Femtosecond transient absorption dynamics of close-packed gold nanocrystal monolayer arrays

https://doi.org/10.1016/j.cplett.2004.01.056Get rights and content

Abstract

Femtosecond transient absorption spectroscopy is used to investigate hot electron dynamics of close-packed 6 nm gold nanocrystal monolayers. Morphology changes of the monolayer caused by the laser pump pulse are monitored by transmission electron microscopy. At low pump power, the monolayer maintains its structural integrity. Hot electrons induced by the pump pulse decay through electron–phonon (e–ph) coupling inside the nanocrystals with a decay constant that is similar to the value for bulk films. At high pump power, irreversible particle aggregation and sintering occur in the nanocrystal monolayer, which cause damping and peak shifting of the transient bleach signal.

Introduction

Hot electron relaxation dynamics in nanoparticles has been an active research area because it can reveal the change of electronic and lattice vibrational properties when the size of materials approaches the nanometer scale [1], [2], [3], [4], [5], [6], [7], [8], [9]. Time-resolved femtosecond spectroscopy is the primary technique used to investigate these fast relaxation dynamics. The principle of this technique is to excite electrons to higher energy levels through a short pump laser pulse. The excited nonthermalized electron distribution quickly relaxes on a femtosecond time scale into a Fermi–Dirac distribution with a higher electron temperature through electron–electron (e–e) interaction. The thermalized hot electron distribution then cools down through electron–phonon (e–ph) coupling, and eventually dissipates the excess energy through phonon–phonon (ph–ph) coupling, either within the nanoparticle or into the surrounding environment. The relaxation process can be monitored by measuring the transient absorption spectrum though a probe laser pulse that is variably delayed from the pump pulse. The change of electronic distribution will cause a change of the dielectric constant, which affects the transient absorption of the probe pulse. Most of the pump–probe experiments carried out so far use either diluted metal nanoparticle colloids, or nanoparticles embedded in a dielectric matrix [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12]. Aside from the potential problem of large particle size distribution in these experiments, the optical density of these samples is typically quite low, and varies from one experiment to another. These issues have complicated the elucidation of the size range the e–ph coupling constant would deviate from its bulk value. Because of the finite particle size, the transient absorption decay caused by e–ph coupling can be entangled with ph–ph coupling, especially with high energy pump pulses. The increase of the e–ph coupling constant with pump power [10] could depend on the details of ph–ph coupling in the nanocrystals as well as the coupling to their environment [11]. Therefore, the true e–ph coupling constant, which reveals the change of intrinsic electronic and vibrational states due to finite size, can only be probed at low pump power when the contribution from ph–ph coupling is relatively weak. The low optical density in various samples creates a challenge to obtain high signal-to-noise ratios for measurements using low pump power. Furthermore, the effect of the pump pulse on the sample structure, and how that would affect the transient dynamics is not well understood, because effort has been made to study the change of sample morphology after the pump–probe experiments [13], [14]. In this work, we study transient absorption spectra and kinetics of a self-assembled close-packed nanocrystal monolayer consisting of monodispersed nanocrystals. The highly uniform particle size reduces the effect of inhomogeneous broadening. The close-packed nature of the monolayer increases the optical density significantly that, in turn, allows low pump power measurements to be carried out on these samples with high signal-to-noise ratio. We also monitored the sample morphology after the pump–probe experiments by transmission electron microscopy (TEM), which allows us to study the effect of the laser pulse on the sample structure.

Section snippets

Experimental

Gold nanocrystals were synthesized according to a procedure previously developed by one of us [15]. Our synthesis is a single-phase reaction in an organic solvent containing cationic surfactant. The as-prepared nanocrystals were then coated with dodecanethiol through ligand exchange. We have also shown that by adopting a digestive ripening process in an environment of excess thiol, the particle size distribution can be greatly narrowed. Further size segregation caused by lowering the colloid

Results and discussion

A typical transient absorption spectrum of the monolayer at its maximum signal is shown in Fig. 1, with 100 nJ pump pulses. There is an increase of absorption in both the short wavelength region (450–480 nm) as well as the long wavelength region (>600 nm). Also, is a negative transient absorption (bleaching) at the same wavelength as the ground state plasmon absorption peak is observed. These transient features reflect the change of electron distribution induced by the femtosecond pump pulse,

Conclusion

We have performed the first femtosecond pump–probe experiments on gold nanocrystal monolayers that are self-assembled on a silicon nitride membrane widow. TEM through the membrane window allows us to correlate the change of the pump–probe transient absorption signal with the change of sample morphology. We found that, for low laser pump power (∼100 nJ pulse), the nanocrystal monolayer maintains its structural integrity, and the e–ph coupling constant for 6 nm diameter particles is close to that

Acknowledgements

This work is supported by the US Department of Energy, Basic Energy Sciences-Materials Sciences, under Contract # W-31-109-ENG-38, National Science Foundation MRSEC Program under Award Number DMR-0213745, DOE Center for Nanoscale Materials, and by the University of Chicago – Argonne National Laboratory Consortium for Nanoscience Research (CNR).

References (32)

  • J. Hodak et al.

    Chem. Phys. Lett.

    (1998)
  • T.L. Morkved et al.

    Polymer

    (1998)
  • T.S. Ahmadi et al.

    J. Phys. Chem.

    (1996)
  • S.L. Logunov et al.

    J. Phys. Chem. B

    (1997)
  • J.Z. Zhang

    Acc. Chem. Res.

    (1997)
  • J.H. Hodak et al.

    J. Phys. Chem. B

    (1998)
  • J.Y. Bigot et al.

    Phys. Rev. Lett.

    (1995)
  • N.D. Fatti et al.

    Chem. Phys.

    (2000)
  • M. Perner et al.

    Phys. Rev. Lett.

    (1997)
  • J. Hranisavljevic et al.

    J. Am. Chem. Soc.

    (2002)
  • J.H. Hodak et al.

    J. Chem. Phys.

    (1999)
  • S. Link et al.

    J. Phys. Chem. B

    (2002)
  • M.J. Feldstein et al.

    J. Am. Chem. Soc.

    (1997)
  • V.P. Safonov et al.

    Phy. Rev. Lett.

    (1998)
  • G.V. Hartland et al.
  • X.M. Lin et al.

    J. Nanoparticle Res.

    (2000)
  • Cited by (24)

    View all citing articles on Scopus

    Work supported by the US Department of Energy, BES-Materials Sciences, under Contract W-31-109-ENG-38, National Science Foundation MRSEC Program under Award Number DMR-0213745, DOE Center for Nanoscale Materials, and by the University of Chicago – Argonne National Laboratory Consortium for Nanoscience Research (CNR).

    View full text