Metal cluster sputtering under reactive ion bombardment investigated by TOF-SNMS-laser-system [Elektronische Ressource] / by Sobhy Ahmed Nassar Ahmed Ghalab

Sohayb

Sujets

• Physik

Informations

• Publié par: universitat_duisburg-essen
• Publié le: 01 janvier 2005
• Nombre de lectures: 39
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

Metal Cluster Sputtering under reactive
Ion Bombardment investigated by
TOF-SNMS-Laser-System
Dissertation
Submitted to the Department of Physics, University of Duisburg-Essen
In order to fulfill the requirements for the academic degree
Doctor rerum naturalium (Dr. rer. nat.)
by
Sobhy Ahmed Nassar Ahmed Ghalab
born in Ghariba, Egypt
1. Referee: Prof. Dr. A. Wucher
2. Referee: Prof. Dr. R. Möller
Public defence held on 13.07.2005
July 21, 2005Contents
1 Introduction 3
2 Physical basis of the sputtering process 7
2.1 Descriptionofsputteringprocess .................. 7
2.2 Classificationofsputteringevents.................. 13
2.2.1 Singleknock-onregime ................... 13
2.2.2 Linear-cascaderegime .................... 14
2.2.3 Collision spike regime (nonlinear cascade) . . . . . . . . . 15
3 Formation of clusters during the sputtering process 16
3.1 Experimental observation of sputtered clusters . . . . . . . . . . 16
3.2 Theoreticalmodelsofsputteredcluster............... 20
3.3 Computersimulation......................... 22
3.4 Formation of sputtered clusters under reactive ion bombardment 27
4 Experimental 31
4.1 Experimentalsetup ......................... 31
4.1.1 Generaldescription ..................... 31
4.1.2 Samplechamber....................... 32
4.1.3 Vacuumsystem ....................... 32
4.1.4 Ionsource .......................... 33
4.1.5 Targetsurface ........................ 37
4.1.6 Theactualtargetcurrent.................. 37
4.1.7 Lasersystem......................... 38
4.1.8 UV/VUV Detector . . . . . . . . . . . . . . . . . . . . . . 40
4.1.9 Opticalcomponents..................... 40
4.2 Time-of-flight mass spectrometer (TOF-MS) . . . . . . . . . . . 41
4.3 Detection of the sputtered species. . . . . . . . . . . . . . . . . . 45
4.3.1 Analogmode ......................... 47
4.3.2 Pulsecountingmode..................... 48
4.3.3 Saturation of microchannel plates and blanking . . . . . . 49
5 Methodology of the measurements 51
5.1 Timesynchronization ........................ 51
5.2 Method ................................ 55
6 Photoionization 57
7Results 60
+7.1 Sputtering from indium under bombardment with SF ( m =m
+1,...,5) and Ar projectiles . . . . . . . . . . . . . . . . . . . . . 61
7.1.1 Experimentalconditions................... 61
7.1.2 Dependence of measured signal on laser intensity . . . . . 64
17.1.3 Neutralclusters........................ 66
7.1.4 Secondaryclusterions.................... 69
7.1.5 Ionizationprobabilities.................... 73
7.1.6 Partialsputteringyields .................. 80
7.1.7 Conclusion .......................... 82
7.2 Formation of sputtered silver clusters under bombardment with
++Xe and SF ions .......................... 845
7.2.1 Experimentalconditions................... 85
7.2.2 Neutralclusters........................ 85
7.2.3 Secondaryclusterions .................... 88
7.2.4 Conclusion .......................... 93
+7.3 Cluster sputtering from silver under bombardment with SF m
( m=1,..,5)projectiles ....................... 94
7.3.1 Experimentalconditions................... 94
7.3.2 Neutralclusters........................ 97
7.3.3 Secondaryclusterions .................... 98
7.3.4 Ionizationprobabilities....................101
7.3.5 Totalsputteringyields....................102
7.3.6 Partialsputteringyields ...................104
7.3.7 Conclusion ..........................106
7.4 Investigation of the bombarded surface by X-ray Photoelectron
Spectroscopy(XPS) .........................107
7.4.1 PhysicalbasisofXPS ....................107
7.4.2 Experimentalconditions...................113
7.4.3 Measurementsprocedure...................113
7.4.4 Results ............................115
8 Summary 120
21Introduction
Sputtering is the removal of materials from the surface of a solid through the
impacting of energetic particles. The materials released from the bombarded
surfaceconsistpredominantlyofneutralatomsandsecondaryionspositivelyor
negatively charged [Be81], [Be91]. The sputtering process has been the subject
of scientific investigations for long time. First discussions over the sputtering
by atomic beams were already underway at the beginning of the last century
[St08], [St09], [Hi27]. Besides atomic species, the flux of sputtered particles
contains an abundant fraction of agglomerates of two or more atoms; these
species were called sputtered clusters. R. Honig [Ho58] was the first to report
on sputtered clusters in 1958 s, he succeeded in detecting positively charged
silver dimers within the flux of particles emitted by sputter erosion of cathode
in a gas discharge. Hortig and Müller [Ho69] observed negative clusters with
−asizeupto 60 Ag atoms when they bombarded polycrystalline silver (whichn
was partly covered in Cs in order to enhance negative formation) with 15 keV
+Kr . This record in cluster size was only broken by Katakuse et al .[Ka85],
+[Ka86],whofoundAg clusterscontaininguptomorethan200atomssputteredn
+from polycrystalline silver under bombardment with 10 keV Xe ions. In fact,
the secondary ion mass spectrometry (SIMS) su ffers primarily from the strong
matrix e ffects i.e. the flux of a specific kind of particles changes not only
as a function of the surface concentration of this specificparticletype,but
mainly due to the presence of other elements at the surface. Due to this e ffect,
the quantification based on the SIMS signal is di fficult. Therefore, it is very
essential to investigate the sputtered neutral species.
Therefore,severalattemptshavebeenmadetostudysputteredneutralclus-
ters that have to be post-ionized prior the detection by analytical techniques.
A highly successful investigation of sputtered neutral clusters was performed in
theseventiesoflastcentury, whenOechsnerandGerharddeterminedtheabun-
danceof neutral dimers and trimers sputtered fromvariousmetallicsamplesby
+sub-keV Ar ions. In these experiments the neutral species were post-ionized
bymeansoflowpressureRFargonplasmasustainedbyelectroncyclotronwave
resonance[Oe74],[Oe78]. AfterseveralyearsthisworkwascontinuedbyGnaser
et al. [Gn89] and Franzreb et al [Fr90] by using an electron beam impact to
post-ionize neutral species. Due to the relatively low ionization e fficiency of
bothplasmaand electron impactpost-ionizationmethods, the sizeof sputtered
neutral clusters detected experimentally was for a long time limited to very
small clusters containing less than five atoms.
C. H. Becker [Be84], was the first to utilize the non-resonant multipho-
ton ionization to post-ionize sputtered neutral clusters. By bombarding copper
+under Ar ions at impact energy of several kilovolts, he observed atoms and
3dimers. As a consequence, laser has became routinely being used to post -
ionizing neutral sputteringspecieswith ahighpost-ionizatione fficiency [Co91],
[Co93], [Co94], [Wu93a], [Wu93b]. Using this technique, a large number of
sputtered neutral clusters for several metals has been detected. In addition,
spectra of kinetic energy distributions of sputtered particles were determined.
The mass spectrometry of secondary neutrals (SNMS) has two major advan-
tages: (i) the ionization e fficiency is increased compared to SIMS process, (ii)
matrix e ffects are reduced because the sputtered neutral particles are ionized
after they have left the target body and when the chemical environment has
lost the most of its influence [Wu001].
So far, numerous studies have been conducted to determine the factors af-
fectingtheenhancementofsputteringyieldi.e.,thenumberofparticlesemitted
per the number of incident [Be81]. These studies pointed out the secondary ion
yieldsdependstronglyontheelectronicandchemicalpropertiesofthesurfaceof
solids. Purposeful modification of surface chemistry to obtain a high secondary
ion yield has been achieved by the judicious selection of the bombarding ion
species. Inparticular,Andersen[An70],[An73]demonstrateddrasticallyhigher
positive secondary ion yields were obtained under bombardment by ions of an
+ +electronegative element e.g. O than by inert gas e.g. Ar . Andersen at-2
tributed this enhancement to the increased surface work function of oxidized
+metal. Onthebasisoftheseobservations,theoxygenionsO areroutinelyused2
for sputtering in positive secondary ion mass spectrometry (SIMS) to enhance
detection sensitivities [Be75]. In more detail, high secondary ion yields are ob-
served fortheseelementsthatcan becompletelyoxidized and formstrongionic
bondswith oxygen[Be75]. However, thesecondaryionyieldsmaybedecreased
up to several orders of magnitude either for elements that form weak bonds
+with oxygen or are only partially oxidized under O bombardment [St77].2
+Since the halogen elements (e. g. fluorine F ions) are higher reactive with
metal than oxygen, one would expect strong metal-fluoride bonds compared
to metal oxides. Therefore, higher secondary ion yields will be expected for
elements that only partially oxidize with oxygen under bombardment by pro-
jectilescontainingfluorine. Toaddressthispoint, Reuterandcoworkers[Re87],
[Re88a], [Re88b] have investigated experimentally the ionization probabilities
and relative sputtered yields produced from di fferent target metals bombarded
+ + + + +with O and F or CF . Their studies demonstrated the