PhD Research Activity

Bicovalent mixed clusters study Motivations

T

he study of matter on various scales is a fascinating subject of interest in those last few years. Understanding of complex objects and research for new materials are few goals which could motivate the increasing interest for atoms, molecules and clusters. Thus, even if applications are also important, the fundamental physics involved in this phenomena are of major importance. In a column of the periodic table of elements properties of atoms are generally similar. But, in the case of carbon and silicon, these elements seems to break this rule. Effectively, in the bulk phase, carbon and silicon do not have the same structures. At a lower level, clusters of silicon and carbon also exhibit such differences. For example, carbon clusters could lead to fullerenes structures but silicon prefers tridimentionnal ones. So the question of doping carbon or silicon clusters with each others occurs naturally to eventually produce and characterize new materials obtained by Low Energy Cluster Beam Deposition. Cluster production and analysis To produce these clusters we are using a laser vaporization source, where a Nd:YAG beam focus on a rod target create a plasma of the desired element. A pulse of helium is then introduced in the source to initiate clusterization. The clusters are ionized by a ArF laser and then analyzed using Time Of Flight Mass Spectroscopy assisted by a reflectron. Another possibility is to study the fragmentation of the clusters under shining of a focused XeCl laser. We are then able to reconstruct the dissociation pathways of a selected size. Those two different techniques (mass spectrometry and photo-fragmentation) gives tools to have a look inside the structure of clusters. In order to obtain mixed we are using mixed target with various composition. We made these target in the laboratory with silicon and carbon powders, as thermodynamic constraints only allow stœchiometric crystal, i.e. 50% carbon and 50% silicon. As this technique is not suitable for all possible compositions (particularly for the silicon rich ones) another source with two rods have been developed in the laboratory. Tuning the parameters of the two lasers allows to modify the plasma average composition and thus the cluster composition. Carbon-doped silicon clusters After a study of the pure silicon clusters already begun for few months in the laboratory, we confirmed different works on that subject. But we also found a amazing behavior during the unimolecular evaporation of size selected silicon clusters. The study of carbon-doped silicon clusters indicates a structure close to pure silicon ones, that could be expected from a reasonable guess. Fortunately, we also found an interesting feature during the photofragmentation of selected SinC+ clusters. Effectively, the statistically favored Si6 fragment

1

versus Si5C one never appears, which is very puzzling. (a article is in progress on that specific part of my work) Evidences for heterofullerenes Mass spectrometry of clusters produced with a carbon rich rod (90% C – 10% Si) proved the existence of fullerenes structures with substitution of one or two carbon atoms by silicon ones. These new structures are often referred as heterofullerenes in literature with other elements. On the figure, one can see another argument for the existence of silicon-doped fullerenes with a photo-fragmentation scheme similar to pure fullerene one. The loss of Si2 as a first fragmentation step and the absence of SiC evaporation steps gives a clue about the relative position of the two silicon atoms in the structure. This assumption is checked by ab initio calculations since the most stable structure found involved 2nd neighbor silicon atoms.

Figure 1:Photo-fragmentation spectrum of C58Si2+ with a ball and stick representation of the most stable structure for the neutral C58Si2

As the mass of 7 carbon atoms equals the mass of 3 silicon atoms, each peak will contain more than one single cluster composition, as C n Sim and C n ±7 q Sim m3 q with q ∈ Z * . So heterofullerenes containing 3 silicon atoms or more are merged in less siliconized series Fortunately we can get ride off the major composition (the richer in carbon) with a low laser intensity photo-fragmentation study, as the different isomass do not have the same stability and absorption cross section. This method allows us to extract the fragmentation pattern of those heterofullerenes (namely through evaporation of Si2 and Si3C) and to get evidences for a cage structure till 12 silicon atoms included in the structure. The knowledge of the geometry of these structure still requires more calculations in order to confirm experimental results. Another way to produce heterofullerenes The study of stœchiometric cluster with same proportion of carbon and silicon, exhibits a surprising feature : a new way to produce heterofullerenes. Assume a structure close to bulk one for stœchiometric clusters with surface reconstruction. Heating those clusters shows the loss of Si2C and Si3C which leads to an impoverishment in silicon during the dissociation sequence. Looking at the fragmentation pattern close enough, one could notice a visible modification with the signature of cage structures, thus the appearance of heterofullerenes. Even if the real dynamic of this phenomenon remains unknown, the structural transition from stœchiometric clusters to heterofullerenes is clear. If we plot the size of the selected cluster versus the mass loss to the structural transition (as shown on figure 2) we can extract information about this transition. Effectively th

2

dependence of these two parameters is clearly linear which indicate a transition for given number of silicon in the structure. The maximum number of silicon it is possible to substitute in a cage structure without destruction is given by the graph and is included between 12 and 16 atoms.

Size n of the selected cluster (SiC) n

90

Expérience Ajustement linéaire Y = 0.031 X + 14.26 Taux de confiance supérieur à 95%

80 70 60 50 40 30 20 10 0

200

400

600

800 1000 1200 1400 1600 1800 2000 2200

Mass loss to structural transition (in a.m.u.) Figure 2 : Mass loss to structural transition versus selected cluster size

Even if the two ways of production are different the maximum doping of the cage seems unchanged. Calculations are in progress to find a possible geometrical reason for that dramatic limitation. So, we have proved the existence of heterofullerenes and shown two original ways to produce these new structures. Low Energy Cluster Beam Deposition (LECBD) of Heterofullerenes After the discovery of these amazing structures, another goal will be to use their exotic properties to create a new family of materials. The properties of these materials obtained by LECBD might be controlled by tuning the composition of the initial target and thus the average doping of the heterofullerenes. The first results are very encouraging and a article is in preparation. A collaboration is also in progress to proceed to macroscopic production of these heterofullerenes through chemical methods.

3