Soundscape
Sound engineering in theatres and clubs relies on placing speakers, adjusting their volume, and
modifying the timing of the signals they emit. These methods assume that the pressure waves that
we hear as sound reach the listener in almost the same way, and rely on using a large number of
speakers. But structures that can steer and focus sound waves even in audible wavelengths are
currently in development, and promise to completely change the way we listen
Structures whose arrangement
creates properties not found in
the base material are known as
metamaterials; since 2007 they
have been used to create lenses and
lters for light. More recently, similar
structures have been used to affect
acoustic waves. A few years ago, a group
of researchers from Sussex and Bristol
Universities used metamaterials in
the outside the range of hearing in the
ultrasound spectrum (20kHz-10MHz or
greater), frequencies used in medicine
for scanning, and in industry for crack
detection.
In 2017, they demonstrated the
ability to focus ultrasound waves into
an intersecting pattern that was stable
enough to hold in place a polystyrene
pellet in otherwise empty air some 10cm
above an electronic transducer.
The secret of this acoustic levitation
trick lies inside the lots of little 3D-printed
bricks that were contained in two 16x16
grids, one stacked on another, which
worked together to cradle the ball. The
bricks were all the same external size, and
all were all open at both ends. But inside,
they contain a twisting passage that
snakes back and forth across the width
of the brick. Some bricks’ insides were
completely open; some were a little bit
bendy; others were completely convoluted.
What these labyrinths do is delay the
time sound takes to pass through the
surface, explains University of Sussex
lecturer Gianluca Memoli. He adds: “If you
take a brick, and take an empty channel to
the side, and send sound through, it takes
a certain time to go through the empty
channel, and more time to go through the
brick, simply because it looks like a maze.
By changing the time delay, you can then
shape what happens to the sound beam
coming out.” He compares the bricks to
a ute, which is a tool for in uencing the
sound by changing the path that air takes
as it passes through. “What we discovered
in 2017 was that the library of bricks only
needs to contain 16 designs, to do any
shape of sound you might think about.”
So a metamaterial grid consists of
a two-dimensional array of 16 different
types of bricks, each with a different type
of internal labyrinth that delays sound
passing through it more or less. Designing
the grid lattice itself – determining which
bricks go where – he characterises as
‘a bit of an art’. (The 2017 grid, colourcoded
to represent different brick types, is
shown at bottom left; a photo of another
grid is above right). Generally, though, to
make a grid the designer works backward
from where the sound is needed; this
determines which bricks can be used
where. Also, the bricks have to be oriented
randomly, or can create directional
artefacts.
Memoli describes the bricks used in
the 2017 grids as big – not as physical
objects, as they were only 4mm by 4mm
by 8mm in the direction of the sound – but
because they were built so their length
corresponded to one whole wavelength.
When applied to the audible spectrum,
that’s a problem, since wavelength
increases as frequency decreases; using
exactly the same technology at say 100Hz
(in the bass region) would require each
block to be 3m long.
CONTINUING DEVELOPMENT
However, since then, researchers have
succeeded in shrinking them. Last year
Memoli presented blocks measuring only
2cm in length that are able to lter sound
between 4-6kHz, the last two octaves on
a piano.
Size isn’t the only problem in moving
metamaterials into the audible range.
Another issue is how to increase the
bandwidth of the focused sound to capture
the full range of tones in the spoken voice
or music. This is possible, but reduces
the transmission power ef ciency, Memoli
explains. He says: “It’s a design question.
The frequency range on which they
22 www.ied.org.uk
/www.ied.org.uk