The top-down approach limits the dimensions of devices to what is technically achievable using

Fig. 1 Epitaxial growth of sharp hetero-interfaces between different semiconductors (like GaAs and AlGaAs) for the formation of QWs and tunnel barriers.
October 2003
2 3
Fig. 2 Comparison of the quantized energy level structure of a ‘real’ and ‘artificial’ atom. In both cases, the attractive potential from the charged proton in the hydrogen atom and the
square well potential formed by the thin layer structure, respectively, binds the electron in discrete states. The value of ~0.3 eV is typical of band offsets between different semiconductors.

---

wavefunction of the electron can penetrate the barrier and
reappear on the other side. This is illustrated in Fig. 3 (top
right), where an incoming wave hits a potential barrier thin
enough to allow the wave to appear, damped, on the other
side of the barrier. This phenomenon is used in many
nanoelectronic and tunnel devices. It should be noted that
the thickness of the barrier is extremely important, since the
probability of tunneling through the barrier decays
exponentially with its thickness. Now we have the
ingredients, the concepts of the QW and tunnel barrier, to
form a double-barrier resonant tunneling (DBRT) device. This
important device structure consists of a QW surrounded by
two tunnel barriers, with low band gap materials on either
side of these barriers, from which electrons are fed into and
extracted from the central quantum region. This is illustrated
in the lower part of Fig. 3. Many of today’s most important
electronic and photonic technologies are based on the
concept of heterostructures, for instance the QW laser, high
electron mobility transistors, DBRT devices, and periodic well-
barrier-structures called superlattices used, for instance, in
quantum cascade lasers
2
.
Until now, the discussion has been limited to what can be
achieved using planar semiconductor technologies, i.e.
multilayer structures of thin semiconducting layers formed
by controlled deposition. The next challenge is that of
forming quantum structures of lower dimensions in which it
will be possible to limit the motion of electrons to one
dimension (1D), in a quantum wire or nanowire, or to lock an
electron into a zero-dimensional (0D) structure, a quantum
dot (QD), described above as an ‘artificial atom’. One
extremely important difference and advantage of the QD in
comparison with ‘real’ atoms is that it is, at least in principle,
possible to attach wires to a QD and electrically feed charge
carriers into it.
The ability to incorporate one- and zero-dimensional
structures into functional electronic and photonic devices has
been seen as a major challenge for many years. The dream
and motivation for such low-dimensional devices has been
discussed for more than 20 years, going back to the
predictions of Hiroyuki Sakaki in 1980, who suggested that a
one-dimensional nanowire would enable suppression of
scattering of charge carriers and lead to highly superior
transport properties
3
. There was also an effort at this time,
started by researchers at Texas Instruments Research
Laboratories, to etch out narrow columns from a
prefabricated planar DBRT-structure
4
, as illustrated in Fig. 4.
Similar approaches have since been tried with different top-
down fabrication techniques, but without satisfactory results,
primarily because of process-induced damage during
patterning and etching. Now, however, the long-time dream
of 1D-0D-1D tunnel devices has been realized by a bottom-
up, or self-assembling, technique.
REVIEW
FEATURE
October 2003
2 4
Fig. 3 (a) illustrates the way attractive potential QWs and potential barriers can be formed by heterostructures. (b) illustrates the concept of a double-barrier tunneling device, in which a
thin QW is sandwiched between two barriers, thin enough to allow electrons to tunnel through. (c) shows how an applied bias may bring the electrons in the emitter into and out of