Metal Clusters: Proceedings of an International Symposium, Heidelberg, April 7–11, 1986

229 Pages·1986·8.19 MB·English

by  E. W. Becker (auth.), Professor Dr. Frank Träger, Professor Dr. Gisbert zu Putlitz (eds.)

Metal Clusters
Metal Clusters
Proceedings of an International Symposium,
Heidelberg, April 7-11, 1986
Editors: F. Trager and G. zu Putlitz
With 219 Figures
Springer-Verlag Berlin Heidelberg New York
London Paris Tokyo
Professor Dr. Frank Trager
Professor Dr. Gisbert zu Putlitz
Physikalisches Institut der Universitiit Heidelberg, Philosophenweg 12, D-6900 Heidelberg
This book originally appeared as the journal
Zeitschrift fiir Physik D - Atoms, Molecules and Clusters, Volume 3, Number 2 and 3
(ISSN 0178-7683) © Springer-Verlag Berlin, Heidelberg 1986
ISBN-13: 978-3-642-71573-0 e-ISBN-13: 978-3-642-71571-6
DOl: 10.1007/978-3-642-71571-6
This work is subject to copyright. All rigths are reserved, whether the whole or part of the material is concerned,
specifically those of translation, reprinting, reuse of illustrations, broadcasting, reproduction by photocopying
machine or similar means, and storage in data banks. Under § 54 of the German Copyright Law where copies
are made for other than private use, a fee is payable to "Verwertungsgesellschaft Wort", Munich.
© Springer-Verlag Berlin Heidelberg 1986
Softcover reprint of the hardcover 1s t edition 1986
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specif
ic statement, that such names are exempt from the relevant protective laws and regulations and therefore free
for general use.
Offset printing: Weihert-Druck GmbH, D-6100 Darmstadt
Bookbinding: 1. Schliffer OHG, D-6718 Griinstadt
2153/3150-543210
Preface
This volume contains papers which have been presented at the International Sym
posium on Metal Clusters in Heidelberg from April 7-11, 1986.
Clusters, and in particular metal clusters, have been the topic of fa~t growing scientific
interest. Indeed, clusters constitute a field of interdisciplinary nature where both physical
and chemical questions have to be addressed. Clusters are offundamental importance for
the deeper understanding of the transition from atoms via molecules and larger aggregates
of particles to the properties of solid materials. Moreover, metal clusters and their character
istics are of vital significance for such applied topics as catalysis or photography.
Experimentally, the field exhibited rapid progress in the last years. Different sources for
clusters have been developed. Intense beams made possible the investigation of free
neutral clusters and cluster ions as well. Even though a number of issues concerning metal
clusters is still discussed controversially, the present volume tries to give an overview of
current work in this field and to illustrate the large variety of experiments as well as the
advances made possible by modern theoretical methods. Looking at the many interesting
questions still to be addressed it is fair to propose a rapid further growth of this field.
The International Symposium on Metal Clusters in Heidelberg was the inaugurate
scientific meeting in the "Internationales Wissenschaftsforum Heidelberg", founded on the
occasion of the 600th anniversary of the Ruprechts-Karls-Universitiit Heidelberg as an
endavour to promote scientific discussion for all disciplines. Already in this very first
symposium the basic goals ofthe "International Science Forum Heidelberg" became appar
ent: Exploitation of a special scientific topic or field with a limited number of participants
ranging from experts to the advanced graduate student, promoting informal scientific
discussions in a truly international atmosphere, and accessing the present knowledge as
well as state ofthe art in the field ofinterest. Research has been and will be the keystone for
the development of our universities.
The present meeting has benefited largely by the advice of the members ofthe Interna
tional Advisory Board. Help in the organization and the local arrangements has been
obtained by the directorate and staff of the Internationales Wissenschaftsforum Heidelberg
and the Physics Institute of the University. Financial support has been provided by the
Deutsche Forschungsgemeinschaft, by the State of Baden-Wiirttemberg, and by the
University of Heidelberg. In addition, some contributions have been made by private spon
sors. To all of those we owe our gratitude.
July 1986 G. zu Putlitz
F. Trager
Contents
On the History of Cluster Beams
By E. W. Becker (With 14 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Shell Structure and Response Properties of Metal Clusters
By W. D. Knight, W. A. de Heer, and W. A. Saunders (With 5 Figures) . . . . . . . . . . . . . 9
Shell Closings and Geometric Structure Effects. A Systematic Approach to the
Interpretation of Abundance Distributions Observed in Photoionisation Mass
Spectra for Alkali Cluster Beams
By M. M. Kappes, P. Radi, M. Schar, C. Yeretzian, and E. Schumacher
(With 1 Figure)................................................................... 15
Evolution of Photoionization Spectra of Metal Clusters as a Function of Size
By C. Brechignac and Ph. Cahuzac (With 8 Figures). . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
Spectroscopy ofNa3
By M. Broyer, G. Delacretaz, P. Labastie, R L. Whetten, J. P. Wolf, and L. Waste
(With 7 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 31
The Formation and Kinetics of Ionized Cluster Beams
By I. Yamada, H Usui, and T. Takagi (With 8 Figures) ............................ 37
On the Phase of Metal Clusters
By 1. Gspann (With 2 Figures) . ............................ ........... ...... . . . . .. 43
General Principles Governing Structures of Small Clusters
By J. KouteckY and P. Fantucci (With 5 Figures) .................................. 47
Geometrical Structure of Metal Clusters
By 1. Buttet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 55
Electronic Structure and Bonding in Clusters: Theoretical Studies
By K Hermann, H 1. Hass, and P. S. Bagus (With 4 Figures) ...................... 59
Metallic Ions and Clusters: Formation, Energetics, and Reaction
By A. W. Castleman, Jr. and R G. Keesee (With 11 Figures) ...................... 67
Experiments on Size-Selected Metal Cluster Ions in a Triple Quadrupole
Arrangement
By P. Fayet and L. Waste (With 8 Figures) .. . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . .. 77
Sputtered Metal Cluster Ions: Unimolecular Decomposition and Collision
Induced Fragmentation
By W. Begemann, S. Dreihafer, K H Meiwes-Broer, and H O. Lutz
(With 5 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . .. 83
A Penning Trap for Studying Cluster Ions
By H-J. Kluge, H Schnatz, and L. Schweikhard (With 6 Figures). . . . . . . . . . . . . . . . .. 89
The Chemistry and Physics of Molecular Surfaces
By A. Kaldor, D. M. Cox, D.1. Trevor, and M. R Zakin (With 9 Figures) . . . . . . . . . .. 95
Analysis of the Reactivity of Small Cobalt Clusters
By A. Rosen and T. T. Rantala (With 2 Figures) ................................... 105
Compound Clusters
By T. P. Martin (With 14 Figures) ................................................. 111
Structural and Electronic Properties of Compound Metal Clusters
By B.KRao, S.N.Khanna, and P.Jena (With 4 Figures) .......................... 119
Binary Metal Alloy Clusters
By K Sattler (With 8 Figures) ..................................................... 123
Cluster Compounds Help to Bridge the Gap between Atom and Solid
By H MUller (With 4 Figures) ..................................................... 133
Systems of Small Metal Particles: Optical Properties and their Structure
Dependences
By U. Kreibig (With 13 Figures) ................................................... 139
Synthesis and Properties of Metal Clusters in Polymeric Matrices
By E.Kay (With 14 Figures) ...................................................... 151
Guest-Host Interaction and Photochemical Transformation of Silver Particles
Isolated in Rare Gas Matrices
By P. S. Bechthold, U. Kettler, H R Schober, and W. Krasser (With 10 Figures) .... 163
Ionized Cluster Beam Technique for Thin Film Deposition
By T. Takagi (With 13 Figures) .................................................... 171
Growth and Properties of Particulate Fe Films Vapor Deposited in UHV on Planar
Alumina Substrates
By HPoppa, c.A.Papageorgopoulos, EMarks, and E.Bauer (With 12 Figures) ... 179
Atom Desorption Energies for Sodium Clusters
By M. Vollmer and E Trager (With 5 Figures) ..................................... 191
The Role of Small Silver Clusters in Photography
By P. Fayet, E Granzer, G. Hegenbart, E. Moisar, B. Pischel, and L. Waste
(With 4 Figures) .................................................................. 199
Magnetic Measurements on Stable Fe(O) Microclusters. Part 2
By E Schmidt, A. Quazi, A. X. Trautwein, G. Doppler, and H M. Ziethen
(With 4 Figures) .................................................................. 203
Photofragmentation of Mass Resolved Carbon Cluster Ions
By M. E. Geusic, M. F. Jarrold, T. J. Mclirath, L. A. Bloomfield, R R Freeman,
and W. L. Brown (With 7 Figures) ................................................. 209
Decomposition Channels for Multiply Charged Ammonia Clusters
By D.Kreisle, KLeiter, O.Echt, and T.D.Miirk (With 2 Figures) ................. 219
An Improved Clusterion-Photoelectron Coincidence Technique for the
Investigation of the Ionisation Dynamics of Clusters
By L. Cordis, G. Gantefdr, J. HeBlich, and A. Ding (With 6 Figures) ............... 223
Study of the Fragmentation of Small Sulphur Clusters
By M. Arnold, J. Kowalski, G. zu Putlitz, T. Stehlin, and E Trager
(With 4 Figures) .................................................................. 229
Metal Clusters and Particles: A Few Concluding Theoretical Remarks
By B. MUhlschlegel ............................................................... 235
List of Contributors .................................................................. 239
On the History of Cluster Beams
E.W. Becker
Kernforschungszentrum und Universitat Karlsruhe,
Institut fur Kernverfahrenstechnik, Karlsruhe, Federal Republic of Germany
Received April 23, 1986; final version May 15, 1986
The methods to produce and investigate cluster beams have been developed primarily
with the use of permanent gases. A summary is given of related work carried out at
Marburg and Karlsruhe. The report deals with the effect of carrier gases on cluster
beam production; ionization, electrical acceleration and magnetic deflection of cluster
beams; the retarding potential mass spectrometry of cluster beams; cluster size measure
ment by atomic beam attenuation; reflection of cluster beams at solid surfaces; scattering
properties of 4He and 3He clusters; the application of cluster beams in plasma physics,
and the reduction of space charge problems by acceleration of cluster ions.
PACS: 36.40
First Demonstration of Cluster Beams
The history of cluster beams dates back to 1956 when,
at the Physics Institute of the University of Marburg,
we tried to produce intense molecular beams at low
temperature. We had built a nozzle source as pro
posed by Kantrowitz and Grey [1], and equipped it
with a cooling system. To measure the velocity distri
bution of the beam, a time-of-flight analyser was ap
plied as depicted in Fig. 1. Figure 2 shows time-of
flight distributions obtained with an argon beam pro
duced at constant source pressure but at different
source temperatures. Obviously, the single peak ob
served at 202 K splits up into two peaks at 187 K. Chopper Nozzle
At still lower temperatures the second peak becomes
source
predominant. The temperature and pressure depen
dence of the effect suggested condensation of the ar Fig. 1. Nozzle source and time-of-flight analyser used in the first
gon atoms as the reason [2]. The lower velocity of demonstration of cluster beams in high vacuum [2]
the second peak exhibits a slip between the clusters
and the uncondensed gas. When hydrogen was used as the feed gas and liq
Condensation or cluster formation in a supersonic uid hydrogen as the cooling fluid, the results were
flow was already well known from wind-tunnel exper similar. In Fig. 3, which shows the dependence of in
iments. It was, however, not previously known that tensity on source pressure of a hydrogen beam, clus
clusters formed under conditions of gas dynamics ter formation is indicated by the steep rise in intensity
could be transferred into a high vacuum and sepa [2].
rated almost completely from the non-condensed re For our basic research, which aimed at measuring
sidual gas. collision properties of single atoms or molecules, the
2 E.W. Becker: History of Cluster Beams
cluster beams became one of the main research topics
of the Institut fUr Kernverfahrenstechnik at Karls
ruhe (founded in 19581).
Considering the tremendous amount of work
To = 169K which has been done in so many laboratories with
cluster beams since that time, it is impossible to dis
cuss here all contributions. Therefore, as agreed with
the chairman of the meeting, the report will be re
stricted to the contributions by the Karlsruhe insti
tute 2. As work there has been with permanent gases,
the short history of cluster beams is a history of van
der-Waals cluster beams. A presentation of this at
To =1 87K a meeting dealing mainly with metal clusters can be
justified by the fact that the methods to produce and
investigate metal cluster beams and van-der-Waals
cluster beams are almost the same, and the boundary
between metal clusters and van-der-Waals clusters is
To =202K
not always clear.
Fig. 2. Time-of-flight distribution of an argon beam at different
temperatures. The second peak which appears at lower temperature Effect of Carrier Gases on Cluster Beam Production
indicates cluster formation [2]
In an attempt to improve the intensity of cluster
beams we investigated the influence of carrier gases,
6 . 10 17 ~------r------'
which, under the conditions applied, should not con
dense but help the feed gas to get rid of its heat
of condensation [5]. It is obvious from the upper part
of Fig. 4 that at liquid hydrogen temperature adding
300% He brings up the intensity of a hydrogen cluster
beam by about a factor of three. For the case where
4
nitrogen is the condensing gas and hydrogen the car
l""U J----,
= rier gas, which is shown in the lower part of Fig. 4
,!! u
u UCIJ) with liquid nitrogen as the cooling fluid, the gain
,!!NE in intensity is more than a factor of thirty. In both
Q
lEu cases, the cluster beam is almost free from the carrier
gas. It should be noted that addition of a light carrier
2
gas enhances the velocity of the clusters, an effect
which may be of interest in special applications.
Onset of
condensation
Ionization, Electrical Acceleration
and Magnetic Deflection of Cluster Beams
o
o
50 100
In 1961 Henkes demonstrated with CO2 [6] and a
Po [Torr]
year later with H2 [7] that cluster beams can be ion
Fig. 3. Dependence of intensity on source pressure of a hydrogen
ized by electron bombardment, so enabling mass
beam at To = 20 K. The steep rise in intensity indicates cluster for
spectra of cluster beams to be produced. Figure 5
mation [2]
shows the distribution of mass number M per unit
formation of clusters was a severe drawback as it 1 It may be of interest that the strong separation of light and heavy
limited the attainable intensity of a beam consisting particles, which we had observed in the course of the work on
of single atoms or molecules. On the other hand, the production of intense molecular beams, was the starting point of
our development at the same institute of the so-called separation
possiblity of producing cluster beams in high vacuum
nozzle process for uranium enrichment [3]
was opening up interesting aspects in basic research
2 Hagena [4] has surveyed most of the work on cluster beams
as well as in practical application. Hence, work on from nozzle sources done up until 1972
E.W. Becker: History of Cluster Beams 3
3-1019r-~~~-'~~~~--r-~~~---' For smaller clusters multiple ionization results in
o
fragmentation [8].
J. (MOleCUleS)
cm2·sec On account of multiple charging and fragmenta
2 tion, the distribution of the specific mass M/Z of
the cluster ions cannot be expected to be identical
with the original mass distribution of the cluster
beam. However, it gives a good measure of the mean
size of the original clusters.
o
o Retarding Potential Mass Spectrometry
of Cluster Beams
8'ld8r-~~~--'-~--""---------'
With higher values of the specific mass the magnetic
j (MOlecules.!
cm2.s ec I deflection of electrically accelerated cluster ions be
6 comes more and more ineffective. Therefore, Bau
chert and Hagena introduced what is called retarding
potential mass spectrometry [9]. It profits from the
fact that all the cluster ions in an ionized cluster beam
have practically the same velocity so that their kinetic
2 energies are proportional to their masses. Figure 6
shows the mean number of molecules N per unit
o charge Z of ionized CO cluster beams measured by
2
o
a retarding electric field. It is obvious from the figure
that the mean specific size N/Z increases considerably
Fig.4a and b. Enhancement of cluster beam intensity by a non
with increasing source pressure and increasing nozzle
condensing carrier gas. a To::::::20 K; b To:::::: 78 K [5]. Ata H2 and
Ata N2=partial pressure ofH2 and N2 respectively, given in atm diameter. Hagena and co-workers derived similarity
3
6 J
I. I N/Z
I
"
po. 220Torr ~
J
4
V
V
2 / 6
I,.t'
ItJ~
o 3
10 190 1000
Mil
Fig. 5. Distribution of mass number M per unit charge Z of hydro
gen cluster ions obtained by electron bombardment, electrical accel-
eration and magnetic deflection [7] 10 2/-----j~--J'+---_t+--_H~---_l
6
charge Z of hydrogen cluster ions obtained by elec
tron bombardment, electrical acceleration and mag
netic deflection.
The ionization process model that emerged from 3 6 104 3
various experiments assumes that the loss in cluster Po [Torr)
mass by partial evaporation is negligible and that the
Fig. 6. Mean number of molecules N per unit charge Z of CO2
observed changes in the cluster ion spectrum are due
clusters measured by a retarding electric field: a = nozzle diameter,
to multiple ionization of the cluster by one electron. Po = source pressure [9]
4 E.W. Becker: History of Cluster Beams
laws for the production of cluster beams [4] from good agreement with the data based on potassium
such measurements. beam scattering [11], thus indicating the internal con
sistency of the various assumptions made.
Cluster Size Measurement
by Atomic Beam Attenuation Reflection of Cluster Beams at Solid Surfaces
From retarding potential mass spectrometry as well One of the most striking effects with cluster beams
as from magnetic mass spectrometry only a mean is the enhancement of flux density attainable by re
size of the cluster ions can be derived. To obtain di flection at a smooth surface, which we first demon
rectly information about the mean number of mole strated in 1967 [12]. The left part of Fig. 8 depicts
cules in the clusters of the original beam, Burghoff the reflection of a nitrogen cluster beam of about
and Gspann measured the attenuation of a potassium 5 mm diameter at a polished steel plate at room tem
oven beam by a nitrogen cluster beam [10]. Figure 7 perature and an angle of incidence of 75°. The re
shows that at a source temperature of 78 K, which flected beam is concentrated in a lobe of about 1 mm
is low enough for nitrogen to form clusters, the effec thickness at an angle of reflection of about 88°. The
tive cross section of a nitrogen molecule in the beam size of the beam perpendicular to the plane of inci
drops by two orders of magnitude when the source dence remains practically unaffected. It can be seen
pressure is brought up to 700 Torr. The effect is a from the right part of the figure that the reflection
consequence of the mutual screening of molecules results in an enhancement of flux density of up to
within a cluster. Assuming the clusters to be spherical a factor of 5. The increase in the reflected flux density,
and having the density of solid nitrogen, Burghoff compared to the incident cluster beam, suggests the
and Gspann calculated the mean number of mole possibility of collimating a cluster beam with suitably
cules per cluster, as shown in the lower part of the shaped reflectors [12]. At a cooled target even He
figure. Later, Gspann repeated the size measurements clusters can be reflected [12 a].
with a time-of-flight mass spectrometer and obtained
incident beam I I
, =5mm 1
103
4 -
6 ~ i =7 5°
3 polished 3
steel plate
To=293K I
2 ;I ;
102 , =1mm 1-m---m. I
~
a eff 6 \ , j ~.
3 -- o
[10-16 cm2] reflected beam 80°
10 To=7BK- Ilr
6 ~ i Fig. 8. Enhancement of flux density j of a nitrogen clusl ;r beam
I"..
"" by reflection at a polished steel plate [12]
3
~
1
molecules II;
:;: I
cluster II
I
102
o 200 400 600 . -
~,.
Po [Torr]
Fig. 7. a Effective cross section (Jeff of an N2 molecule in a molecular
beam (To = 293 K) and in a cluster beam (To = 78 K), varying with Fig. 9. Time-of-flight distribution of a helium beam at To = 4 K,
source pressure Po. b Number of N2 molecules per cluster calculated Po = 700 Torr. The extremely high Mach number which follows
from the drop of (Jeff [10] from the time-of-flight distribution indicates cluster formation [14]