Page 51 - Nanotechnology an introduction

Light Scattering Techniques
These are in principle more useful, especially for characterizing statistical (ir)regularity. Conventional scattering techniques include specular
reflexion, total integrated scattering, and angle-resolved scattering; the newer speckle techniques (speckle contrast, speckle pattern illumination,
and angle- or wavelength-dependent speckle correlation [105]) are of particular interest. In the speckle pattern illumination method [106], based on
doubly scattered coherent light, the (specularly reflecting) surface is illuminated with a monochromatic speckle pattern, whose phase distribution is
then modulated by the rough surface. In polychromatic speckle autocorrelation [107], the (diffusely scattering) surface is illuminated with a
collimated, partially coherent (i.e., polychromatic) light beam, either discrete (produced by a combination of laser diodes) or continuous (produced
by superbright light-emitting diodes, for example).
The distribution of collimated, typically partially coherent light scattered from a diffusely reflecting surface is suitable for determining its statistical
roughness up to about a quarter of a wavelength (i.e., about 150 nm for typical visible light sources). If the surface is specularly reflecting, the
illuminating light should itself be a speckle pattern, whose phase distribution is modulated by the asperity.
Evanescent optical wave-based techniques, especially its most recent variant, optical waveguide lightmode spectroscopy (OWLS), can yield
structural data on ultrathin layers, including geometric thickness, refractive indices, molecular orientation (e.g., [139]), the refractive index profile
perpendicular to the interface [115] and the lateral distribution of adsorbed objects (e.g., [8]), with some limitations regarding substrata (Table 5.1).
These techniques are especially useful because they can be used to measure processes in situ with good time resolution (Section 5.3) and the
nano/bio interface (Section 5.5), described in more detail in those sections.
Table 5.1 Summary of the ranges of applicability of optical techniques for investigating thin films
Thin film material
Technique Transparent dielectrics Opaque materials Metals
Scanning angle reflectometry (SAR) ✓ ✓ ✓
Ellipsometry ✓ ✓ ✓
Surface plasmon resonance (SPR) ✓
Optical waveguide lightmode spectroscopy (OWLS) ✓
Imaging Nanostructures
Ever since the invention of the microscope in the 17th century, science has been confronted with the challenge of exploring phenomena that are not
directly visible to the human eye. The same extension of the senses applies to “colors” only revealed using infrared or ultraviolet radiation, sounds
of a pitch too low or too high to be audible, and forces too slight to be sensed by the nerves in our fingers. Although artists sometimes maintain that
there is a qualitative distinction between the visible and the invisible, scientists have not found this distinction to be particularly useful; for them, the
problem of “visualizing” atoms is only technical, not conceptual.
Improvements in lenses, and other developments in microscope design, eventually enabled magnifications of about 2000-fold to be reached. With
that, objects around 100 nm in size could just be visualized by a human observer peering through the eyepiece of the microscope. The classical
microscope runs into the fundamental limitation of spacial resolving power Δx, due to the wavelike nature of light (Abbe's limit):

(5.2)
where λ is the wavelength of the illuminating light and N.A. is the numerical aperture of the microscope condenser. To address this problem, one
can

• reduce the wavelength of the light
• operate in the near field rather than the far field, as in SNOM (Figure 5.3)
• renounce direct imaging
• use a totally different approach (profilers, Section 5.1.1).
Wavelength Reduction
Although shorter-wavelength varieties of radiation (ultraviolet, X-rays) are well known, as the wavelength diminishes it becomes very hard to
construct the lenses needed for the microscope. However, one of the most important results emerging from quantum mechanics is the de Broglie
relationship linking wave and particle properties:

(5.3)
where λ is the wavelength associated with a particle of momentum p = mv, where m and v are the mass and velocity, respectively, and h is
Planck's constant, with a numerical value of 6.63 × 10 −34 J s. Hence, knowing the mass and velocity of a particle, we can immediately calculate its
wavelength.
The electron had been discovered not long before the formulation of the de Broglie relationship, and was known to be a particle of a certain rest
mass (m = 9.11 × 10 −31 kg) and electrostatic charge e. We know that opposite charges attract, hence the electron can be accelerated to a
e
desired velocity simply by application of an electric field. In other words, the wavelength can be tuned as required! Furthermore, ingenious
arrangements of magnetic fields can be used to focus electron beams. The transmission electron (ultra)microscope was invented by Ernst Ruska
and Max Knoll in the 1930s. Nowadays, high-resolution electron microscopy can indeed image matter down to atomic resolution. The space
through which the electrons pass, including around the sample, must be evacuated, because gas molecules would themselves scatter, and be
ionized by, fast-moving electrons, completely distorting the image of the sample. If the sample is very thin, the modulation (according to electron
density) of electrons transmitted through the sample can be used to create an electron density map (transmission electron microscopy, TEM).
Otherwise, a finely focused beam can be raster-scanned over the sample and the reflected electrons used to create a topographical image
(scanning electron microscopy, SEM, first developed by Manfred von Ardenne, also in the 1930s, although not commercialized until the 1960s). In
this case, if the sample is not electrically conducting, a thin layer of a metal, typically palladium, must be evaporated over its surface to prevent the
accumulation of those electrons that are not reflected, which is liable to obscure the finest features.

Alternatively, if the sample is a semiconductor with a not-too-large band gap, it might be practicable to heat it in order to make it sufficiently