The Influence of an Air Exposure on the Secondary Electron Yield of Copper
©1997
Diplomarbeit
69 Seiten
Zusammenfassung
Inhaltsangabe:Abstract:
The influence of different air exposure times on the secondary electron emission of clean copper surfaces as well as on technical copper surfaces has been studied in the context of the phenomenon of multipacting, which can limit the performance of superconducting radio-frequency (RF) cavities for particle acceleration.
The copper samples were prepared by heat treatments and in situ sputter-etching and they were investigated with a dedicated instrument for SEY measurements, by scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDX), and by Auger electron spectroscopy (AES).
After short air exposures of some seconds the maximum secondary electron yield dmax of clean copper is reduced from 1.3 to less than 1.2, due to the oxidation of the copper surface. Subsequent air exposure increases the secondary electron yield (SEY) until, after about 8 days exposure dmax is higher than 2.
Clean copper samples were also exposed to the single gases present in air to find out the reasons for the dramatic increase of the SEY after long lasting air exposures. Only oxygen and water were found to affect secondary electron emission. An oxygen exposure decreases the SEY, while pure water exposure increases the SEY, but no single gas exposure changes dmax more than 0.2.
Different methods have been tried in order to reduce the secondary electron yield of technical copper surfaces. For instance a 5 minutes air exposure of copper at 350 °C followed by a 350 °C bakeout reduces dmax to values close to unity.
This procedure was applied to the outer, copper plated conductor of the LEP2 power couplers and its influence on pre-conditioning was tested. The results are promising but further tests are needed to confirm a beneficial effect of this treatment.
Inhaltsverzeichnis:Table of Contents:
Glossaryiv
1.Introduction1
2.Basics3
2.1Secondary electron emission3
2.1.1The energy distribution of the emitted electrons3
2.1.2The secondary electron yield (SEY)4
2.1.3The SEY as a function of the primary electron energy4
2.1.4Influence of adsorbed layers of another species on the SEY5
2.1.5Influence of the work function on the SEY5
2.1.6Influence of the surface structure on the SEY5
2.2Air6
2.3Copper and copper oxidation7
2.4Vacuum basics9
2.4.1Kinetic theory of gases9
2.4.2The mean free path of a gas molecule9
2.4.3The monolayer time W9
2.4.4Gas flow regimes9
2.4.5Pumping speed S and throughput Q10
2.4.6Conductance […]
The influence of different air exposure times on the secondary electron emission of clean copper surfaces as well as on technical copper surfaces has been studied in the context of the phenomenon of multipacting, which can limit the performance of superconducting radio-frequency (RF) cavities for particle acceleration.
The copper samples were prepared by heat treatments and in situ sputter-etching and they were investigated with a dedicated instrument for SEY measurements, by scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDX), and by Auger electron spectroscopy (AES).
After short air exposures of some seconds the maximum secondary electron yield dmax of clean copper is reduced from 1.3 to less than 1.2, due to the oxidation of the copper surface. Subsequent air exposure increases the secondary electron yield (SEY) until, after about 8 days exposure dmax is higher than 2.
Clean copper samples were also exposed to the single gases present in air to find out the reasons for the dramatic increase of the SEY after long lasting air exposures. Only oxygen and water were found to affect secondary electron emission. An oxygen exposure decreases the SEY, while pure water exposure increases the SEY, but no single gas exposure changes dmax more than 0.2.
Different methods have been tried in order to reduce the secondary electron yield of technical copper surfaces. For instance a 5 minutes air exposure of copper at 350 °C followed by a 350 °C bakeout reduces dmax to values close to unity.
This procedure was applied to the outer, copper plated conductor of the LEP2 power couplers and its influence on pre-conditioning was tested. The results are promising but further tests are needed to confirm a beneficial effect of this treatment.
Inhaltsverzeichnis:Table of Contents:
Glossaryiv
1.Introduction1
2.Basics3
2.1Secondary electron emission3
2.1.1The energy distribution of the emitted electrons3
2.1.2The secondary electron yield (SEY)4
2.1.3The SEY as a function of the primary electron energy4
2.1.4Influence of adsorbed layers of another species on the SEY5
2.1.5Influence of the work function on the SEY5
2.1.6Influence of the surface structure on the SEY5
2.2Air6
2.3Copper and copper oxidation7
2.4Vacuum basics9
2.4.1Kinetic theory of gases9
2.4.2The mean free path of a gas molecule9
2.4.3The monolayer time W9
2.4.4Gas flow regimes9
2.4.5Pumping speed S and throughput Q10
2.4.6Conductance […]
Leseprobe
Inhaltsverzeichnis
ID 5432
Scheuerlein, Christian: The Influence of an Air Exposure on the Secondary Electron Yield of
Copper / Christian Scheuerlein - Hamburg: Diplomica GmbH, 2002
Zugl.: Berlin, Technische Fachhochschule, Diplomarbeit, 1997
Dieses Werk ist urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die
der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen,
der Funksendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen und der
Speicherung in Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Verwertung,
vorbehalten. Eine Vervielfältigung dieses Werkes oder von Teilen dieses Werkes ist auch im
Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der
Bundesrepublik Deutschland in der jeweils geltenden Fassung zulässig. Sie ist grundsätzlich
vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des
Urheberrechtes.
Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in diesem
Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme, dass solche
Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei zu betrachten
wären und daher von jedermann benutzt werden dürften.
Die Informationen in diesem Werk wurden mit Sorgfalt erarbeitet. Dennoch können Fehler nicht
vollständig ausgeschlossen werden, und die Diplomarbeiten Agentur, die Autoren oder
Übersetzer übernehmen keine juristische Verantwortung oder irgendeine Haftung für evtl.
verbliebene fehlerhafte Angaben und deren Folgen.
Diplomica GmbH
http://www.diplom.de, Hamburg 2002
Printed in Germany
Contents ii
Glossary ... iv
1. Introduction ... 1
2. Basics... 3
2.1 Secondary electron emission ... 3
2.1.1 The energy distribution of the emitted electrons ... 3
2.1.2 The secondary electron yield (SEY) ... 4
2.1.3 The SEY as a function of the primary electron energy ... 4
2.1.4 Influence of adsorbed layers of another species on the SEY ... 5
2.1.5 Influence of the work function on the SEY ... 5
2.1.6 Influence of the surface structure on the SEY... 5
2.2 Air ... 6
2.3 Copper and copper oxidation ... 7
2.4 Vacuum basics... 9
2.4.1 Kinetic theory of gases... 9
2.4.2 The mean free path
l
of a gas molecule... 9
2.4.3 The monolayer time
W ... 9
2.4.4 Gas flow regimes... 9
2.4.5 Pumping speed S and throughput Q... 10
2.4.6 Conductance C ... 10
2.5 Analytical techniques employed to characterise the sample surfaces... 11
2.5.1 Scanning electron microscopy (SEM)... 11
2.5.2 Energy dispersive X-ray analysis (EDX) ... 11
2.5.3 Auger electron spectroscopy (AES)... 12
3. The Experimental System... 13
3.1 The electron gun... 15
3.2 The vacuum system... 16
3.2.1 The pumping system... 17
3.2.2 Total pressure measurement... 20
3.2.3 Partial pressure measurement... 23
4. Experimental Procedures... 25
4.1 The cleaning of the samples ... 25
4.1.1 Bakeout... 25
4.1.2 Glow discharge cleaning ... 25
iii Contents
4.2 Gas exposures... 28
4.2.1 Air exposure... 28
4.2.2 Pure water vapour exposure... 28
4.2.3 Pure oxygen exposure... 29
4.3 Error estimation ... 31
5. Results... 32
5.1 Influence of an air exposure on the SEY of initially clean copper... 32
5.2 Influence of a pure oxygen exposure on the SEY of initially clean copper ... 34
5.3 Influence of a pure water vapour exposure on the SEY of initially clean copper ... 35
5.4 Influence of the other gases in air... 35
5.5 Influence of a pure water vapour exposure on oxidised copper ... 36
5.6 Influence of a bakeout on the SEY... 36
5.7 Influence of an air exposure at high temperature ... 38
5.8 Influence of an air exposure on the SEY of copper oxidised at 350
qC in air... 40
5.9 Conditioning of the LEP2 power couplers together with copper plated extensions which
were heated in air at 350
qC ... 41
6. Discussion and Outlook... 43
References ... 45
Appendix ... 47
Acknowledgments ... 65
Introduction 1
1. Introduction
Superconducting cavities have been developed to accelerate particles in powerful particle
accelerators like the Large Electron Positron collider (LEP) at CERN. They are necessary to
provide the high accelerating gradients that are needed to compensate the large energy loss of the
particles by synchrotron radiation. Superconducting cavities are also more efficient in converting
the mains power to the beam power and, hence, less power consuming than normal conducting
cavities /1/.
Figure 1; Module consisting of four superconducting cavities installed in the LEP tunnel
The performance of such superconducting radio frequency (RF) resonators can be severely
hampered by the phenomenon of multipacting.
Multipacting or resonant electron loading is an avalanche-like electron multiplication that can
occur in high vacuum. If multipacting is initiated the growing electron avalanche absorbs the
input RF power. Furthermore, the bombarding electrons can desorb high amounts of tightly
adsorbed gases from the cavity surface, thus spoiling the cavity vacuum. Finally, the cooled
cavity walls can be locally heated up by absorption of a part of the electrons energy, resulting in
a transition from the superconducting to the normalconducting state (quench).
Two conditions have to be fulfilled to initiate multipacting:
x
the primary electrons are resonant with the RF field, i.e., they hit the same target again after
an even number of RF half cycles (one surface multipacting) or they hit another target after an
odd number of RF half cycles (two surface multipacting) /2/.
x
each impacting primary electron liberates more than one secondary electron, i.e., the
secondary electron yield (SEY) of the target material is higher than unity.
Therefore, it is important to know the SEY of the materials used in superconducting cavities in
order to avoid, or at least to limit multipacting. Copper is used to coat auxiliary equipment like
power couplers, where multipacting is often observed.
2 Introduction
The SEY of pure copper can be found in literature. However, secondary electrons are emitted
from the upper monolayers of a surface only and the SEY is strongly influenced by surface
contamination. The large LEP cavities have to be exposed to air during installation or
modification. The gases present in air interact with the surfaces during the time of exposure and,
as a consequence, the SEY after exposure differs from the yield of a clean surface.
This work studies the effects which air exposures of different duration have on the SEY of clean
and technical, i.e., contaminated copper surfaces at room temperature and at elevated
temperatures.
Basics 3
2. Basics
2.1 Secondary electron emission
When electrically charged particles with sufficient kinetic energy impinge on a solid surface, this
surface emits electrons. The emitted electrons are called secondary electrons and the bombarding
particles are called primary particles. For this work only the emission induced by primary
electrons is of interest.
2.1.1
The energy distribution of the emitted electrons
The energy of emitted secondary electrons varies from very low energies up to the primary
electrons energy. A typical energy distribution for all metals is shown in Figure 2.
Figure 2; Energy distribution of electrons emitted by silver upon bombardment with 155 eV primary electrons /3/
One distinguishes mainly between three groups of secondary electrons. These are true secondary
electrons, inelastically backscattered primary electrons and elastically reflected primary electrons.
The peak indicated by a corresponds to elastically reflected primary electrons.
Most emitted electrons have low energies, corresponding to the peak indicated by c. These are
referred to as true secondary electrons, indicating that they are electrons which originally
occupied bound states in the crystal. As it is impossible to determine the origin of individual
electrons one usually regards all emitted electrons with an energy below 50 eV as true secondary
electrons.
The emitted electrons with energies between 50 eV and the primary energy are then referred to as
inelastically reflected primary electrons, i.e., incident electrons which have been backscattered
after losing part of their energy in the solid.
The two small peaks near b correspond to emitted electrons with discrete energies. They are
called Auger electrons and can be used to determine the different atomic species present on a
surface (see 2.5.3).
electron energy in eV
N (
E
)
0
50
100
150
4 Basics
2.1.2
The secondary electron yield (SEY)
Different definitions for the SEY exist. The following four coefficients can be found in literature
/4/:
1)
K: backscatter coefficient, i.e., the number of backscattered electrons (energies !50 eV) per
incident primary electron.
2)
G
s
:
the number of true secondary electrons per incident primary electron.
3)
G
t
:
ratio of the true secondary electrons and the number of the primary incident electrons
without the backscattered electrons. This coefficient is sometimes called the true yield,
because the backscattered electrons are lost for the process of electron emission.
4)
G: the total number of emitted secondary electrons per incident primary electron, i.e., the
secondary electron current i
sec
divided by the primary electron current i
P
.
G
i
i
P
sec
Equation 1
The first three coefficients are useful for the experimental verification of the theory of secondary
electron emission. As far as multipacting is concerned only the total SEY
G is relevant and hence
this definition will be followed throughout this work.
2.1.3
The SEY as a function of the primary electron energy
The SEY of a solid depends on the energy of the primary electrons. Primary electrons with very
low energy can only liberate few secondary electrons. On the other hand, high energetic electrons
will penetrate deeply into the bulk and deposit most of their energy in regions where the liberated
secondary electrons are not able to escape. Between those two cases the maximum SEY
G
max
is
found.
Figure 3; Total SEY
G of copper in dependence of the primary electron energy
Figure 3 shows the SEY of copper as a function of the primary electron energy. This curve can
be described by the following characteristic values:
0.6
0.8
1
1.2
1.4
0
500
1000
1500
2000
2500
3000
primary electron energy in eV
SEY
E
m ax
E
1
E
2
G
max
Basics 5
G
max
:
maximum SEY
E
max
:
energy at which
G
max
is attained
E
1
:
energy below which the SEY is less than unity
E
2
:
energy above which the SEY is less than unity
For metals the maximum SEY varies between 0.5 (lithium) and 1.8 (platin). For copper it is 1.3.
The yield of elemental semiconductors is also of the order of unity /5/. The maximum SEY of
insulators is considerably higher than that for metals and semiconductors. Quartz, as an example,
has a maximum SEY of about 3. For other insulators yields up to 10 have been measured /6/.
2.1.4
Influence of adsorbed layers of another species on the SEY
The maximum escape depth of secondary electrons from metals is estimated as 10 atomic
layers /7/. Assuming a thickness of 4 Å for one atomic layer, the emitted electrons origin from the
upper 4 nm of the solid. The likeliness for the low energetic secondary electrons to pass through a
condensate layer of adsorbed gases is higher. The escape depth of secondary electrons in such
layers is about 100 monolayers.
The SEY of a condensate layer can differ strongly from that of the base material. The SEY of
water condensed on a silver target at 77 K is about 2.3 /8/. If the condensate layer is only a few
atomic layers thick, the proportion of secondary electrons which are emitted there is small
compared with the number of secondaries emitted from the base material. On the other hand, the
electron emission will be determined by the condensate layer only if it is thicker than the
maximum escape depth of the secondary electrons. For a condensate some 10 monolayers thick
one can assume a SEY between that of the base material and that of the pure condensate.
However, it is not evident if the change of the SEY is due to electrons which have been liberated
in the adsorbed layer or to a change of the work function of the base material.
2.1.5
Influence of the work function on the SEY
The work function of a metal can be changed by the adsorption of atoms of another species onto
the base material. The work function of copper for instance can be increased by the adsorption of
oxygen and it can be decreased by the adsorption of water vapour.
Electron emission processes like photo-electric emission or thermionic emission depend strongly
on the work function of the emitting material. Compared to these processes the influence of the
work function on the SEY is much smaller. This can be explained by the fact that the mean
energy of the secondary electrons is considerably higher than that of electrons emitted by the
other processes.
Effects on the SEY are reported mainly for the adsorption of atoms which decrease the work
function /3/. For atoms which increase the work function like oxygen on tungsten it is reported
that they have only a slight influence on the SEY.
In a more recent theoretical investigation Hachenberg and Brauer claim that the influence of the
surface on the SEY may be neglected in satisfactory approximation /9/.
2.1.6
Influence of the surface structure on the SEY
The surface structure of a material can have a decisive influence on the SEY. Layers consisting
of very small crystallites do not present a reflecting surface but are rather optically black. Such
6 Basics
microcrystaline layers have much lower yields than smooth layers /3/. Carbon in the form of soot
for example has a SEY of 0.5. The yield of a smooth carbon layer is higher than 1.
The small SEY of a microcrystaline layer can be explained as follows. Once a secondary electron
passes through the surface of a smooth layer it meets no more obstacles and hence it contributes
to the SEY. On the other hand, secondary electrons which have been liberated in a labyrinthine
structure are likely to face other microcrystals and a big fraction will be absorbed before they can
escape from the surface. This is illustrated in Figure 4.
Figure 4; Influence of the surface roughness on the SEY. From a smooth surface the emitted electrons are more
likely to escape than from a rough surface.
2.2 Air
At sea level air consists mainly of 791 mbar N
2
, 212 mbar O
2
, 9.4 mbar Ar and 0.3 mbar CO
2
. It
also contains gaseous pollutants like SO
2
, NH
3
, NO or ozone which can have a strong impact on
corrosion processes.
Water vapour is another constituent of air and it is the most detrimental for vacuum applications.
Furthermore, it is essential to the formation of an electrolyte solution that supports
electrochemical corrosion reactions. The concentration of water vapour in the atmosphere is
usually expressed in terms of the relative humidity. This is defined as the ratio of the actual
partial pressure of the water vapour in the atmosphere compared to that which would saturate the
atmosphere at the same temperature. Most corrosion processes depend strongly on the relative
humidity. The relative humidity of common indoor atmospheres is about 60%. In other
environments like storage huts or the housing of large accelerators it can be much more critical.
The behaviour of the water molecules is strongly influenced by their shape. The two hydrogen
atoms are attached to the oxygen atom in such a way to make an angle of 104.5
q. This makes the
water molecule a dipole because the electrons are closer to the oxygen atom. Hence, the oxygen
corner of the molecule has a slightly negative charge and the hydrogen ends are slightly positively
charged. The negative part of one water molecule attracts the positive part of another molecule,
which causes a comparatively strong sticking of the molecules to each other. This is the reason
for the unusual properties of water. It has a very high heat capacity and the ability to dissolve
many different substances. It is also responsible for the high vapour pressure of H
2
O at room
temperature (23 mbar at 20
qC) compared to gases with a similar mass.
Basics 7
2.3 Copper and copper oxidation
Copper is the metallic, chemical element of atomic number 29. It has the electronic configuration
2.8.18.1. Loss of the outermost electron gives the cuprous ion Cu
+
, and a second electron may be
lost in the formation of the cupric ion Cu
2+
.
Cu consists of the two stable isotopes Cu63 (69.1%) and Cu 65 (30.9%) and has an atomic
weight of 63.5. The melting point of pure copper is 1083
qC and its boiling point is 2563qC.
Copper is frequently used for many technical applications, mainly because of its excellent thermal
and electrical conductivity, its good workability and its high corrosion resistance. For ultra high
vacuum applications Cu is favourable because of its low gas solubility (relatively low outgassing)
and its vacuum tightness which allows the employment of relatively thin-walled apparatus.
After the extraction of copper from its ores it is available as low purity crude copper with about
2% impurities. The crude copper can than be refined by electrolysis (electrolytic copper). To
purify Cu electrolytically, the impure Cu is made the anode in an electrolytic cell. A thin sheet of
already purified Cu acts as the cathode. With a suitable electrolyte and appropriate operating
conditions very pure ETPC (electrolytic tough pitch copper) can be deposited on the cathode.
ETPC contains some oxygen, chiefly as Cu
2
O. Oxygen lowers the melting point, the strength and
the conductivity of the pure copper. It also leads to fine cracks if heated in H
2
atmosphere
(`hydrogen sickness'). This is due to H
2
, which readily diffuses into heated Cu and combines with
the oxygen present to water vapour:
Cu O
H
Cu
H O
2
2
2
2
o
Equation 2
The water vapour has a low diffusion rate in copper and so builds up pressures of several
thousand bar in the bulk, leading to microscopic cracks /10/.
The amount of oxygen can be reduced by heating the copper in a carbon monoxide atmosphere.
CO is not soluble in Cu but combines readily with the oxygen at the copper surface. This
phenomenon is utilised to produce the so called OFHC (oxygen-free high-conductivity) copper.
OFHC copper is immune against the hydrogen embrittlement (as long as it has not been oxidised
by heating it in air or oxygen).
For ultra high vacuum equipment, which usually has to be bakeable, OFHC copper can be used.
OFHC-regular contains less than 0.05% impurities. Particularly suitable for high vacuum
applications is OFHC-certified, which contains less than 0.02% impurities. The purest grade of
copper commercially available is GFHP (gas-free high purity) copper with a minimum copper
content of 99.993%.
The vapour pressure and the rate of evaporation of copper are relatively high. Therefore, the
operating temperature of vacuum equipment should not exceed 550
qC.
Copper can be oxidised to cuprous oxide (Cu
2
O or copper(I)-oxide) and to cupric oxide (CuO or
copper(II)-oxide).
The reddish Cu
2
O is a p-type semiconductor formed by the chemical process:
4
2
2
2
Cu
O
Cu O
o
Equation 3
Pure CuO is black and it can be formed when powdered copper is heated in air or oxygen:
2
2
2
Cu
O
CuO
o
Equation 4
8 Basics
Bulk copper under the same conditions forms a mixture of CuO and Cu
2
O. At room temperature
in dry air a thin, invisible film of Cu
2
O is formed. At temperatures higher than 250
qC the surface
tarnishes. Then it consists of an adherent inner layer of Cu
2
O and a comparatively thin outer
layer of the black CuO. At higher temperatures the amount of CuO existing is determined by the
pressure of oxygen /11/.
100 200 300 400 500 600 700 800 900 1000
10E-4
10E3
10E2
10
1
10E-1
10E-2
10E-3
CuO
Cu
p(ai
r)
in
m
b
ar
Temperature in
°C
Cu O
2
Figure 5; Temperature/pressure regions of formation of Cu
2
O and CuO /12/
Figure 5 shows the dependence of the oxygen pressure and temperature on the oxide film
composition. The oxygen pressure/temperature curves enclose the regions in which CuO or Cu
2
O
is the outermost oxide of the film. The decomposition pressure of CuO at 1000
qC corresponds
approximately with the partial pressure of oxygen in air. Therefore, at temperatures above
1000
qC only Cu
2
O is formed in air.
For the corrosion reaction to proceed, copper ions and electrons have to migrate through the
Cu
2
O layer, which is predominantly responsible for protection /13/. Hence, the rate of oxidation
decreases with the thickness of the Cu
2
O layer.
At temperatures up to 100
qC the oxide film grows linearly with the logarithm of time and above
400
qC the oxidation of Cu follows the parabolic law /14/.
In dry air at atmospheric pressure and room temperature Cu is not attacked superficially.
However, in moist air containing carbon dioxide Cu initially darkens because of the build up of a
thin layer of black CuO. After several years a bluish green layer of basic copper carbonate called
patina is formed.
An oxidation of Cu by pure water vapour takes not place even at high temperatures up to 1000
qC
/15/.
Besides heating copper in air, cuprous oxide can be formed for example by anodic oxidation of
copper or by reactive sputtering of copper in gas mixtures including oxygen.
Because of its comparatively low dissociation pressure cuprous oxide can be used as oxidiser in a
catalytic pump for hydrogen pumping. In such a pump molecular hydrogen is dissociated by
means of a hot metal filament (atomising catalyst) and the atomic hydrogen is absorbed on a
Basics 9
cuprous oxide layer where it is oxidised to H
2
O. The water vapour is desorbed and condenses
finally on a liquid nitrogen trap. A Cu
2
O temperature of 200
qC has been found to be sufficient to
rapidly desorb the produced H
2
O from the cuprous oxide /16/.
2.4 Vacuum basics
2.4.1
Kinetic theory of gases
For calculations in vacuum technology one assumes the gas molecules as pointlike particles
which only interact elastically and are not able to condense at the ambient temperature. Such a
gas is called a perfect gas and it can be described by the perfect gas law:
p V
N k T
Equation 5
where p, V and T correspond to the gas pressure, volume and absolute temperature respectively.
N
is the number of molecules and k is Boltzmann's constant.
The result of continual elastic collisions and exchange of energy of the residual gas molecules is
the Maxwell-Boltzmann velocity distribution. The mean velocity
c
(arithmetic mean value) of a
gas particle can be calculated:
c
R T
M
8
S
Equation 6
with the molar mass M and the molar gas constant R.
2.4.2
The mean free path l of a gas molecule
The mean free path
l
is the distance which every molecule travels on the average between two
collisions with other particles. It is a function of the particle diameter 2r and the number density
of molecules n. In good approximation it can be calculated as:
l
n
r
1
2
2
2
S
Equation 7
2.4.3
The monolayer time
99
The time which is needed for the formation of a monomolecular or monatomic layer on a
substrate surface is called the monolayer time
W.
W
4 a
n c
Equation 8
With the assumption that every particle arriving at this surface sticks to it,
W is proportional to the
number of free places a on the surface and inversely proportional to the number of molecules
hitting the surface per unit time and hence the pressure.
2.4.4
Gas flow regimes
Whenever there is a directed net movement of gas in a system under the influence of pumps, the
gas flows. Three different regimes of gaseous flow exist. These are viscous flow, Knudsen flow
10 Basics
and molecular flow. The kind of flow depends on the geometry of the vacuum system as well as
the pressure, temperature and type of the present gas.
The molecular flow occurs at low gas densities when the mean free path of the gas molecules is
greater than the diameter of the conducting tube. In this region the gas molecules have virtually
no influence on each other and their motion is strictly random.
Viscous flow prevails in rough vacuum when the mean free path of the gas molecules is much
smaller than the dimensions of the vacuum system. The mutual interactions of the particles with
each other determine the character of this regime so that the flow is limited by its viscosity. At
low gas velocities the flow is laminar where parallel flow lines may be imagined. At higher
velocity the flowing gas layers are no longer parallel but cross each other. This condition is called
turbulent flow.
At pressures where the mean free path of the molecules is similar to the dimensions of the
vacuum enclosure, the flow of the gas is governed by viscosity as well as by molecular
phenomena. This regime is called Knudsen flow or intermediate flow.
The expressions rough-, medium- and high vacuum are related to the flow regimes viscous-,
Knudsen- and molecular flow respectively.
2.4.5
Pumping speed S and throughput Q
The pumping speed S is defined as the volume V of gas passing through a pump per unit time t.
S
dV
dt
Equation 9
The throughput Q describes the gas load which is removed by a pump per unit time. It is defined
as the product of the pumping speed and the pressure p at the pump inlet, i.e.,
Q
S p
Equation 10
2.4.6
Conductance C
The throughput of gas through any conducting element is proportional to the difference between
the pressures at the entrance and at the exit of the conducting element.
Q
C p
p
2
1
Equation 11
The constant C in this equation is called the conductance between the two points where p
1
and p
2
were measured. Hence, the conductance is a figure which describes the likeliness for the gas
molecules to travel through a vacuum enclosure according to its geometry. Therefore, short pipes
with large cross sections have a high conductance.
In the molecular flow regime the conductance is independent of the pressure. In the viscous flow
region the conductance is a function of the pressure and its effective value is larger than that in
the molecular flow regime.
For a tube which length l is big compared with its inner diameter d
i
, the conductance for gas
molecules with a mean velocity
c
can be calculated:
C
d
l
c
i
S
12
3
Equation 12
Basics 11
When conductances are joined in series, the system conductance C
sys
is given by:
1
1
1
1
1
2
3
C
C
C
C
sys
...
Equation 13
One can imagine a perfect pump as a conductance with an inlet pressure p and an outlet pressure
of zero. The pumping speed of such a pump S is than equal to its conductance and it is
determined by its effective cross section only. If this pump is joined to a vacuum vessel via a
vacuum pipe with a conductance C, the effective pumping speed S
eff
can be calculated:
1
1
1
S
S
C
eff
Equation 14
2.5 Analytical techniques employed to characterise the sample surfaces
2.5.1
Scanning electron microscopy (SEM)
A commonly used tool to investigate surface structures is the SEM. The samples have to be
placed in a high vacuum chamber and an electron beam is scanned over the chosen surface area.
Several signals resulting from this primary electron bombardment can be detected and imaged.
The most common image mode detects the low energetic secondary electrons (in 2.1.1 referred to
as true secondary electrons). They originate from a surface depth no larger than several Å. The
signal is captured by a so called Everhart-Thornley detector which consists basically of a
scintilator-photomultiplier combination. The detector output serves to modulate the intensity of a
cathode ray tube, which is rastered in synchronism with the primary beam.
Contrast variation can be caused by the difference in secondary electron emission of the materials
present on the investigated surface. The topological contrast, however, has much more influence
on the image. Different incident angles of the primary electrons result in variations of the
secondary electron emission. At edges the volume from which secondary electrons can escape is
particularly high and as a result edges appear brightest on the screen.
The image magnification is the ratio of scan lengths on the cathode ray tube to that on the
specimen. If the area on the screen is kept constant, the magnification is determined by the size of
the surface area above which the primary beam is scanned.
The resolution of the SEM is mainly limited by the minimum spot size of the primary beam which
can be achieved with a sufficiently high signal to noise ratio. The highest with a SEM attainable
lateral resolution is about 50 Å /17/. Apart from the high resolution, the great depth of focus is a
major advantage of the SEM.
Backscattered electrons can also be detected for imaging. The probability of backscattering
increases with the atomic number Z of the sample material. Therefore, contrast can develop
between regions of the sample that differ widely in Z. Since the escape depth for high energetic
backscattered electrons is much greater than for low energetic secondary electrons, there is much
less topographical contrast in such images.
2.5.2
Energy dispersive X-ray analysis (EDX)
X-rays are another signal emitted by the samples in a SEM under electron bombardment. Besides
bremsstrahlung, a discrete spectrum of characteristic X-rays can be detected in order to identify
the emitting atoms. Characteristic X-rays are emitted, when an electron from an outer shell
12 Basics
lowers its energy by filling an electron vacancy in an inner shell, which has been created by
interaction of the atom with a primary electron. If the number of X-rays of a certain energy is
counted, the concentration of atoms in a specimen can be determined.
An X-ray detector registers the emitted photons and delivers pulses with a voltage amplitude
proportional to the photon energy. These pulses are amplified and then sorted according to their
amplitude by a multichannel analyser. This analyser also counts and stores the number of pulses
within given increments of the voltage range. The result is an X-ray spectrum, characteristic for
the elements present on the investigated sample area.
Lowenergetic photons are absorbed by the window which protects the X-ray detector from
contamination. For this reason, it is difficult to detect elements with a low Z. With a standard
detector only elements with an atomic number higher than 10 can be registered /18/. The spectral
resolution of energy dispersive X-ray analysis systems is limited to about 150 eV.
2.5.3
Auger electron spectroscopy (AES)
Instead of characteristic X-rays, electrons with a characteristic energy can be emitted by excited
atoms when they decay to their ground state. These are called Auger-electrons and they can serve
to identify the emitting atoms. The processes of Auger transition and X-ray emission go on
simultaneously. In the low-Z elements, the probability is greater that an Auger transition will
occur, whereas X-ray emission is favoured for elements with a high Z.
When the Auger transitions occur within a few angstroms of the surface, the Auger electrons may
be ejected from the surface without loss of energy and contribute to the total spectrum of ejected
electrons at a distinct characteristic energy (see 2.1.1). The Auger peaks appearing in the energy
distribution N(E) are small compared with the background current and the use of N(E) is in
general inadequate for analytical purposes. Instead the differential distribution dN(E)/dE is
recorded. The differentiation can largely eliminate the background and makes the peaks more
easy to identify.
The sensitivity of the Auger technique is determined by the transition probability of the Auger
transitions involved, the incident beam current and energy and by the collection efficiency of the
analyser. With a 3 keV, 50
PA beam and a high sensitivity cylindrical mirror analyser, the limit
of detection for the elements varies between approximately 0.02 and 0.2 atomic percent with
scanning rates of 1 eV per second /19/. Apart from hydrogen and helium all elements can be
detected by AES.
Since the escape depth of the Auger electrons is only a few atomic layers, AES has a high depth
resolution. A depth profile can be obtained if the sample surface is continuously removed by
sputter etching and the Auger peaks for the elements of interest are monitored simultaneously.
Details
- Seiten
- Erscheinungsform
- Originalausgabe
- Jahr
- 1997
- ISBN (eBook)
- 9783832454326
- ISBN (Paperback)
- 9783838654324
- DOI
- 10.3239/9783832454326
- Dateigröße
- 5.4 MB
- Sprache
- Englisch
- Institution / Hochschule
- Beuth Hochschule für Technik Berlin – Verfahrens- und Umwelttechnik
- Erscheinungsdatum
- 2002 (Mai)
- Note
- 1,0
- Schlagworte
- sekundärelektronen vervielfachung kupfer oberflächen luftexposition
