S C

IE N

T IF

IC H

IG H

LI G

H T

S C

O M

P L

E X

S Y

S T

E M

S A

N D

B IO

M E

D IC

A L

S C

IE N

C E

S

5 9 I H I G H L I G H T S 2 0 2 2

Finding the best efficiency for laser- machining of gold colloids

A detailed understanding of the heating efficiency and property change of colloidal gold nanoparticles during the fragmentation process was missing. Using X-ray scattering, the local excitation efficiency of gold nanoparticles was examined on ultrashort timescales. Although the absorption efficiency is higher, the heating efficiency at the plasmon resonance of gold nanoparticles is significantly lower compared to the intraband excitation.

The synthesis of quantum-sized metal particles without any organic impurities can be achieved by laser ablation and laser fragmentation of matter in an inorganic, liquid environment [1]. However, in order to enable independent and flexible material design, which is required for the different demands of technological applications such as catalysis, the mechanisms and processes that occur during laser excitation, heating and structure transformations of nanoparticles need to be understood [2]. In this work, the excitation (heating) efficiency of gold colloids by laser pulses was studied with picosecond time resolution, and was compared for 532 nm under plasmon excitation and 400 nm at interband excitation with 1 ps laser pulses.

Plasmon excitation generally shows a higher ground state extinction and absorption (as seen in Figure 48), which is beneficial for resonant applications, where a very local strong electric field can be harnessed. The inset in Figure 48 additionally depicts an electron micrograph of the used nanoparticles with an impression of the near- field that is excited by the incoming polarised laser.

Fig. 48: Spectral extinction of a gold colloid of 53 nm particles in water from measurement (crosses) and calculation (line), together with the optical excitation wavelengths (arrows).

The inset depicts some gold nanospheres together with the excited near-field distribution when illuminated by a linear

polarised light source.

It was found that the heating of the nanoparticles within several hundred picoseconds after laser excitation, despite the higher absorption efficiency at 532 nm (optical plasmon excitation), is considerably lower compared to an interband excitation at 400 nm. Hereby, the particle heating and temperature change, as well as the subsequent structure and size change of the laser- excited nanoparticles, was measured with the help of wide-and small-angle X-ray scattering, respectively, using the laser-pump-X-ray-probe setup at beamline ID09. At this setup, the transient state of the lattice expansion of colloidal gold nanoparticles can be caught with a 50 ps time resolution and be calorically (e.g., by the heating efficiency) compared to the incoming laser intensity.

Figure 49 shows the measured lattice expansion (interpreted as nanoparticle heating) at the two wavelengths and a pump-probe delay of 55 ps, together with a theoretical prediction including the gold expansion coefficient and latent heat. The lattice expansion is expected to stop at about 1.8 %, where the melting point of gold is reached, and the powder scattering should vanish. Indeed, the increase of the expansion follows the prediction with an onset of melting around 40 80 J/m2. Yet, the expansion for 532 nm (green line in Figure 49) shows a much smaller slope than predicted by the absorption coefficient, indicating a lower heating efficiency despite the higher absorptivity.

The reason for this behaviour is currently hypothesised to originate from an ultrafast bleaching of the plasmon resonance due to non-equilibrium excitation, as well as yet-to-be-validated additional dissipation channels

Fig. 49: Transient lattice expansion of the colloid as a function of applied laser fluence for 400 nm (blue symbols) and 532 nm (green symbols) laser pulses. The lines show the predicted expansion (dashed line for 532 nm) and the

fitted expansion for both wavelengths.