3.2. Pure glycerol
The results with crude glycerol produced in a biodiesel plant
were compared with the behavior of pure glycerol. With that purpose,
high purity (>99.5%) glycerol was injected into the DCF with
the same initial diameter and experimental conditions. The results
are shown in Fig. 7 in terms of the evolution of normalized d2 along
the axial distance. Up to 60 mm the curves are almost identical,
regarding both the initial heating period and the slope of the subsequent
linear decay. The difference is that the pure glycerol
evolves with constant slope during the full lifetime of the droplet,
whereas the evolution of the crude glycerol droplets is suddenly
modified due to the microexplosions. The actual combustion history
of the droplets produced by this secondary atomization was
not recorded. In any case, if the size of the droplet is reduced by
a factor >3 (which, according to the images recorded, happens in
most cases), their combustion time is reduced by a factor >10;
i.e., the length until total conversion of the crude glycerol droplets
can be estimated as slightly longer than the axial distance to
microexplosions. Since disintegration occurs when (d/d0)
2 0.42,
and according to the d2 law, the time elapsed until the microexplosion
is 58% of the time for total evaporation of pure glycerol droplets.
If the fragments produced in the microexplosion are
consumed about 10 times faster than the parent droplet, this rough
analysis suggests that the total combustion time of the crude glycerol
droplets could be of the order of 65% of the time required to
fully evaporate the pure substance.
Incidentally, only a very weak blue luminosity could be seen
along the axis of the tube. This further confirms the very low sooting
tendency of pure glycerol, in clear contrast with the intense
radiation observed around fuel-oil droplets (see Fig. 5(c)) caused
by a high concentration of soot surrounding the burning droplets.
3.3. Blends with pure glycerol
The absence of any disruptive behavior of the pure glycerol
indicates, as expected, that microexplosions are induced by some
other species contained in the crude glycerol. A systematic investigation
was followed to determine the actual role of different candidate
components in this regard.
Crude glycerol typically contains 60–85% of glycerol, combined
with variable amounts of water, methanol, NGOM and salts. In
order to investigate which of those components is responsible for
the disintegration of the droplets, a wide range of pure glycerol