Dynamics of O2 photodesorption from metal clusters: A significant difference from bulk behaviour

Abstract

Photodesorption of O 2 from size-selected Ag n O À 2 cluster anions with n = 2, 3, 4 and 8 was studied using time-resolved photoelectron spectroscopy (TR-PES). The spectra indicate that relaxations of photo-excited Ag n O À 2 clusters with n = even numbers accompany ultrafast direct O 2 photodesorption. For the odd-numbered cluster Ag 3 O À 2 , in contrast, a long-lived excited state is observed, since O 2 might be dissociatively chemisorbed, suppressing direct photodesorption of oxygen. Both, direct desorption and long-lived excited states, have not been observed from adsorbate covered metal surfaces, suggesting unique photochemical properties of such small clusters.

O2 photodesorption from AuO 2 − and Au2O 2 −

“…For Au 2 O − 2 , O 2 desorption is observed. The data can be compared to the ones of the silver species: Ag 3 O − 2 [10] and AuO − 2 behave similarly, and in both cases the two O atoms are probably bound dissociatively [13,14]. The spectra of Au 2 O − 2 , although with a poor signal to noise ratio, exhibit certain similarities to the ones of Ag 2 O − 2 [9], where our TRPES data indicate a direct desorption process.…”

“…The experimental setup has been described in detail elsewhere [9,10]. Gold cluster anions are generated in a pulsed arc cluster ion source (PACIS) and O 2 is introduced into the source to generate Au n O − m clusters.…”

“…Recently, we started a systematic search for long-living excited states of metal clusters reacted with O 2 [9,10] using time-resolved photoelectron spectroscopy (TRPES). We began with Ag and Au clusters reacted with oxygen, because oxides might have a more rigid geometric structure suppressing thermalization by internal conversion [7].…”

“…In addition, the metal cores have a low density of states at energies close to the highest occupied molecular orbital even for larger clusters such as Au 20 [11]. In the case of Ag [10]. It was not possible to determine the decay channel of this long-living state, but oxygen desorption and fragmentation could be ruled out.…”

“…It was not possible to determine the decay channel of this long-living state, but oxygen desorption and fragmentation could be ruled out. Fast O 2 photodesorption was the main decay channel for even-numbered Ag n O − 2 clusters with n = 2, 4 and 8 [9,10]. Unfortunately, in these earlier studies we lacked the higher probe photon energy necessary to expand these studies to Au n O − 2 clusters.…”