O2 Oxidation and Sublimation Kinetics of Single Silicon Nanoparticles at 1200 to 2050 K: Variation of Reaction Rates, Evolution of Structural and Optical Properties, and the Active-to-Passive Transition

Abstract

Sublimation and O2 etching kinetics for a series of individual silicon (Si) nanoparticles (NPs) were studied for NP temperatures (TNP) from 1200 to 2050 K, using a single NP mass spectrometry technique. Sublimation was significant for TNP > 1700 K, with rates reasonably well fit to Arrhenius kinetics, but evolving, particularly during initial heating. O2 etching efficiencies varied from NP-to-NP and with changing TNP, but also evolved dramatically over time. For TNP ≤ 1500 K, NPs were observed to passivate after losing 30 to 50% of the initial NP mass. At higher TNP, etching efficiency decreased over time, but never passivated. Interestingly, bulk Si passivation has not been observed for the range of TNP and O2 pressures used here, and a model was developed to test the effects of several NP-specific mechanistic parameters on both the initial and time-dependent etching behavior. The optical properties of the hot NPs were also found to evolve as the NPs etched, particularly during the initial fast mass loss, and correlations between emission intensities and etching kinetics were examined.

High Temperature Transformation, O2 Etching, and Passivation of Single SiOx Nanoparticles: Kinetics and Optical Properties as Structure Probes

“…3. Pre-heated silica NPs held in O2-containing atmosphere etched vigorously at 1200 and 1300 K, with many similarities to the etching behavior of Si NPs at 1200 K. 13 4. Active etching continued at roughly constant rates as the NPs lost >50% of Minitial 5.…”

“…Elemental silicon absorbs strongly in the visible, and as expected, Si NPs readily absorb 532 nm laser radiation, resulting in strong blackbody-like emission in the visible and near-IR. 13 In contrast, bulk silica is transparent in the visible, 20 and we find that silica NPs, as they are initially trapped, cannot be heated to Just before time = 0 in Fig. 1A, NP injection was stopped and the CO2 laser was turned on for a few seconds at 1% duty factor, irradiating any NPs in the trap with ~0.1 W average power, focused to a ~1 mm beam waist (intensity ≈ 10 W/cm 2 ).…”

“…3 shows a variable TNP oxidation experiment for a silica NP, done using a similar protocol to analogous experiments done on Si NPs. 13 The NP was CO2 laser pre-heated briefly as it was trapped, and then Q-stepped prior to the start of the data record shown. Minitial was 37.4 MDa Fig.…”

“…The right axis reports the same data as etching efficiencies (EEO2), in units of Daltons of mass lost per O2 collision with the NP surface, allowing comparison with experiments at different PO2, The same EEO2 units were used in our previous studies of Si NP oxidation 13 and carbon oxidation [11][12] allowing the reactivity of different NPs under different conditions to be compared.…”

“…For Si NPs, similar sublimation and oxidative etching kinetic experiments were reported. 13 Si also etched and sublimed, but the kinetics depended strongly upon both TNP and oxygen pressure (PO2). At low TNP, sublimation was slow, and oxidation led to substantial (up to ~70%) mass loss, with eventual passivation, where EEO2 dropped several orders of magnitude.…”