Aluminum nanoparticle arrays push the high-energy limits of plasmonics

When an electromagnetic (EM) field interacts with a metallic nanoparticle (NP), it can promote a resonantly oscillating displacement of the metal free electrons over the NP, that goes under the name of Localized Surface Plasmon Resonance (LSPR). In correspondence of the LSPR, the EM field in the near vicinity of the NP surface is resonantly enhanced, thus strongly altering the optical response of the nanostructured material and creating “hot” EM field spots that are most useful, e.g. for ultrasensitive “plasmonic” molecular detectors. LSPR are a very general feature of metals, yet most of the research in plasmonics has so far focused on the noble metals Ag and Au, due to their low dielectric losses and their unparalleled low reactivity. Exploiting other metals in plasmonics would however have a strong impact on the field, integrating chemical, catalytic or magnetic functionalities, to name a few, with plasmonics.
Aluminum is considered of the most promising materials for the future of plasmonics, due to its low cost, and its predicted possibility of exhibiting a LSPR in the deep-ultraviolet (DUV) region of the EM spectrum, precluded to most other materials. Successfully exploiting aluminum in plasmonics is however more challenging compared to the noble metals. Aluminum is reactive, hard to synthesize in small-particle form, and quickly oxidizes when exposed to atmosphere, thus degrading its plasmonic performances and redshifting its LSPR. Altogether, these technical issues have stood in the way of the full development of Al-based plasmonics, preventing in particular the achievement of the theoretically-predicted DUV limit of the LSPR.
In our work, we experimentally tested the limits of DUV plasmonics in ultradense arrays of Al/Al2O3 core-shell NPs. Taking advantage of the capabilities of the BEAR beamline to measure the optical response of materials in the UV-DUV spectral range, we addressed the DUV optical response of the Al NP arrays, and found the LSPR in the Al NPs at energies of 5.8 eV, by far the highest value reported so far for Al nanostructures.


Figure 1:   Left: Schematic representation of the Al nanoparticle array fabrication procedure. Middle: AFM image of the Al/LiF system obtained by dewetting of an ultrathin Al film (image size: 800x800 nm2). Right: XPS spectra of the Al 2p core level.

The Al NPs were realized by a self-organization approach (Fig.1, left). The systems were fabricated on spontaneously-nanopatterned LiF(110) surfaces, that exhibit a regular ridge-valley morphology with nanometric periodicity. Grazing-incidence metal evaporation leads to the formation of arrays of nanowires, that are thermally dewetted and exposed to atmosphere to form arrays of disconnected core/shell Al/Al2O3 NPs.
In Fig.1, middle, we report a representative AFM image of an Al NP array obtained by dewetting a ultrathin Al film (image size 800x800 nm2). The small agglomerates arranged in “chains” aligned along the nanopatterned LiF surface are Al NPS with in-plane diameter below 20 nm.
High-resolution X-ray photoelectron spectroscopy (XPS) spectra, in the energy region of the Al 2p core level (Fig.1, right) reveal the presence of a metallic-Al core enclosed by an Al oxide shell. Thus, the small Al NPs retain a metallic core even after exposure to atmosphere, making it possible to observe the LSPR excitation.
In Fig. 2A, we show the optical extinction of the Al NP arrays as a function of photon energy, measured at the BEAR beamline in normal-incidence, with the electric-field vector aligned either parallel (longitudinal) or perpendicular (transverse) to the Al-NP “chains” (Fig. 2, top). In both configurations, a clear extinction peak indicates the successful detection of the LSPR. In the transverse configuration, the LSPR is found at the strikingly-high energy of 5.8 eV, the highest ever observed in optically-excited metallic NPs. In the longitudinal case, the plasmon hybridization along the NP chains redshifts the LSPR to a slightly lower value.
Theoretical calculations by the Finite-Integration Technique (Fig. 2B) reproduce the high-energy character of the LSPR excitation, along with the redshift of the longitudinal LSPR induced by the plasmon hybridization along the NP chains.
Observing a DUV LSPR in aluminum NPs represents a fundamental step forward for ushering plasmonics into the ultraviolet regime. Furthermore, the ease of fabrication of Al particles in the small-size regime and in ultradense arrays is extremely promising for application of these systems as high-efficiency plasmonic substrates for DUV plasmon-enhanced optical spectroscopies.

Figure 2:   Panel A: optical extinction of the Al/LiF NP arrays in longitudinal and transverse geometries (open and full markers, respectively). Panel B: calculated extinction spectra.

This research was conducted by the following research team:

  • Giulia Maidecchi, Luca Anghinolfi, Maurizio Canepa, Lorenzo Mattera, CNISM and Dipartimento di Fisica, Università di Genova, Genova, Italy
  • Grazia Gonella, Hai-Lung Dai, Temple University, Philadelphia (PA) U.S.A.
  • Remo Proietti Zaccaria, Istituto Italiano di Tecnologia, Genova, Italy
  • Angelo Giglia,  CNR-IOM, Trieste, Italy
  • Stefano Nannarone, CNR-IOM and Università di Modena e Reggio Emilia, Italy
  • Riccardo Moroni, Francesco Bisio, CNR-SPIN, Genova Italy

Contact person:
Angelo Giglia, email:


G. Maidecchi, G. Gonella, R. Proietti Zaccaria, R. Moroni, L. Anghinolfi, A. Giglia, S. Nannarone, L. Mattera, H. Dai, M. Canepa, and F. Bisio, “Deep Ultraviolet Plasmon Resonance in Aluminum Nanoparticle Arrays”, ACS Nano 7, 5834 (2013); DOI: 10.1021/nn400918n  
Last Updated on Friday, 04 October 2013 12:36