Pneumatic methods
There have previously been some attempts to create a consistant aerosol by entraining par-ticles in a stream of air. One of these methods have previously been tested at the Eindhoven University of Technology [5]. A similar method of dispersing fine powders was developed by TSI and tested by Chen et al. [12]. This dispersion system, which can be seen in Figure 7.1a, relies on the principle of a turntable onto which the powder is dispensed. After rotating 180 degrees, the powder is flushed from the turntable and delivered to a pipe through a capillary tube. The downside of this design is leakage of the powder and a lack of consistency in the flow. This was also found by Chen et al.
Another method that uses flushing with air is developed by Cassel, the concept for this apparatus can be seen in Figure 7.1b. A rotating drum is used which prevents agglomeration and keeps the powder in motion. A stationary outlet tube is placed inside the drum and air is supplied through a hypodermic needle to flush the powder into the outlet tube, creating a constant aerosol [25]. The air flux and density of the aerosol can be varied by varying the air flow and the rotational speed. In this experiment, particles of 6µm were used, and aerosol concentrations did not vary more than ±5% [25].
A fairly simple concept for flushing air from a packed bed is presented in Figure 7.2. This concept was created by Risha et al., and was used to study the effect of liquid and gas phases of water as the oxidizer combusting with both micron- and nano-sized aluminum particles [13]. Aluminium particles of 5 - 7 µm were used to create aerosols. In their research, Risha et al. used steam as fluid to entrain the aluminium particles. These particles are continuously replenished by means of a linear actuator which conveys the powder to the steam inlet, the particles are then carried away by the flow and are then ignited by a pilot flame on top of the device. The density of the aerosol is varied by the speed of the linear actuator, and the gas flux can be changed by changing the steam pressure.
CHAPTER 7. APPENDIX
in nitrogen [14]. The fluidized bed is very simple, but the consistency of the dust cloud density is less than that of other methods described.
(a) (b)
Figure 7.1: (a) Schematic of the TSI small-scale powder disperser, reprinted from Chen et al. [12]. (b) Dust disperser, reprinted from Cassel [25]
Figure 7.2: Schematic diagram of an aluminum dust cloud disperser, reprinted from Risha et al [13].
Design of a dispersion system for metal fuels 43
CHAPTER 7. APPENDIX
Figure 7.3: Schemtaic diagram of a fluidized bed aerosol generator with a feed system, reprinted from Prenni et al. [14].
CHAPTER 7. APPENDIX
Dispersion of powder is already being done by means of pressurised air in modern day dis-persion systems. A venturi system is popular, and a figure describing the concept can be seen in Figure 7.4. Because of the primary pressure in the air flow rate, a lower pressure gets introduced at the vacuum port of the venturi, and the powder is sucked towards the nozzle.
Pneumatic methods seem ideal, since the air that is used to flush and entrain the parti-cles can also (partly) be used as an oxidizer. A recurring problem however, is consistency of the aerosol. An ideal metal-air burner needs to have an aerosol with a consistent density and flux, and this consistency needs to be proven before a concept with this principle can be chosen.
Screw/auger methods
Screw or Auger methods usually take larger clumps of powder and deliver them in a flow.
A method of delivering fine powders in an air stream at a constant rate has previously been described by Atkins. This method, shown in Figure 7.5a, was developed to entrain fine dry coal powder in an air flow at a constant rate. This design uses an air flow which flows per-pendicular to the hopper, and makes the deposition of the powder more constant. Although the author claims to deliver the powder at a reasonably constant rate [15], no quantification of the accuracy of the system is given, and thus the density fluctuations in the aerosol might be too big for the use in this project.
Another method that may be relevant for the constant dispersion of fine powders is de-veloped by Affleck and displayed in Figure 7.5b. Affleck reviewed and tested several devices for the aerosolization of fine powders, including an open weir system. This system is based on a screw system which pushes a bulk of powder towards a slit. The powder is sucked into the slit and entrained into a stream of air [26]. Although this system is interesting in the large bulk applications, Affleck reported that the quality of dispersion was extremely variable.
Design of a dispersion system for metal fuels 45
CHAPTER 7. APPENDIX
(a) (b)
Figure 7.5: (a) Flow diagram of pulverized fuel feeder, reprinted from Atkins [15]. (b) Experimental weir-type dispersion generator, reprinted from Affleck [26].
Screw and Auger methods are interesting because of their ability to convey large quantities of powder. However, conveying small amounts of powder, and reducing fluctuations in the density of the aerosol are common problems that need to be addressed when working with a screw or auger system. Ideally, an auger or screw conveying system should be combined with a way to consistently disperse the powder, which would create a highly dynamic system with the least amount of fluctuations possible. While this may be challenging for smaller systems, such a concept would be useful for larger systems. In these systems, small fluctuations in the aerosol density are more acceptable.
CHAPTER 7. APPENDIX
Electrostatic methods
Shoshin & Dreizin created a method for electrostatic dispersion of fine powders. The ap-paratus with which they created a constant aerosol can be seen in Figure 7.6a. A master graduation project at the Eindhoven University of Technology was performed in order to create a similar apparatus. A sketch of this device, as made by Stiphout [17], can be seen in Figure 7.6b. Both devices use two electrodes, of which one is charged and the other one is grounded. The electrodes enable particles to ’bounce’ between them, and a carrying gas is used to create a flux of gas and particles through the small hole in the center of the top electrode. This method can be used to create a constant aerosol, however, most of the par-ticles that are carried away had a positive charge, and a shroud flow was needed to prevent diffusion of the powder [17].
Another method used for metering and retardation of solid powders using an electric field is demonstrated by Chen et al. As shown in Figure 7.7a, an experimental apparatus was cre-ated to retard the flux in a straight pipe connected to a hopper. With the use of electrodes, the dipole-dipole interactions influence the flow of solid particles between ∼ 6.5gs at 0 kV and ∼ 2.5gs at 6 kV [18]. The medium used as solid particles were granular nickel particles which had an average diameter of 0.25 mm, the particles were flowing through a pipe with an inner diameter of 3 mm. The nickel had a thin oxide layer to prevent interaction between the particles. A method was also found to bring the granular particles to a complete stop in another project execute by the same author. Here, granular nickel particles with an average diameter of 0.35 mm were used in a pipe with an inner diameter of 2 mm. When a voltage of 3kV was applied, the flow of the particles was halted [27]. This implies that the flow rate of metallic particles in an electric field relies on the particle size, the applied voltage and the diameter of the pipe. To the knowledge of the author, no experiments with finer particles or iron particles have been conducted, and performing these experiments would be useful for consideration of these concepts in the dispersion of iron powder.
An earlier adaption of the concept from Shoshin & Dreizin comes from Yu [16], this concept is shown in Figure 7.7b and shows an open system where charged particles are continuously replenished using an auger which transports the powder from a hopper to the reservoir. For these experiments, copper particles with a diameter between 26-80 µm were used.
Lastly, a method which relies on an electric field as the primary source for creating a constant aerosol is shown in Figure 7.8. This device was designed to operate without the use of a car-rying gas or any mechanically moving parts. The concept relies on a center electrode which is charged, and two outer electrodes which are grounded. The metallic powder (silver, coated brass, copper and zinc particles) is placed on the center electrode and lifted by potential difference. Because of the shape of the top electrode, the powder moves radially outwards and is then guided through the bottom hole [19]. The flow rate of the powder is controlled by varying the current.
Design of a dispersion system for metal fuels 47
CHAPTER 7. APPENDIX
(a) (b)
Figure 7.6: (a) Schematic design diagram of the aerosol feeder. The aerosol jet is shown to be fed into a flame zone to produce a steady, laminar aerosol, reprinted from Shoshin &
Dreizin [22]. (b) 2D schematical display showing the fundemental working of the electrostatic dispersion system, reprinted from Stiphout [17].
(a) (b)
Figure 7.7: (a) Apparatus for retardation of the flow field in a vertical pipe, reprinted from Chen et al. [18]. (b) Apparatus for the creation of an aerosol with the use of electrodes, reprinted from Yu & Colver [16].
CHAPTER 7. APPENDIX
Figure 7.8: Apparatus for the creation of an aerosol with the use of electrodes, reprinted from Olansen et al. [19].
dispersion is the charging of particles. If particles exiting the dispersion system have the same charge, dipole-dipole interactions will cause particles to repel each other and generate diffusion of particles in the burner. If a concept using electrostatic dispersion is chosen, it needs to be proven that this can be avoided.
Design of a dispersion system for metal fuels 49
CHAPTER 7. APPENDIX
Vibratory methods
Vibration has been used a lot in dispersion systems which disperse fine powders. The vibra-tion helps to break up agglomerated bulks of powders. A method that uses vibravibra-tion as an actual metering method is shown in Figure 7.9. This system uses vibration to reduce the adhesive forces of the powder within the capillary tube, and thus increase the flow rate. As is shown by Matsusake et al., the vibration frequency and the average diameter of the particles in the powder are critical in deciding the flow rate [20]. This system is capable of constant dispersion of powder with an average diameter of 6− 20µm, and because of the hopper, this system is also capable to sustain an aerosol for an extended period of time.
Figure 7.9: Schematic diagram of the experimental apparatus for micro feeding of a fine powder, reprinted from Matsusaka et al. [20]
CHAPTER 7. APPENDIX