Etude théorique des alternances pair-impair dans les propriétés des amas Cn, C+n et C- n linéaires

Abstract

2014 Nous étudions les amas de carbone linéaires Cn, C+n, C-n (n ~ 7). Nous analysons d'abord la stabilité et l'énergie d'ionisation des Cn. Dans cette dernière étude il apparaît qu'il faut aller au-delà du modèle « à un électron » de Pitzer et Clementi pour rendre compte des résultats expérimentaux (pas d'alternance des énergies d'ionisation avec n). L'introduction des corrélations diélectroniques est faite à l'aide de la méthode de Gutzviller étendue aux petits amas, technique décrite dans d'autres travaux. Nous montrons aussi que, dans le cas de C3, il faut envisager l'ionisation d'un niveau 03C3 et non 03C0 comme pour les autres Cn. En tenant compte de ces nouveaux éléments le calcul des stabilités de Cn, C+n et C-n est effectué ; l'accord avec les résultats expérimentaux, obtenus par sublimation (Cn, n impairs plus stables) par source à étincelle (C+n, n impairs plus abondants) et par émission ionique secondaire (C-n , n pairs plus abondants), est satisfaisant. Pour montrer comment, en émission ionique (et aussi par source à étincelle car les deux techniques donnent en général les mêmes résultats), la stabilité d'une espèce donnée est reliée à son intensité d'émission nous discutons les mécanismes d'émission. Nous montrons en particulier que la charge positive de bout de chaîne qui apparaît sur les Cn avec n impair peut être importante pour favoriser la création d'ion C-n avec n pair. Abstract. 2014 We study the linear carbon aggregates Cn, C+n, C-n (n ~ 7). We first analyse the stabilities and the ionization potentials of Cn. In the latter study it appears that it is necessary to go beyond the « one-electron » model of Pitzer and Clementi to reproduce experimental results (no alternation with the parity of n of the ionization potentials). We introduce the dielectronic correlations by using a Gutzwiller technique adapted to small aggregates. This method is developed in other works. It also appears that, in the case of C3, the ionized level is of 03C3 type, not of 03C0 type as for the other Cn studied. By taking these new factors into account we calculate the stabilities of Cn, C+n and C-n and find a satisfactory agreement with experimental results : in Cn, odd n's are more stable ; in C+n, odd n's are more abundant in spark source discharge ; in C-n, even n's are more abundant in SIMS. To show how in SIMS (but also in spark source discharge because the two techniques give, in general, the same spectra) the stability of a given aggregate is related to its intensity in the mass spectrum, we discuss the emission mechanism. In particular, we show that the positive charge located on the end-atom of the Cn chains when n is odd may be important in the production of C-n clusters with an even number of atoms.

Energetics of carbon clusters C8 and C10 from all-electron quantum Monte Carlo calculations

“…The recovery of correlation energy is 89%-94%. For C 8 , relative values are found to be in agreement with earlier calculations: the cyclic C 4h 1 A g structure, the linear 3 ⌺ g Ϫ ͑cumulenic͒, and the linear 1 ⌺ g ϩ ͑cumulenic͒ structures are found lowest in energy and within 10 kcal/mol of each other. For C 10 , relative values are found to be in agreement with earlier calculations: the cyclic D 5h 1 A 1 Ј ͑distorted cumulenic͒ and D 10h 1 A 1g ͑cumulenic͒ structures are found to be lowest in energy, within 5 kcal/mol of each other, 60 kcal/mol below the linear 3 ⌺ g Ϫ ͑cumulenic͒ structure.…”

“…A partial list of results of these calculations is given in Tables I-IV. The likely stable structures for C 8 and C 10 include the singlet and triplet states of the linear and cyclic structures with either acetylenic ͑-CϵCϪCϵC-͒ or cumulenic ͑vCvCvCvCv͒ bonding-a total of eight possibilities for each. However, as shown in Tables I-IV the lowestenergy structures given by the earlier high-level ab initio calculations are the linear cumulenic 3 ⌺ g Ϫ and the cyclic acetylenic C 4h 1 A g structures for C 8 …”

“…The earliest calculations for C 8 and C 10 were carried out by Pitzer and Clementi, 4 Hoffmann, 5 Slanina and Zahradnik, 6 Ewing and Pfeiffer, 7 Joyes and Leleyter, 8 and Bernholc and Phillips, 9 using either semiempirical methods or self-consistent field ͑SCF͒ calculations with small basis sets.…”

Determination of some C60 electronic properties. Application to the cohesive energy in C60 ionic compounds

“…Thus, by bombarding a graphite sample interesting odd-even alternations were observed in the C+n and C; spectra, n 15 [5]. These effects were interpreted as due to larger stabilities of the corresponding aggregates and various theoretical descriptions confirmed these conclusions [6]. In particular, it was possible to show that, in this range of sizes, the shape was linear.…”

Optimum structures and vibrational frequencies of (SIC)2 clusters

“…Gas-phase neutral and ionic clusters of carbon and of silicon have been the subject of intense experimental [1][2][3][4][5][6][7][8][9][10][11][12][13] and theoretical [14][15][16][17][18][19][20][21][22][23] scrutiny. The most recent theoretical studies of carbon clusters [20] predict that such clusters exist as either linear or cyclic planar structures, for C, to C,o.…”