^{1}, C. Dahmen

^{2}, G. von Plessen

^{2}, F. Springer

^{3}and A. Plech

^{3,a)}

### Abstract

Intense nonequilibrium femtosecond laser excitation of goldnanoparticles in water leads to a transient heating of the nanoparticles, which decays via heat transfer to the water phase. It is shown that the water temperature rises to near the critical temperature and the water undergoes an explosive evaporation in the subnanosecond range. The formation of vapor bubbles shows a threshold dependence on laser fluence. The nascent nanoscale vapor bubbles change the heat dissipation drastically. The nanoscale structure is resolved directly with a combination of x-ray scattering methods sensitive to the particle lattice expansion and the change in the water structure factor.

Help from M. Wulff, Q. Kong, and M. Lorenc at the beamline ID09B and discussions with F. Lang and J. Boneberg are acknowledged. This project is funded by the ESRF, DFG (Sonderforschungsbereich 513 and Graduiertenkolleg “Biointerface”) and the Center for Junior Research Fellows Konstanz.

I. INTRODUCTION

II. METHODS

A. X-ray scattering setup

B. Laser excitation

C. Thermal calculations

III. RESULTS AND DISCUSSION

A. Cooling kinetics

B. Bubble formation

C. Discussion

IV. CONCLUSION

### Key Topics

- Gold
- 22.0
- X-ray scattering
- 22.0
- Water heating
- 20.0
- Nanoparticles
- 18.0
- Heat transfer
- 17.0

## Figures

(Color online) Extinction spectra for several gold nanoparticle solutions rescaled to the cuvette diameter of . The plasmon resonance shifts to larger wavelengths for increasing particle sizes.

(Color online) Extinction spectra for several gold nanoparticle solutions rescaled to the cuvette diameter of . The plasmon resonance shifts to larger wavelengths for increasing particle sizes.

(a) Thermal expansion coefficient and (b) molar heat capacity of bulk gold as functions of temperature. (c) Resulting specific expansion coefficient, as the ratio of the two material constants.

(a) Thermal expansion coefficient and (b) molar heat capacity of bulk gold as functions of temperature. (c) Resulting specific expansion coefficient, as the ratio of the two material constants.

(Color online) Lattice expansion of gold particles of 52 and diameters as determined from the peak shift of the (111) powder reflection at (circles) and (crosses) delays together with a calculation of the thermal expansion (lines; dashed line without rescaling, see text). Above no powder reflection at delay is detectable, indicating particle melting.

(Color online) Lattice expansion of gold particles of 52 and diameters as determined from the peak shift of the (111) powder reflection at (circles) and (crosses) delays together with a calculation of the thermal expansion (lines; dashed line without rescaling, see text). Above no powder reflection at delay is detectable, indicating particle melting.

(Color online) Temperature evolution of (엯) and (●, Ref. 14) particles together with the calculated temperatures of the particle lattice (upper solid curves) and the adjacent water shell (lower solid curves). The calculated particle lattice temperatures are rescaled so that the calculated temperature rise at matches the experimental particle temperature rise . For better comparison, the dashed lines represent a convolution of the particle temperature curves with the resolution function of the x-ray pulse.

(Color online) Temperature evolution of (엯) and (●, Ref. 14) particles together with the calculated temperatures of the particle lattice (upper solid curves) and the adjacent water shell (lower solid curves). The calculated particle lattice temperatures are rescaled so that the calculated temperature rise at matches the experimental particle temperature rise . For better comparison, the dashed lines represent a convolution of the particle temperature curves with the resolution function of the x-ray pulse.

Calculated temperature profiles around a particle of diameter as function of the radial coordinate from the particle center for delays of 100 and . The temperature within the particle is assumed to be uniform. At the short delay of a discontinuous drop of the temperature across the particle-water interface is a result of the thermal boundary resistance. The temperature before excitation and the water critical temperature are marked by the dashed lines.

Calculated temperature profiles around a particle of diameter as function of the radial coordinate from the particle center for delays of 100 and . The temperature within the particle is assumed to be uniform. At the short delay of a discontinuous drop of the temperature across the particle-water interface is a result of the thermal boundary resistance. The temperature before excitation and the water critical temperature are marked by the dashed lines.

(Color online) Difference in liquid scattering of a sol with gold particles at delay at a fluence of (+) and (●). The solid line is the static derivative , determined for pure water and scaled in amplitude to match the signal height of . A volume change of the bulk water of is derived from that amplitude, which is equivalent to a bubble diameter of . The reduction of powder scattering at the fcc reflections is indicated by the bars.

(Color online) Difference in liquid scattering of a sol with gold particles at delay at a fluence of (+) and (●). The solid line is the static derivative , determined for pure water and scaled in amplitude to match the signal height of . A volume change of the bulk water of is derived from that amplitude, which is equivalent to a bubble diameter of . The reduction of powder scattering at the fcc reflections is indicated by the bars.

Ratio of bubble volume to particle volume as a function of laser fluence for particle sizes of 39 and at the delay of maximum bubble radius ( for particles, for particles). The dashed vertical lines indicate the threshold.

Ratio of bubble volume to particle volume as a function of laser fluence for particle sizes of 39 and at the delay of maximum bubble radius ( for particles, for particles). The dashed vertical lines indicate the threshold.

Dependence of the system properties on particle size. The absorbed proportion (a) of the laser fluence (as calculated using the Mie theory) decreases with particle size due to increased light scattering. The required temperature rise (b) to the calculated melting point (Ref. 34) is reduced through the finite size effect for small particles. The maximum temperature rise (c) in the water phase is derived as a function of particle size from the calculated spatial temperature profiles at the particle melting point. The vapor nucleation temperature relative to the particle temperature at the melting transition is shown for the different particle sizes under study. The open symbols are derived from Fig. 3, while full symbols are deduced from the diffuse scattering. The lines are based on the bubble nucleation threshold occurs at (solid line) and 85% (dashed line), respectively.

Dependence of the system properties on particle size. The absorbed proportion (a) of the laser fluence (as calculated using the Mie theory) decreases with particle size due to increased light scattering. The required temperature rise (b) to the calculated melting point (Ref. 34) is reduced through the finite size effect for small particles. The maximum temperature rise (c) in the water phase is derived as a function of particle size from the calculated spatial temperature profiles at the particle melting point. The vapor nucleation temperature relative to the particle temperature at the melting transition is shown for the different particle sizes under study. The open symbols are derived from Fig. 3, while full symbols are deduced from the diffuse scattering. The lines are based on the bubble nucleation threshold occurs at (solid line) and 85% (dashed line), respectively.

Article metrics loading...

Full text loading...

Commenting has been disabled for this content