Agriculture Reference
In-Depth Information
the sand from the bed at the end of the co-firing experiments. The visual appearance of the
bed after the co-firing tests was however different from that after coal firing, see Figure 19
which suggests that the composition of the outer layers of the sand particles are different in
the two cases.
6.
C
OMPARISON OF
SEM
R
ESULTS
The levels of potassium with coal-PP co-firing were low but were considerably higher for
co-firing test with raffinate and vinasse. However there appeared to be a gradual increase in
potassium content when co-firing coal and pressed pulp. Armesto et al
.
(2002) also observed
the accumulation of potassium in the bed when bed material was not changed between the
runs. They observed increased amounts of potassium-calcium silicates in the bed. After 8
days (about 56 hours) of testing during current study potassium content had risen to 1.08%
(which is around 4 times higher than that observed after 3 days testing with coal, see Figure
20. This is still a low level, when compared with that for raffinate, and as mentioned earlier
there is no sign of bed clinkering. Consequently agglomeration is unlikely to be major
problem when co-firing pressed pulp with coal particularly when size degradation of the sand
during long term operation is taken into account. However a safe level for the potassium
content cannot be predicted so that much longer timescale tests may be required for a
definitive conclusion.
The levels of sodium on the outside of the ash and sand particles after co-firing were
lower than those of potassium, see Figure 21 with no evidence of a gradual increase over the
duration of the tests. Again however the sodium content after co-firing with pressed pulp is
substantially below the measured value after the short duration raffinate test. The highest
detected sodium levels occurred after 3 days of coal firing. However, no problems of
agglomeration were experienced after these tests so that the presence of relatively high levels
of sodium do not appear to be associated with bed clinkering. This is probably a result of the
relatively high melting point of sodium silicate (1088 °C) which is significantly above the
fluidized bed temperature. The relatively high level of sodium with Thoresby coal is in line
with the typical value of about 4.5% in the ash of this coal. The amount of coal used in the co-
firing tests was greater than in the 3 day coal alone case so that the low sodium levels with
co-firing may be due to the pulp “neutralising” the deposition of sodium silicate. It is worth
mentioning here that alkalis in coal and biomass behave differently during combustion and
the output of gaseous alkali metals is not steady [Hansen et al. 1995].
It can be observed from the Figures 20 and 21 that during coal only tests both sodium and
potassium started accumulating but sodium accumulation rate is considerably higher than
potassium, possibly due to higher sodium content in the coal. Comparison of the figures show
that during coal-PP tests potassium content in the bed increased with time but sodium content
first increased and then decreased which indicates interactive behaviour of sodium
accumulation is different when PP is co-fired with coal. The figures also show that potassium
and sodium content in the bed is very high after rafinate and vinasse co-firing tests. During
natural gas-raffinate test relative sodium content is higher than potassium content as
compared to coal-raffinate test which again indicates different behaviour of these alkalis
during co-firing.
