(688d) Kinetic Modeling of Low-Temperature PM Combustion with Relative Velocity in the Presence of Water Vapor

Particulate
matter (PM) is a well-known
air pollutant, and mainly soot emitted from fossil fuel combustion. The
diameter of PM has decreased in recent years with improvement of combustion
technology. Especially, the concentration of PM_2.5 (diameter less
than 2.5 µm) increases. While bag filter, diesel particulate filter, and
electrostatic precipitator are conventionally used to remove PM from exhaust
gas of combustor, collection efficiency of these devices is not sufficiently
high for fine PM. In our research group, fluidized bed is used as a PM removal
device to collect fine PM by adhesion force, which is dominant force for PM diameter
of less than several tens µm. This device collects PM by depositing PM on bed
particle, and collection efficiency is 100% for PM_2.5 due to the
collection principle of this device. However, the amount of PM deposition
increases through continuous PM collection and PM aggregation is formed on bed
particle. PM aggregation becomes large with time. Large PM aggregation easily
separates from bed particle by getting large flow resistance, and collection
efficiency decreases. It is necessary to take counter measure for maintaining
high collection efficiency. PM is mainly composed of
combustible substances, and fluidized bed is well-known as a
low-temperature combustor. Thus, this device can burn PM at low-temperature. PM
combustion inhibits the formation of PM aggregation, and collection efficiency easily maintains initial value.
It is observed that the fluidized bed type PM removal
device is continuously regenerated at the bed temperature of 400 ‹C while
maintaining the collection efficiency of 100%, in contrast to the conventional
continuous regeneration devices that require temperatures of 600 to 650
‹C for PM combustion [1].

Water vapor is
generated by fossil fuel combustion. Exhaust gas of combustor includes water
vapor at concentration of 5-15 vol.%, but the effect of water vapor on
continuous regeneration is not investigated in our previous work [1]. It is
reported that water vapor promotes PM combustion at above several hundred
degrees. Moreover,
the combustion rate of PM increases with relative velocity between gas and
solid on PM collection and combustion experiments by means of this device [1]. In order to further develop this device as a continuous
regenerating type PM removal device, it is
necessary to understand the transport phenomena and PM combustion
characteristics in this device. The effects of water vapor and relative
velocity on carbon combustion were investigated at low temperatures, and the
kinetic model was constructed to take the relative velocity into account. 
The kinetic model was applied to numerical simulation of a fluidized bed type
PM removal device to confirm the validity of this model and to investigate transport
phenomena, PM adhesion characteristics, and PM combustion behavior in the
presence of water vapor.

PM combustion rate in the presence of water vapor should be
measured with considering relative velocity between gas and PM. A new
thermogravimetric apparatus was built to create collision between gas and PM. Figure
1 shows a schematic diagram of the experimental apparatus. PM is
maintained in thermogravimetric apparatus by hanging on electric balancer. Water vapor is added at
concentration of 0-15 vol.% in this experimental apparatus by saturating air with
water vapor. Air
including water vapor was collided with PM to represent contact between gas and
PM in the fluidized bed type PM removal device. Relative
velocity between the gas and PM was expressed as the superficial velocity. Carbon
black was used as a PM on collection and combustion experiments by means of fluidized
bed. However, it is difficult to be placed
carbon black in this thermogravimetry apparatus while direct collision with
air. Carbon block was used to collide with gas in this thermogravimetric apparatus,
and it has almost same combustion rate per unit Brunauer-Emmett-Teller (BET)
surface area as carbon black. Combustion rate was obtained from measuring
weight change of carbon block by electric balance.

Figure 2 shows the plot of combustion rate versus the relative
velocity between the carbon block and the gas at temperature of 400 ‹C, water
vapor of 0 and 10 vol.%. Combustion rate at water vapor 10 vol.% is larger than
that at water vapor 0 vol.%. The promoting effect of water vapor is detected
from this result. It is reported that complex groups (e.g., carboxyl groups)
are generated with
water vapor even at several hundred degrees, and increase with the water vapor
concentration. Carboxyl groups transfer into CO₂ and OH at
temperatures ranging from 100-400 ‹C. It is considered that this is the main
reaction for the promotion of PM combustion by water vapor at bed temperature of
250-400 ‹C. Moreover, combustion rate increases with increasing gas-PM relative
velocity. The carbon combustion rate depends on oxidant concentration at carbon
surface. Oxidant
concentration decreases by PM combustion and increases by oxidant mass
transfer. When the consumption of oxidant is larger
than the supply of oxidant, the oxidant concentration decreases and the combustion
rate becomes small. High combustion rate is obtained owing to supplying more oxidant
to the carbon surface by a rise of relative velocity. Especially, the
dependence on relative velocity at water vapor of 10 vol.% is larger than that
at water vapor of 0 vol.%. Oxidant consumption increases due to the promoting
effect of water vapor on PM combustion. It is necessary to supply more oxidant to
maintain PM combustion rate. Thus, the effect of relative velocity becomes
large as water vapor.

Numerical
analysis of fluidized bed type PM removal device is performed under the same
condition as those for our previous experimental work. The Euler-Euler method
was used for the simulation of fluidized bed type device. PM adhesion is
represented by adhesion model. Adhesion model is constructed in our previous
study to represent adhesion behavior [2]. It describes the relationship between
the amount of the PM deposition and collection efficiency. Collection
efficiency is calculated as a function of the amount of PM deposition and
decreases with increasing amount of PM deposition. Kinetic model is constructed
in the thermogravimetric experimental work, and kinetic parameters are
determined by fitting. Kinetic model of PM combustion was incorporated into the
simulation code for the fluidized bed type PM removal device. The effect of
relative velocity was taken into account as a mass transfer term in this
kinetic model [3]. PM combustion amount is calculated
from the kinetic model. The amount of PM deposition decreases with increasing
PM combustion amount. Figure 3 shows time histories of collection
efficiency at superficial velocity of 0.4 m/s, PM diameter of 100-125 µm, PM
concentration of 100 mg/m³, bed temperature of 20, 400 ‹C, water vapor
0 and 10 vol.%. PM diameter and concentration are severe
conditions to clearly observe the effects of PM combustion and water vapor on
continuous regeneration of this device. The numerical and experimental
collection efficiencies decrease with time at all conditions by
formation and separation of PM aggregation on bed particles. Numerically
obtained collection efficiency is good agreement with experimental data at
ambient temperature. Thus, if numerical results agree with experimental results
at continuous regenerating mode, the validity of kinetic model is confirmed. Collection
efficiency at 400 ‹C is larger than that at ambient temperature, because PM
combustion decreases the amount of PM deposition and inhibits formation of PM
aggregates on the bed particle. Furthermore, collection
efficiency increases in the presence of water vapor. The amount of PM
deposition at water vapor 10 vol.% are maintained smaller than that at water
vapor 0 vol.% due to promoting effect of water vapor on PM combustion. The
differences between numerical and experimental obtained collection efficiencies are
at most 1.6%. The numerically obtained collection
efficiency shows a reasonably good agreement with the experimental collection
efficiency. As a result, the validity of the kinetic model was confirmed.

[1] Yamamoto, T.,
Yokoo, K., Kusu, A., & Tatebayashi, J. Highly efficient particulate matter
removal by a fluidized-bed-type device operated in continuous regeneration mode.
Powder Technology 2018, 323, 86-94.

[2] Yamamoto, T.,
Tsuboi, T., & Tatebayashi, J. A numerical simulation of PM adhesion
characteristics in a fluidized bed type PM removal device by a finite volume
Eulerian–Eulerian method. Powder Technology 2016, 288, 26-34.

[3] Guedea, I., Díez,  L.I.,
Pallarés, J., & Romeo, L.M. On the modeling of oxy-coal combustion in a
fluidized bed. Chemical Engineering Journal 2013, 228, 179-191.