Eighth Condensed Phase and Interfacial Molecular Science (CPIMS)

Integrated O 2 PSD (ML)
Integrated O 2 PSD (ML)
0.1
0.08
0.06
0.04
0.02
0
0.06
0.05
0.04
0.03
0.02
0.01
a)
b)
18 O2
16 O 18 O
16 O2 (x 0.2)
0
0 100 200 300 400 500 600
T
anneal
Figure 1: Integrated O2 PSD yields versus annealing
temperature, Tanneal. 18 O2 was deposited at 30 K
and annealed. Then physisorbed 16 O2 was
deposited and the system was irradiated with
UV light while the desorption of the O2 isotopes
was monitored with a mass spectrometer.
11 th -20 th cycle:
Physisorb 18 O
2
Figure 1 shows the time-integrated 16 O2,
18 O2, and 16 O 18 O photon-stimulated desorption
(PSD) yields versus annealing temperature, Tanneal,
of chemisorbed oxygen. 18 O2 was adsorbed at 30 K
and annealed at Tanneal for 240 s. Then 0.4 ML 16 O2
was adsorbed and the sample was irradiated (both
at 30 K). The 16 O 18 O PSD signal results from
photon-stimulated reactions involving the exchange
of atoms between the chemisorbed oxygen and
physisorbed O2. The 18 O2 PSD yield decreases approximately monotonically as Tanneal increases and is
essentially zero above 400 K (Fig. 1a, red triangles). The 16 O2 PSD yield is nearly independent of Tanneal
(Fig. 1a, blue squares). In contrast, the 16 O 18 O PSD yield initially increases with increasing annealing
temperature, goes through a maximum at Tanneal ~ 375 K, and then rapidly drops to zero for high
temperatures (Fig. 1b, black circles). Annealing above ~400 K brings Ti interstitials to the surface that
react with the adsorbed oxygen to form TiOx islands, resulting in the observed decrease in the 16 O2 and
16 O 18 O PSD yields. The results in Figure 1 suggest that annealing TiO2(110) with adsorbed oxygen leads
to the creation of a distinct, stable species on the surface, and changes in the angular distribution of the
desorbing oxygen versus Tanneal support this hypothesis (data not shown).
Remarkably, the chemisorbed oxygen species responsible for the photon-stimulated exchange
reactions is not destroyed by repeated cycles of UV irradiation. Figure 2 shows the 16 O 18 O PSD signals
versus time for an experiment with repeated UV irradiations. For this experiment, 18 O2 was annealed at
350 K. The sample was subsequently dosed with physisorbed 16 O2 at 30 K and irradiated with UV light.
The black trace in the upper left corner of Figure 2 (labeled “1 st ”) shows the 16 O 18 O PSD versus time for
the first irradiation. This was followed by nine more cycles where the sample was heated to 100 K (to
desorb any remaining physisorbed O2), dosed with 16 O2 and UV irradiated. The red and blue traces on the
left side of Figure 2 show the 16 O 18 O PSD signals for the third and tenth irradiation cycles, respectively.
By the tenth irradiation cycle, the 16 O 18 O PSD is essentially zero. Following the first ten UV irradiation
116
16 O 18 O PSD Signal (arb. units)
1 st -10 th cycle:
Physisorb 16 a)
O
2
1 st
3 rd
10 th
11 th
13 th
20 th
Integrated 16 O 18 O PSD (ML)
0 50 100 150 0 50 100 150
time (s)
Figure 2: 16 O 18 O PSD signals versus time for
repeated UV irradiations. See text for details.