Electronic excitations induced by the impact of coinage metal ions and clusters on a rare gas matrix: Neutralization and luminescence

Low-energy collisions of atomic ions and cluster ions with rare gas covered metal substrates can lead to strong light emission which is element and size specific. Instead of a direct energy transfer into the electronic system of the cluster the observed emission originates from excitons trapped in the rare gas layer which carries the excitation energy. The nature of the processes involved in the neutralization and excitation of the neutral clusters electronic system are discussed. Strong analogies to thermoluminescence are found.

Figure 1

Schematics of the deposition region. Inset: Light collection with an optical fiber placed in the vicinity of the deposition spot.

Figure 2

Evolution of the ${\mathrm{Ag}}_2^{+}$ current as a function of matrix thickness measured at a temperature of $20\mathrm{K}$. Since Ar is dosed at a constant rate of $0.3\mathrm{nm}∕\mathrm{s}$ the maximum thickness corresponds to a time of $1200\mathrm{s}$.

Figure 3

Current reading versus the potential difference $\Delta V$ between the sample holder and the last lens element before the sample holder (the distance between the two planes is $\sim 5\mathrm{mm}$). Note that the fluctuations in the current increase dramatically even for very small potential difference. See text for details.

Figure 4

Spectral signature of the argon discharges. See text for details.

Figure 5

(Color online) This diagram sketches the energetics of the studied system in the absence of external electric fields. The system is considered as ideal, i.e., without structural defects or impurities. See text for details.

Figure 6

(Color online) Evolution of light emitted by ${\mathrm{Ag}}_2^{+}$ collisions measured at a matrix temperature of $20\mathrm{K}$. (a) The deposition starts with no argon matrix on the metal surface. The clusters are deposited together with argon to form a seeded matrix. No light emission is observed at zero coverage, the luminescence starts with a matrix thickness of $\sim 20\mathrm{nm}$ and saturates after $\sim 200\mathrm{nm}$. The deposited current reading is identical to the one of Fig. 2 and the photon counts have been corrected for current fluctuations. (b) Simulated intensity of the emission at $470\mathrm{nm}$ parallel to a gold surface (see text). (c) The matrix is first built without clusters until a thickness of $700\mathrm{nm}$ and then the clusters are directed towards the matrix together with argon to form a seeded matrix. In this case the emitted light is directly proportional to the cluster current. For technical reasons, the cluster current is slowly increased over time.

Figure 7

(Color online) (a), (b) Spectrally resolved luminescence from ${\mathrm{Ag}}_1^{+}$ and ${\mathrm{Au}}_1^{+}$ collisions and laser induced fluorescence spectra of ${\mathrm{Ag}}_1$ and ${\mathrm{Au}}_1$ in argon for comparison. (c) Luminescence spectrum from ${\mathrm{Cu}}^{+}$ collisions.

Figure 8

(Color online) (a) Spectrally resolved luminescence of ${\mathrm{Ag}}_1^{+}$, ${\mathrm{Ag}}_2^{+}$, and ${\mathrm{Ag}}_3^{+}$ during the deposition in an argon matrix. (b) Corresponding luminescence of ${\mathrm{Au}}_1^{+}$ and ${\mathrm{Au}}_2^{+}$.

Figure 9

(Color online) Spectrally resolved luminescence of ${\mathrm{Ag}}_1^{+}$, during the deposition in an argon and a krypton matrix.

Figure 10

Spectrally resolved luminescence of ${\mathrm{Au}}_2^{+}$, during the deposition into an argon matrix. The comparison with Fig. 8 shows that when the impact energy is raised the gold dimer fragments. The fact that we observe the luminescence of the fragment means that the luminescence process happens after the fragmentation has taken place. The spectra are normalized in order to keep the area below them constant.

Figure 11

Ratio between the intensity of the main peak associated with the luminescence of ${\mathrm{Au}}_2$ $\left(745\mathrm{nm}\right)$ and the main peak associated with the luminescence of ${\mathrm{Au}}_1$ $\left(465\mathrm{nm}\right)$ for different deposition energies. The line is a guide to the eye. The total current varies with the deposition energy. It is, however, impossible to normalize the signal with the deposition current, because of the electron avalanche process discussed above and illustrated in Fig. 3.

Figure 12

(Color online) Luminescence intensity of ${\mathrm{Ag}}_2^{+}$ $(E_{\mathrm{dep}}=30\mathrm{eV})$ versus time. The continuous line shows the luminescence and the dotted line the measured current. Three experiments are shown: a fresh matrix is grown at $18\mathrm{K}$ with no cluster current for the first $30\mathrm{s}$ [(a) and (b)] and for $60\mathrm{s}$ (c). After the “preparation” the cluster beam is switched on at $t=0$.

Figure 13

Luminescence intensity versus the matrix temperature during the collision of ${\mathrm{Ag}}_2^{+}$. At $34\mathrm{K}$ the evaporation of the matrix sets in. The experiment is conducted under steady argon gas deposition and with a deposition energy for ${\mathrm{Ag}}_2^{+}$ of $30\mathrm{eV}$. The line is a guide to the eye.

Figure 14

(Color online) Matrix annealing experiments. Here again the continuous line shows the luminescence and the dotted line the measured current. A fresh matrix is first deposited without clusters at $18\mathrm{K}$ for $60\mathrm{s}$, it is then annealed by heating it to $32\mathrm{K}$ for $\sim 30\mathrm{s}$ and cooled down again to $18\mathrm{K}$; when the cluster current is switched on almost no luminescence is seen. Argon is then let into the system; this provides a reference for the luminescence intensity at $18\mathrm{K}$. Later the same experiment is renewed but this time with a short annealing cycle (heat and cool) the luminescence intensity is smaller due to incomplete annealing of the matrix.

Figure 15

Free excitons (FE) are created in the collision process. Due to the strong interaction with the matrix deposited at low temperature most of FE become self-trapped excitons (STE). When the STE recombine a high energy photon is emitted that has a 50% probability to be absorbed by the close-by metal substrate and produce a photoelectron responsible for the neutralization. When the matrix is grown at low temperature the trapping of the electron on defects results in two stable entities the self-trapped hole (STH) and the trapped electron $\left({\mathrm{Te}}^{-}\right)$. As long as the electron is trapped the excitation is stable. This is the initial situation in the thermally stimulated experiments (TSL, TSC, TSEE, see text). In our experiment the electron is liberated by the local annealing resulting from the collision process. Due to the close proximity with the cluster an energy exchange mechanism between the STE and the atom or cluster can take place and result in the observed fluorescence.