• enhancement factor
• microscopy
• Raman

Nowadays the study of Surface-enhanced Raman scattering (SERS) effect involves
many physics research groups worldwide. Since its first discovery in the 1973 by
Martin Fleischmann considerable improvements have been carried out. The principal mechanism proposed to explain the origin of SERS enhancement is named the
electromagnetic mechanism. It is directly connected to the local electromagnetic field
enhancement due to the excitation of surface plasmon resonances (SPR). Actually,
the main limit of the Raman scattering process is its very low effciency, and thus
very low signal intensity detectable from small amount of material. The SERS effect
allows the overcoming of this limitation, giving access to the prominent informations contained in the Raman spectra. This enables the distinction between similar
molecules due to the identification of their specific finger-prints and the possibility to
study the chemical and structural state of a limited number of molecules located at
a sub-micrometer scale near the SERS active sites. All these properties have important applications in the fields of biology (DNA recognition), analytical and applied
chemistry (catalysis) and renewable energy (photovoltaics).
Since the SPR is the main source of the SERS effect, the fabrication of metal micro
and nanostructures is requested. The goals of this thesis are the fabrication of metallic structures of different sizes and types and the analysis of their enhanced SERS
signals. The structural properties of the fabricated systems have been analysed by
Atomic Force Microscopy (AFM), the SERS signal by Confocal Raman Microscopy
(CRM). A monolayer of a Raman active dye (methylene blue) was deposited on the
surface of metal objects.
The work has been developed in collaboration between the Experimental Physics
Department of Ulm University and the Physics Department of Pisa. The SERS
active systems investigated in this thesis work are: (i) silver Fischer's patterns
(FP) with lateral sizes of 1 μm, 300 and 100 nm, (ii) silver micro and nano-bowls
(MNB) with diameters of 3, 0.9 and 0,4 μm, (iii) gold micro-cavities of 3 μm diam-
eter with size-controlled clusters of nanoparticles (400 nm diameter) located inside
them. In the first case a well established production technique has been employed,
the nanospheres-lithography (NSL), while in the case of the gold structures a novel
fabrication process is realized. The FP are 2-D triangular metal structures located
at the surface according to an hexagonal pattern, while the MNB are mainly 3-D
convex structures (i.e. 2-D lattices of overturned bowls) with the same periodical
structure. The cavities are instead hemispherical metal holes.
The FP structures are partially studied SERS emitters, but no systematic study of
length scale dependence is present in literature. Concerning micro- and nano-bowls,
just a few studies are present. The specific capability of the studied substrates
to boost the Raman signal is quantified as the enhancement factor (EF), namely
the ratio of the intensity of the enhanced Raman signal, over the signal measured
on a reference sample. Such quantities are normalized by the number of involved
Raman active molecules. At the state of the art, the Raman reference spectra are
measured on the dye powder or in-solution. This leads to arbitrary assumptions
and drawbacks. In this thesis, a new approach is proposed, adopting a dye single
crystal of micrometer size. A careful analysis of the localized changes in Raman
signal intensity over the surface of FB and MBN structures was performed using
statistical analysis methods and combining the CRM images with AFM morphology
data. In the case of the micrometer size objects, the Raman active sites have been
discriminated from the elastic scattering local maxima. The general behaviour in the
case of the FP and MNB systems shows considerable SERS EF increase in the case
of the smallest fabricated structures, in good agreement with the trend predicted
by the theory and the literature, when existing. EF passes from the order of some
thousands, as in case of the biggest structures, up to 7 orders of magnitude, for the
sub-micrometer structures.
Finally, a novel technique was used to fabricate gold micro-cavities containing clusters
of a controlled size of gold nanoparticles. In fact, many experimental and theoretical
studies have been carried out about the SERS signal from nanoparticles clusters, but
no reliable production processes are available at the state of the art for fabrication
of aggregates with the specified characteristics. The preparation method allowed the
production of periodically arranged, gold hemispherical cavities with the diameter
of 3 microns, acting as to trap and confine on their bottom limited clusters of gold
nanoparticles of 400 nm diameter. As expected, the gold cavities themselves did not
exhibit a significant SERS effect, but SERS active areas with repeatable size were
detected inside the hemispherical metal holes with trapped particles clusters. As a
future perspective, a significant increase of the SERS signal could be obtained by
scaling down to a smaller (sub-micrometer) scale the metal hemispherical holes and
reducing the size of the trapped nanoparticles. The preliminary results obtained
here can be also of interest for the development of SERS substrates with nanoscale
precision.