Discovered in 1890 by Edouard Branly, "the
coherer effect" or "Branly effect" is an electrical transition from an
insulating to a conductive state of an oxidized metallic powder when an
electromagnetic wave is emitted in its vicinity. Such a wave
detector was at the origin of the first wireless radio transmissions at
the end
of the 19th century. However, the physical origin of this
phenomenon
still remains not well understood. A similar phenomenon of
conduction
transition is also observed when a DC voltage is directly applied to
the
sample and exceeds a certain value (we will call it herafter "DC Branly
effect").
At the Physics Lab. of the Ecole Normale Supérieure de Lyon,
various experiments (1D or 2D network of balls, and metallic powder)
were performed in order to understand the origin of this electrical
conduction transition. These studies were rewarded by the 2004 Branly Prize. All our
experiments are performed in DC, the influence of the wave frequency
will be undertaken later on.
By means of a model experiment with a chain of metal balls, we
show, for the first time, that the mechanism of the electrical
conduction transition (DC Branly effect) results from the local heating
of the microcontacts between each ball where microwelding occurs [1, 3, 4, 5, 7]. The increase in
temperature
reached 1050°C for an applied voltage as low as 0.4 V!
The electrical
conduction transition is connected to the local properties of the
contact
between two grains. It constitutes a first step towards more
realistic
granular media such as a 2D network of ordered balls (including the
disorder
of the contacts), or a metal powder sample (including the disorder of
position).
Electrical noise
& intermittency
We apply a DC voltage to a metal powder. Under certain
conditions, the temporal evolution of the current is then very
noised. We show that this electric noise has interesting
properties of scale invariance (over 4 decades in time) and of
intermittency which come from thermal expansions locally creating or
destroying the electrical contacts [2, 6, 7, 8].
These expansions can take place on various scales (size of the grain,
size of force network, size of
the sample): The small scales depend on the large ones with
similarities and differences with hydrodynamic turbulence. These
astonishing
phenomena of self-similarity are connected to the collective effects of
the granular matter.
Réferences
International articles :
[1] E. Falcon, B. Castaing and M. Creyssels, European Physical Journal B 38, 475 - 483 (2004)
