A novel method for the synthesis of core-shell nanoparticles for functional applications based on long-term confinement in a radio frequency plasma

Abstract

A novel combined setup of a Haberland type gas aggregation source and a secondary radio frequency discharge is used to generate, confine, and coat nanoparticles over much longer time scales than traditional in-flight treatment. The process is precisely monitored using localized surface plasmon resonance and Fourier-transform infrared spectroscopy as in situ diagnostics. They indicate that both untreated and treated particles can be confined for extended time periods (at least one hour) with minimal losses. During the entire confinement time, the particle sizes do not show considerable alterations, enabling multiple well-defined modifications of the seed nanoparticles in this synthesis approach. The approach is demonstrated by generating Ag@SiO₂ nanoparticles with a well-defined surface coating. The in situ diagnostics provide insights into the growth kinetics of the applied coating and are linked to the coating properties by using ex situ transmission electron microscopy and energy dispersive X-ray spectroscopy. Surface coating is shown to occur in two phases: first, singular seeds appear on the particle surface which then grow to cover the entire particle surface over 3 to 5 minutes. Afterwards, deposition occurs via surface growth which coincides with lower deposition rates. Our setup offers full control for various treatment options, which is demonstrated by coating the nanoparticles with a SiO₂ layer followed by the etching of the part of the applied coating using hydrogen. Thus, complex multi-step nanofabrication, e.g., using different monomers, as well as very large coating thicknesses is possible.

Fig. 1. (a) Side view of the experimental setup combining a gas aggregation source (GAS) with a secondary treatment plasma setup for long-term confinement. Particles are generated at the GAS at a higher pressure, creating a flow directed through the feed pipe to the differential pumping stage. Here, the buffer gas is pumped away symmetrically to minimize the loss of particles. The particles then flow through a second feed pipe through a shutter system directly into the treatment plasma volume, where charging effects ensure their confinement. While a coil electrode is used for the option of an inductively coupled plasma, in this work it is only used in a capacitively coupled mode. The plasma volume is surrounded by a grounded cage for better confinement through higher electric fields. The hatched area in red represents the IR/UV-vis beam area oriented perpendicular to the sketch plane. (b) Top view of the treatment plasma chamber. The IR/UV-vis beam path is shown in red. Argon is let in through the marked nozzles to create a protective gas flow for the white cell mirrors.

Fig. 2. (a) In situ UV-vis transmission spectra for 7 different times during the NP injection. 0 s corresponds to the plasma ignition in the GAS. At 60 s the injection of Ag NPs is stopped. Plasma emission lines were removed from the spectra. (b) Long-term measurement of injected Ag NPs, which are trapped for 1 h in the CCP plasma. The peak position λ_peak corresponds to the wavelength of the minimum in the transmission spectrum.

Fig. 3. (a) UV-vis absorption spectrum showing a LSPR peak shift. The starting time t = 0 s corresponds to the start of the coating treatment. Plasma emission lines were removed from the spectrum. (b) Time evolution of the LSPR peak position for four different treatment durations. (c) FTIR absorption spectrum with the inset showing a decomposition of the Si–O–Si asymmetric stretching absorption and the sum of fits. (d) Time evolution of the integrated absorbance and calculated Si–O–Si concentration in the plasma volume for four different treatment durations.

Fig. 4. Total concentration of Si determined through fitting the four Si–O–Si stretching modes in the IR spectrum. The growth rate changes after about 150 seconds from 0.23 μmol L⁻¹ s to 0.17 μmol L⁻¹ s and is indicated with dashed lines.

Fig. 5. Prolonged argon plasma treatment of SiO₂-coated Ag NPs. The particles have been generated in the GAS and coated using silane injection for 90 s (marked in grey) followed by an hour-long confinement. Gas pressure: 11 Pa, CCP power: 100 W, and argon flow: 30 sccm.

Fig. 6. Deposition and etching experiment with SiO₂ coated Ag NPs. Ag NPs were injected into the treatment chamber, coated using SiH₄ and subsequently etched using H₂ (both marked in grey). This procedure was done once with UV-vis measuring the LSPR peak position (right axis) and once with FTIR monitoring measuring the Si–O–Si concentration (left axis). The increase and decrease in coating can be clearly seen in both methods.

Fig. 7. TEM micrographs for different coating times. It is visible that the NPs are not homogeneously coated for 1 to 5 min. After 10 min the coating looks homogeneous (a–d). The growth of the coating on the Ag NPs is schematically shown (e–h). Additionally, an EDX map is shown for a sample with a coating time of 10 min, which shows that the particles have an Ag core and a SiO_x shell (i–l).

Fig. 8. Atomic concentration of Si measured with EDX during TEM measurements for extracted Ag@SiO₂ NPs with different coating times. Ten different areas of the TEM grid were used for statistical analysis. Only silver and silicon were taken into account to minimize the influence of oxidation after extraction and the background carbon support on the TEM grid.