.B. Fenn. Electrospray ionization mass spectrometry: how it all began. Journal of Biomolecular Techniques 2002, 13, 101. 96. H.D. Beckey. Field ionization mass spectrometry. Research/Development 1969, 20, 26. 97. M.B. Comisarow, A.G. Marshall. Fourier transform ion cyclotron resonance spectroscopy. Chem. Phys. Lett. 1974, 25, 282. 98. R.D. Macfarlane, D.F. Torgerson. Californium-252 plasma desorption mass spectroscopy. Science 1976, 191, 920. 99. E.R. Hilf. Approaches to plasma desorption mass spectrometry by some theoretical physics concepts. Int. J. Mass Spectrom. Ion Proc. 1993, 126, 25. 100. http://www.asms.org/asms99pdf/095.pdf 101. S. Borman, H. Russell, G. Siuzdak. A mass spec timeline. Today’s chemist at work. 2003, 47. 102. M. Karas, D. Bachmann, F. Hillenkamp. Inﬂuence of the wavelength in highirradiance ultraviolet laser desorption mass spectrometry of organic molecules. Anal. Chem. 1985, 57, 2935. 103. J. Krause, M. Stoeckli, U.P. Schlunegge. Studies on the selection of new matrices for ultraviolet matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 1996, 10, 1927. 104. C.R. Jiménez, L. Huang, Y. Qiu, A.L. Burlingame. Sample preparation for MALDI mass analysis of peptides and proteins. Current Protocols in Protein Science. 2001, Unit 16.3. 105. K. Markides, A. Gräslund. Advanced information on the Nobel Prize in Chemistry 2002. 106. M. Karas, U. Bahr. Laser desorption ionization mass spectrometry of large biomolecules. Trends Anal. Chem. 1990, 9, 321.

119

107. M.C. Huberty, J.E. Vath, W. Yu, S.A. Martin. Site-specific carbohydrate identification in recombinant proteins using MALD-TOF MS. Anal. Chem. 1993, 65, 2791. 108. S. Sekiya, Y. Wada, K. Tanaka. Derivatization for stabilizing sialic acids in MALDI-MS. Anal. Chem. 2005, 77, 4962. 109. Y. Fukuyama, S. Nakaya, Y. Yamazaki, K. Tanaka. Ionic liquid matrixes optimized for MALDI-MS of sulfated/sialylated/neutral oligosaccharides and glycopeptides. Anal. Chem. 2008, 80, 2171. 110. D.I. Papac, A. Wong, A.J. Jones. Analysis of acidic oligosaccharides and glycopeptides by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal. Chem. 1996, 68, 3215. 111. D.J. Harvey. Matrix-assisted laser desorption/ionization mass spectrometry of carbohydrates. Mass Spectrom. Rev. 1999, 18, 349. 112. E. Lattová, V.C. Chen, S. Varma, T. Bezabeh, H. Perreault. Matrix-assisted laser desorption/ionization on-target method for the investigation of oligosaccharides and glycosylation sites in glycopeptides and glycoproteins. Rapid Commun. Mass Spectrom. 2007, 21, 1644. 113. L. Becker, R.J. Cotter, W. Brinckerkoff, F. Robert, P. Ehrenfreund. Detection of organic compounds in polar ices using AP MALDI. Geophysical Research Abstracts 2005, 7, 09866. 114. R. Lidgard, M.W. Duncan. Utility of matrix-assisted laser desorption/ionization time-of- flight mass spectrometry for the analysis of low molecular weight compounds. Rapid Commun. Mass Spectrom. 1995, 9, 128. 115. H.Y. Wang, X. Chu, Z.X. Zhao, X.S. He, Y.L. Guo. Analysis of low molecular weight compounds by MALDI-FTICR-MS. J. Chromatogr. B 2011, 879, 1166. 116. S.D. Pangavhane, L. Hebedová, M. Alberti, J. Havel. Laser ablation synthesis of new phosphorus nitride clusters from α-P3N5 via laser desorption ionization and matrix assisted laser desorption ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25, 917. 117. W. Schrepp, H. Pasch. Maldi-Tof mass spectrometry of synthetic polymers (Springer Laboratory). Berlin: Springer-Verlag. 2003, ISBN 3-540-44259-6. 118. W.F. Nielen, M.M. Sabine. Characterization of polydisperse synthetic polymers by size- exclusion chromatography/matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 1997, 11, 1194. 119. K.J. Wu, R.W. Odom. Characterizing synthetic polymers by MALDI MS. Anal. Chem. 1998, 70, 456A. 120. D.C. Schriemer, L. Li. Mass discrimination in the analysis of polydisperse polymers by MALDI Time-of-Flight mass spectrometry. 2. Instrumental issues. Anal. Chem. 1997, 69, 4176. 121. P Seng, M Drancourt, F Gouriet, S.B. La, P.E. Fournier, J.M. Rolain, D. Raoult. Ongoing revolution in bacteriology: routine identification of bacteria by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Infect Dis. 2009, 49, 552. 122. S.L. Cohen, B.T. Chait. Influence of matrix solution conditions on the MADLI-MS analysis of peptides and proteins. Anal. Chem. 1996, 68, 31. 123. R.C. Beavis, B.T. Chait. Matrix-assisted laser desorption ionization mass-spectrometry of proteins. Methods in Enzymol. 1996, 270, 519. 124. O. Vorm, P. Roepstorff, M. Mann. Improved resolution very high sensitivity in MALDI TOF of matrix surfaces made by fast evaporation. Anal. Chem. 1994, 66, 3281. 125. J. E. Coligan, B. M. Dunn, H. L. Ploegh. Matrix-assisted laser desorption/ionization time-

120 of-flight mass analysis of peptides. D. W. Speicher and P.T. Wingfield, eds. (Contributed by W.J. Henzel and J.T. Stults), Current Protocols in Protein Science, Volume 1, John Wiley & Sons, New York, N. Y., 1995, Unit 16.2. 126. K. Hjerno, P. Hojrup. Calibration of matrix-assisted laser desorption/ionization time-of- flight peptide mass fingerprinting spectra. In R. Matthiesen, editor. (ed): Methods in Molecular Biology : Mass Spectrometry Data Analysis in Proteomics. Totowa, NJ, USA: Humana, 2007, 367, 49. 127. J. Gobom, M. Mueller, V. Egelhofer, D. Theiss, H. Lehrach, E. Nordhoff. A calibration method that simplifies and improves accurate determination of peptide molecular masses by MALDI-TOF MS. Anal. Chem. 2002, 74, 3915. 128. M. Bantscheff, B. Duempelfeld, B. Kuster. An improved two-step calibration method for matrix-assisted laser desorption/ionization time-of-flight mass spectra for proteomics. Rapid Commun. Mass Spectrom. 2002, 16, 1892. 129. E. Moskovets, H.S. Chen, A. Pashkova, T. Rejtar, V. Andreev, B.L. Karger. Closely spaced external standard: a universal method of achieving 5 ppm mass accuracy over the entire MALDI plate in axial matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2003, 17, 2177. 130. W.A. Harris, D.J. Janecki, J.P. Reilly. Use of matrix clusters and trypsin autolysis fragments as mass calibrants in matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2002, 16, 1714. 131. D.C. Chamrad, G. Koerting, J. Gobom. Interpretation of mass spectrometry data for high- throughput proteomics. Anal. Bioanal. Chem. 2003, 376, 1014. 132. F. Levander, T. Rognvaldsson, J. Samuelsson, P. James. Automated methods for improved protein identification by peptide mass fingerprinting. Proteomics 2004, 4, 2594. 133. R. Zenobi, R. Knochenmuss. Ion formation in MALDI mass spectrometry. Mass Spectrom. Rev. 1998, 17, 337. 134. G.L. Glish, R.W. Vachet. The basics of mass spectrometry in the twenty-first century. Nat. Rev. Drug Discov. 2003, 2, 140. 135. R.H. Perry, R.G. Cooks, R.J. Noll. Orbitrap mass spectrometry: instrumentation, ion motion, and applications. Mass Spectrum. Rev. 2008, 27, 661. 136. W.C. Wiley, I.H. MacLaren. Time‐of‐Flight mass spectrometer with improved resolution. Rev. Sci. Instrum. 1955, 26, 1150. 137. J.T. Watson, O.D. Sparkman. Introduction to mass spectrometry: instrumentation, applications, and strategies for data interpretation. John Wiley & Sons Ltd, England, 4th Edition, 2008. 138. A. Westman, G. Brinkmalm, D.F. Barofsky. MALDI induced saturation effects in chevron microchannel plate detectors. Int. J. Mass Spectrom. Ion 1997, 169, 79. 139. J.J. Lennon, R.S. Brown. Proceedings of the 42nd annual conference on mass spectrometry and allied topics, Chicago, IL, ASMS, Santa Fe, 1994, p501. 140. S.M. Colby, T.B. King, J.P. Reilly. Improving the resolution of matrix assisted laser desorption ionization time-of-flight mass spectrometry by exploiting the correlation between ion position and velocity. Rapid Commun. Mass Spectrom. 1994, 8, 865. 141. R.S. Brown, J.J. Lennon. Mass resolution improvement by incorporation of pulsed ion extraction in a matrix-assisted laser desorption/ionization linear time-of-flight mass spectrometer. Anal. Chem. 1995, 67, 1998.

121

142. R.M. Whittal, L. Li. High-Resolution Matrix-Assisted Laser Desorption/Ionization in a Linear Time-of-Flight Mass Spectrometer. Anal. Chem. 1995, 67, 1950. 143. M.L. Vstal, P. Juhasz, S.A. Martin. Delayed extraction matrix-assisted laser desorption time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 1995, 9, 1044. 144. P. Juhasz, M.T. Roskey, I.P. Smimov, L.A. Haff, M.L. Vestal, S.A. Martin. Applications of delayed extraction matrix-assisted laser desorption ionization time-of-flight mass spectrometry to oligonucleotide analysis. Anal. Chem. 1996, 68, 941. 145. R. Kaufmann, P. Chaurand, D. Kirsch, B. Spengler. Post-source decay and delayed extraction in matrix-assisted laser desorptiod ionization-reflectron time-of-flight mass spectrometry. Are there trade-offs? Rapid Commun. Mass Spectrom. 1996, 10, 208. 146. X. Tang, W. Ens, F. Mayer, K.G. Standing, J.B. Westmore. Measurement of unimolecular decay in peptides of masses greater than 1200 units by a reflecting time-of-flight mass spectrometer. Rapid Commun. Mass Spectrom. 1989, 3, 443. 147. X. Tang, W. Ens, F. Mayer, K.G. Standing, J.B. Westmore. Daughter ion mass spectra from cationized molecules of small oligopeptides in a reflecting time-of-flight mass spectrometer. Anal. Chem. 1988, 60, 1791. 148. R. Kaufmann, B. Spengler, F. Lutzenkirchen. Mass spectrometric sequencing of linear peptides by product-ion analysis in a reflectron time-of-flight mass spectrometer using matrix- assisted laser desorption ionization. Rapid Commun. Mass Spectrom. 1993, 7, 902. 149. R. Kaufmann, D. Kirsch, B. Spengler. Sequenching of peptides in a time-of-flight mass spectrometer: evaluation of postsource decay following matrix-assisted laser desorption ionisation (MALDI). Int. J. Mass Spectrom. Ion Processes 1994, 131, 355. 150. J. Machold, Y. Utkin, D. Kirsch, R. Kaufmann, V. Tsetlin, F. Hucho. Photolabeling reveals the proximity of the alpha-neurotoxin binding site to the M2 helix of the ion channel in the nicotinic acetylcholine receptor. Proc. Natl Acad. Sci. USA 1995, 92, 7282. 151. J.C. Rouse, W. Yu, S.A. Martin. A comparison of the peptide fragmentation obtained from a reflector matrix-assisted laser desorption-ionization time-of-flight and a tandem four sector mass spectrometer. J. Am. Soc. Mass Spectrom. 1995, 6, 822. 152. B. Spengler, D. Kirsch, R. Kaufmann. Fundamental aspects of postsource decay in matrix- assisted laser desorption mass spectrometry. 1. Residual gas effects. J. Phys. Chem. 1992, 96, 9678. 153. B. Spengler, D. Kirsch, R. Kaufmann, E. Jaeger. Peptide sequencing by matrix-assisted laser-desorption mass spectrometry. Rapid Commun. Mass Spectrom. 1992, 6, 105. 154. B. Spengler. Post-source decay analysis in matrix-assisted laser desorption/ionization mass spectrometry of biomolecules. J. Mass Spectrom. 1997, 32, 1019. 155. A. Zakery, S.R. Elliott. Optical properties and applications of chalcogenide glasses: a review. J. Non-Cryst. Solids 2003, 330, 1. 156. A. Gruszecka, M. Szymanska-Chargot, A. Smolira, J. Cytawa, L. Michalak. Role of the supportmaterial on laser desorption/ionization mass spectra. Rapid Commun. Mass Spectrom. 2008, 22, 925. 157. P. Němec, J. Jedelský, M. Frumar, M. Štábl, M. Vlček. Structure, thermally and optically induced effects in amorphous As2Se3 films prepared by pulsed laser deposition. J. Phys. Chem. Solids 2004, 65, 1253. 158. M.S. Iovu, E.I. Kamitsos, C.E. Varsamis, P. Boolchand, M. Popescu. Raman spectra of AsxSe100-x glasses doped with metals. J. Optoelectron. Adv. Mater. 2005, 7, 1217. 159. M. Ystenes, W. Brockner, F. Menzel. Scaled quantum mechanical (SQM) calculations and

122 vibrational analyses of the cage-like molecules P4S3, As4Se3, P4Se3, As4S3, and PAs3S3. Vib. Spectrosc. 1993, 5, 195. 160. M. Ystenes, F. Menzel, W. Brockner. Ab initio quantum mechanical calculations of energy, geometry, vibrational frequencies and IR intensities of tetraphosphorus tetrasulphide, α- P4S4(D2d), and vibrational analysis of As4S4 and As4Se4. Spectrochimica Acta 1994, 50A, 225. 161. P. Nagels, R. Callaerts, M. Roy, M. Vlček. Plasma deposition of amorphous As-S films. J. Non-Cryst. Solids 1991, 137, 1001. 162. M. Frumar, B. Frumarová, T. Wágner, P. Němec. In photo-induced metastability in amorphous semiconductors. A.V. Kolobov (ed). Wiley-VCH: Weinheim, Germany, 2003, 23. 163. P. Němec, J. Jedelsky, M. Frumar, M. Stabl, Z. Cernosek. Structure, optical properties and their photo-induced changes in AsxSe100-x (x = 50, 57.1, 60) amorphous thin films prepared by pulsed laser deposition. Thin Solid Films 2005, 484, 140. 164. M. Popescu, F. Sava, A. Lorinczi. Self-organization and anisotropy in amorphous chalcogenides. J. Non-Cryst. Solids 2006, 352, 1506. 165. R. Golovachak, O. Shpotyuk, A. Kozdras, B. Bureau, M. Vlček, A. Ganjoo, H. Jain. Atomistic model of physical ageing in Se-rich As-Se glasses. Philos. Mag. 2007, 87, 4323. 166. R. Golovachak, A. Kovalskíy, A.C. Miller, H. Jaín, O. Shpotyuk. Structure of Se-rich As-Se glasses by high-resolution x-ray photoelectron spectroscopy. Phys. Rev. B 2007, 76, 125208. 167. N. Mateleshko, V. Mitsa, M. Veres, M. Koos. Investigation of nanophase separation in glassy As40Se60 using Raman scattering and ab initio calculations. J. Optoelectron. Adv. Mater. 2005, 7, 991. 168. A.N. Rosli, H.A. Kassim, K.N. Shrivastava. Vibrational frequencies in AsxSe1-x glass. Current Issues of Physics in Malaysia 2008, 1017, 429. 169. B.H. Christian, R.J. Gillespie, J.F. Sawyer. Preparation, x-ray crystal structures and + + vibrational spectra of some salts of the arsenic-sulfur (As3S4 ) and arsenic-selenium (As3Se4 ) cations. Inorg. Chem. 1981, 20, 3410. 170. J.C. Mauro, A.K. Varshneya. Multiscale modeling of arsenic selenide glass. J. Non-Cryst. Solids 2007, 353, 1226. 171. O. Šedo, M. Alberti, J. Havel. Laser ablation synthesis of new binary chalcogen molecules from the selenium–sulfur system. Polyhedron 2005, 24, 639. 172. K. Tanaka. Structural phase transitions in chalcogenide glasses. Phys. Rev. 1989, 39, 1270. 173. S.R. Elliott. Physics of amorphous materials, Essex, in Longman Scientific & Technical, New York, Wiley, ed. 2nd, 1990. 174. C. Julien, S. Barnier, M. Massot, N. Chbani, X. Cai, A.M. Loireaulozach, M. Guittard. Raman and Infrared spectroscopic studies of Ge-Ga-Ag sulfide glasses. Mat. Sci. Eng. B-Solid 1994, 22, 191. 175. N. Chbani, A.M. Loireaulozach, F. Keller, S. Benazeth. Glass-forming tendency in Ga-Ge-S glasses - a structural approach. J. Phys. IV 1994, 4, 113. 176. B.G. Aitken, C.W. Ponader. Physical properties and Raman spectroscopy of GeAs sulphide glasses. J. Non-Cryst. Solids 1999, 257, 143. 177. G. Lucovsky, F.L. Galeener, R.C. Keezer, R.H. Geils, H.A. Six. Structural interpretation of the infrared and Raman spectroscopy of glasses in the alloy system Ge1-xSx. Phys. Rev. B 1974, 10, 5134. 178. S. Sugai. Stochastic random network model in Ge and Si chalcogenide glasses. Phys. Rev. B 1987, 35, 1345. 179. V. Nazabal, P. Nemec, A.M. Jurdyc, S. Zhang, F. Charpentier, H. Lhermite, J. Charrier, J.P.

123

Guin, A. Moreac, M. Frumar, J.L. Adam. Optical waveguide based on amorphous Er3+-doped Ga-Ge-Sb-S(Se) pulsed laser deposited thin films. Thin Solid Films 2010, 518, 4941. 180. H. Higuchi, R. Kanno, Y. Kawamoto, M. Takahashi, K. Kadano. Local structures of Er3+ containing Ga2S3-GeS2-La2S3 glass. Phys. Chem. Glasses 1999, 40, 122. 181. J.H. Song, Y.G. Choi, K. Kadono, K. Fukumi, H. Kageyama, J. Heo. EXAFS investigation on the structural environment of Tm3+ in Ge-Ga-S-CsBr glasses. J. Non-Cryst. Solids 2007, 353, 1251. 182. Y.G. Choi, J.H. Song, Y.B. Shin, J. Heo. Chemical characteristics of Dy-S bonds in Ge-As- S glass. J. Non-Cryst. Solids 2007, 353, 1665. 183. M. Munzar, K. Koughia, D. Tonchev, S.O. Kasap, T. Sakai, K. Maeda, T. Ikari, C. Haugen, R. Decorby, J.N. McMullin. Influence of Ga on the optical and thermal properties of Er2S3 doped stoichiometric and nonstoichiometric Ge-Ga-Se glasses. Phys. Chem. Glasses 2005, 46, 215. 184. B.G. Aitken, C.W. Ponader, R.S. Quimby. Clustering of rare earths in GeAs sulfide glass. C. R. Chim. 2002, 5, 865. 185. L. Bigot, A.M. Jurdyc, B. Jacquier, J.L. Adam. Inhomogeneous and homogeneous linewidths in Er3+-doped chalcogenide glasses. Opt. Mater. 2003, 24, 97. 186. L. Petit, N. Carlie, F. Adamietz, M. Couzi, V. Rodriguez, K.C. Richardson. Correlation between physical, optical and structural properties of sulfide glasses in the system Ge–Sb–S. Mater. Chem. Phys. 2006, 97, 64. 187. M.F. Churbanov. High-purity chalcogenide glasses as materials for fiber optics. J. Non- Cryst. Solids 1995, 184, 25. 188. T. Kavetskyy, R. Golovchak, O. Shpotyuk, J. Filipecki, J. Swiatek. On the compositional trends in IR impurity absorption of Ge–As(Sb)–S glasses. Chalcogenide Letters 2004, 1, 125. 189. V. Moizan, V. Nazabal, J. Troles, P. Houizot, J.L. Adam, F. Smektala, G. Gadret, S. Pitois, J.L. Doualan, R. Moncorgé, G. Canat. Er3+-doped GeGaSbS glasses for mid-IR fibre laser application: Synthesis and rare earth spectroscopy. Opt. Mater. 2008, 31, 39. 190. M.L. Feng, Z.L. Xie, X.Y. Huang. Two gallium antimony sulfides built on a novel heterometallic cluster. Inorg. Chem. 2009, 48, 3904. 191. A. Povolotskiy, T. Ivanova, A. Manshina, Y. Tver'yanovich, S.K. Liaw, C.L. Chang. Er3+ as glass structure modifier of Ga-Ge-S chalcogenide system. Appl. Phys. A 2009, 96, 887.

124

List of abbreviations and acronyms

ANP 2‐amino‐5‐nitropyridine a.u. arbitrary unit CHCA α-cyano-4-hydroxycinnamic acid ChG Chalcogenide Glass ChG’s Chalcogenide Glasses CVD Chemical Vapour Deposition DE Delayed Extraction DFT Density Functional Theory DHB 2,5-dihydroxybenzoic acid DIT 1,8‐dihydroxy‐9[10H]‐anthracenone EDX Energy Dispersive X-ray analyzer ESI Electrospray Ionization EDX Energy-Dispersive X-ray analyzer EPR Electron paramagnetic resonance EXAFS Extended X-ray Absorption Fine Structure FT-ICR MS Fourier Transform Ion Cyclotron Resonance Mass Spectrometry HPA 3‐hydroxypicolinic acid ICP MS Inductively Coupled Plasma Mass Spectrometry ICR Ion Cyclotron Resonance KeV Kiloelectron Volt KE Kinetic Energy LASER Light Amplification by Stimulated Emission from Radiation LA Laser Ablation LD Laser Desorption LDI Laser Desorption Ionisation LI Laser Ionization MS Mass Spectrometry MS Mass Spectrometer m/z Mass-to-charge ratio MALDI Matrix Assisted Laser Desorption Ionisation MCP Micro-Channel Plate NMs Nanomaterials PDMS Plasma Desorption Ionization Mass Spectrometry PSD Post Source Decay PLD Pulsed Laser Deposition ppm parts per million QIT Quadrupole Ion Trap QTOF Quadrupole Time of Flight QIT-TOF Quadrupole Ion Trap Time-Of-Flight RT Room Temperature SA Sinapinic acid SEM Scanning Electron Microscopy TEM Transmission Electron Microscopy

125

TFA Trifluoroacetic acid TMN trans‐2‐[3‐(4‐tert‐butylphenyl)‐2‐methyl‐2‐propenylidene]malononitrile TOF MS Time-Of-flight Mass Spectrometry US EPA US US Environmental Protection Agency XRD X-Ray Diffraction

126

APPENDIX

List of Publications

Publications as part of the PhD Thesis

1. S.D. Pangavhane, L. Hebedová, M. Alberti, J. Havel. Laser ablation synthesis of new phosphorus nitride clusters from α-P3N5 via laser desorption ionization and matrix assisted laser desorption ionization time of flight mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25: 917.

2. S.D. Pangavhane, J. Houška, T. Wágner, M. Pavlišta, J. Janča, J. Havel. Laser ablation of ternary As-S-Se glasses and time of flight mass spectrometric study. Rapid Commun. Mass Spectrom. 2010, 24: 95.

3. S.D. Pangavhane, P. Němec, T. Wágner, J. Janča, J. Havel. Laser desorption ionization time- of-flight mass spectrometric study of binary As-Se glasses. Rapid Commun. Mass Spectrom. 2010, 24: 2000.

4. S.D. Pangavhane, P. Němec, V. Nazabal, A. Moreac, J. Havel. Clusters synthesis via laser desorption ionization time-of-flight mass spectrometry and evidence for structural motifs of erbium doped Ga-Ge-Sb-S glasses. to be published.

Other papers - not part of the PhD Thesis

5. R.G. Guillermo, E.M. Peña-Méndez, S.D. Pangavhane, M. Alberti, J. Havel. Ab initio + structure modeling of AsSn (n = 1-7) cluster ions. Polyhedron 2010, 29: 1567.

6. S.D. Pangavhane, J. Havel, M. Valiente. Arsenite and arsenate adsorption on impregnated cellulose sponge with maghemite and magnetite nanoparticles. In preparation.

7. S.D. Pangavhane, N.R. Panyala, E.M. Peña-Méndez, J. Havel. Nanodiamonds: Applications in medicine. Review. In preparation.

Poster Presentation

S.D. Pangavhane, P. Němec, V. Nazabal, A. Moreac, J. Havel. Analyses of erbium doped Ga- Ge-Sb-S glasses by laser desorption ionization time-of-flight mass spectrometry. 10th International Conference Solid State Chemistry (SSC), June 10 - 14, 2012, Faculty of Chemical Technology, University of Pardubice, Pardubice, Czech Republic.

List of Oral Presentations

1. S.D. Pangavhane, P. Němec, J. Havel. Laser ablation or laser desorption ionization of binary As-Se glasses: a time-of-flight mass spectrometric study. 15th International Conference on Thin

127

Films, Kyoto, Japan, Meeting Guide, 8. - 11. 11. 2011, p. 18. (http://www.ictf15.jp/)

2. S.D. Pangavhane, J. Havel. Chalcogenide glasses: preparation, applications, and laser desorption ionisation time-of-flight mass spectrometric study, XD107 seminar, Department of Chemistry, Masaryk University, Czech Republic, 27.10.2011 (https://is.muni.cz/auth/el/1431/podzim2011/XD 107/um/)

3. S.D. Pangavhane, J. Havel. Elucidation of chalcogenide glasses chemical structure via Time of Flight Mass Spectrometry (TOF MS), Conference "Potential and Applications of Surface Nanotreatment of Polymers and Glass 2011", NANO contact, Department of Physical Electronics, Masaryk University, Blansko, Czech Republic, 17. - 19. 10. 2011 (http://nanocontact.cz)

4. S.D. Pangavhane. Arsenic: an aqueous adsorption, Group of Separation Technology, Department of Analytical Chemistry, University Autonoma of Barcelona, Barcelona, Spain, 08.04.2011.

5. S.D. Pangavhane. Characterization of drug molecules via HPLC & Mass Spectrometric Analysis of Chalcogenide Glasses, Group of Separation Technology, Department of Analytical Chemistry, University Autonoma of Barcelona, Barcelona, Spain, 19.01.2011.

6. S.D. Pangavhane, T. Wagner, P. Němec, J. Janča, J. Havel. Do Mass Spectra Reflect Condensed-Phase chemistry of glasses? XV. annual training seminar of PhD students, INORGANIC NON-METALLIC MATERIALS, Department of Glass and Ceramics, Faculty of Chemical Technology, Institute of Chemical Technology, Prague, Czech Republic, 2010 (http://tresen.vscht.cz/sil/en/node/97).

7. S.D. Pangavhane. Mass spectrometric analysis of chalcogenide glasses- formation of clusters, Internal Seminar, Department of Chemistry, Masaryk University, Brno, Czech Republic, 2010.

8. S.D. Pangavhane. Ternary As-S-Se glasses and AspSqSer clusters, Internal Seminar, Department of Chemistry, Masaryk University, Brno, Czech Republic, 2009.

128

Curriculum vitae

Mr. Sachinkumar D. Pangavhane, Ph.D. fellow Department of chemistry (A14/107), Masaryk University, Brno, Czech Republic E-mail [email protected]

Career objective To pursue knowledge in the current research field of analytical chemistry by effectively utilizing my technical and strong interpersonal skills, thereby fulfilling my deepest desire to learn and achieve the best.

Educational qualification 1. M.Sc. (Analytical chemistry), Pune University, India, 2007, First class (60.85%). 2. B.Sc. (Chemistry), Pune University, India, 2005, Distinction (78.50%)

Research Synopsis of Ph.D. thesis Binary, ternary, or multi-component chalcogenide glasses are important for optics, computers, material science and technological applications but their structure is still un-apprehended. The formation of clusters in plasma plume from different glasses by laser desorption ionization (LDI) or laser ablation (LA) was studied by time-of-flight mass spectrometry (TOF MS) in positive and negative ion modes. The stoichiometry of clusters reflects to a certain extend chemical structure of glasses and thus TOF MS represents suitable experimental technique yielding important information about glasses chemical structure.

Publications 5 Research papers, 8 Oral/1 Poster presentations Experience 5 Years Masaryk University, Czech Republic: 3 Years - MALDI- TOF MS University Autonoma of Barcelona, Spain: 1 Year - Synthesis and characterization of nanomaterials National Chemical Laboratory, India: 1 Year - HPLC + 3 months - NMR

Personal profile 1. Name Mr. Pangavhane Sachinkumar Dagurao 2. Permanent address A/P Mahegaon deshmukh, Kopargaon (Taluka), Ahmednagar (District), Maharashtra (State), India, Postal code 423 602 3. Date of birth 16th Feb 1984 4. Gender Male 5. Marital status Single 6. Nationality Indian 7. Languages Marathi, English, and Hindi

129

Awards

1. Chemistry prize for best Ph.D. fellow, 2012, Deapertment of Chemistry, Masaryk Univeristy 2. Masaryk University, Rector’s fellowship for best Ph.D. student has been awarded three times 2009/10/11. 3. Masaryk University, Erasmus fellowship for abroad stay 2010/11.

References

1. Prof. Josef Havel 2. Prof. Manuel Valiente Department of Chemistry, Department of Chemistry, Faculty of Science, Faculty of Science and Bioscience Masaryk University, University Autonoma of Barcelona, Brno, Czech Republic Barcelona, Spain. E-mail [email protected] E-mail [email protected] Ph. +420 549 49 4114, 8677 Ph. +34 935812903

I hereby declare that all above information is true for the best of my knowledge and belief.

Place - Brno, Czech Republic Date - 24th July 2012 Sachinkumar D. Pangavhane

130