COMPOSITE DISCHARGE ELECTRODE FOR ELECTROSTATIC PRECIPITATOR

A thesis presented to the faculty of the Russ College of Engineering and Technology of Ohio University

In partial fulfillment of the requirements for the degree Master of Science

Jason M. Morosko March 2007

This thesis entitled COMPOSITE DISCHARGE ELECTRODE FOR ELECTROSTATIC PRECIPITATOR

by JASON M. MOROSKO

has been approved for the Department of Mechanical Engineering and the Russ College of Engineering and Technology by

Khairul Alam Moss Professor of Mechanical Engineering

Dennis Irwin Dean, Russ College of Engineering and Technology

Abstract MOROSKO, JASON M., M.S., March 2007, Mechanical Engineering COMPOSITE DISCHARGE ELECTRODE FOR ELECTROSTATIC PRECIPITATOR (54 pp.) Director of Thesis: Khairul Alam This thesis focuses on the design, manufacture, and testing of discharge electrodes made from composites for use in electrostatic precipitators (ESP). Standard industry ESP’s use carbon steel, or stainless steel as the base material for constructing discharge electrodes. The electrodes need to be conductive and resist corrosive environments. It is proposed that a conductive polymer matrix composite material can be used in place of the steel to achieve both the necessary conductivity performance, and withstand the corrosive environments in the ESP.

Approved: Khairul Alam Moss Professor of Mechanical Engineering

Acknowledgments I would specifically like to thank and acknowledge Dr. Khairul Alam for his shear devotion and dedication, first to the advancement in education of all of his current and prior students, second to the contributions he has already and will make to the engineering community, and third to the needed help he provides through the university engineering department to the people and businesses in the Athens community.

- thank you

5

Table of Contents Page Abstract .............................................................................................................................. 3 Acknowledgments .............................................................................................................. 4 List of Tables ..................................................................................................................... 6 List of Figures .................................................................................................................... 7 Chapter 1: Topic Introduction ........................................................................................... 9 1.1: Goal ....................................................................................................................... 9 1.2: Background ......................................................................................................... 10 1.2.1: Electrostatic Precipitation ......................................................................... 10 1.2.2: ESP Electrodes ......................................................................................... 14 1.2.3: Wet and Dry ESP...................................................................................... 18 1.2.4: Conductive Polymers................................................................................ 19 1.3: Summary ........................................................................................................... 20 Chapter 2: Experiment .................................................................................................... 21 2.1: Design and Fabrication of the ESP Test Chamber ............................................ 21 2.2: Electrode Fabrication ......................................................................................... 27 Chapter 3: Testing ........................................................................................................... 34 3.1: Introduction ........................................................................................................ 34 3.2: Results ................................................................................................................ 35 Chapter 4: Discussion and Future Work ......................................................................... 48 4.1: Introduction ......................................................................................................... 48 4.2: Conclusion .......................................................................................................... 48 4.3: Future Work ........................................................................................................ 48 References ........................................................................................................................ 50 Appendix A ..................................................................................................................... 52

6

List of Tables Table

Page

3.1: Sample 001 VI Curve Test Data ...........................................................................36 3.2: Sample 002 VI curve and collection comparison data …………………….........37 3.3: Sample 003 VI curve and collection comparison data …………………….........38 3.4: Sample 004 VI curve and collection comparison data …………………….........39 3.5: Sample 005 VI curve and collection comparison data ………………….............40 3.6: Sample 006 VI curve and collection comparison data ………………….............41 3.7: Sample 007 VI curve and collection comparison data ………………….............42 3.8: Sample 008 VI curve and collection comparison data …………………….........43 3.9: Sample 009 VI curve and collection comparison data …………………….........44 3.10: Sample 010 VI curve and collection comparison data ………………………...45

7

List of Figures Figure

Page

1.1: Graphical Representation of ESP (top)....................................................................12 1.2: Graphical Representation of ESP (side) ..................................................................12 1.3: Full Scale ESP Example .........................................................................................13 1.4: Pipe and Spike Discharge Electrodes .....................................................................14 1.5: Collection Plates .....................................................................................................15 1.6: Elex Discharge Electrode Detail.............................................................................15 1.7: Elex Discharge Electrode Picture ...........................................................................16 1.8: Small Scale Rod and Star Electrode .......................................................................17 2.1: ESP Top Assembly ..................................................................................................22 2.2: ESP End View .........................................................................................................22 2.3: ESP Inside Detail .....................................................................................................23 2.4: ESP Electrode Detail ...............................................................................................23 2.5: Typical Test Electrodes (Iten Industries).................................................................24 2.6: Typical Test Electrodes (Other)...............................................................................24 2.7: ESP Chamber Drawing............................................................................................25 2.8: ESP Chamber End View..........................................................................................25 2.9: ESP Chamber Component Layout...........................................................................26 2.10: Detailed Drawing of an Electrode........................................................................ 26 2.11: Sample 001 Stainless Steel................................................................................... 28 2.12: Sample 002 Iten Rod.............................................................................................28

8 2.13: Sample 003 Iten Rod.............................................................................................29 2.14: Sample 004 Iten Rod.............................................................................................29 2.15: Sample 005 Iten Rod.............................................................................................30 2.16: Sample 006 Iten Rod.............................................................................................30 2.17: Sample 007 Sandwich Carbon Matt .....................................................................31 2.18: Sample 008 Carbon Disks ....................................................................................31 2.19: Sample 009 Carbon Plate Stars.............................................................................31 2.20: Sample 010 Carbon Matt ......................................................................................32 3.1: Collection Comparison ...........................................................................................35 3.2: VI Curves Overlay ..................................................................................................46

9

Chapter 1: Topic Introduction This thesis concentrates on a specific element of both wet and dry electrostatic precipitators (ESP) known as the discharge electrode. Electrostatic precipitators are well known industrial devices used to capture exiting particulate (or mist) being created from an industrial manufacturing type process. Examples of such processes include coal fired power plants, insulation manufacturing plants, pulp and paper mills, petrochemical and chemical plants, glass production plants, and so forth (Southern Environmental, 2006a). Ohio University has had a long collaborative research and development program with Southern Environmental, Inc. (SEI) of Pensacola, Florida for many years. This thesis will focus on electrode designs made by SEI, which are to be constructed of polymer composite materials. Electrically conductive polymer composites can be highly advantageous in corrosive applications. Conductive polymers can be made by adding in small volume fractions of carbon nanofiber into the mix before processing. The carbon will not significantly change the mechanical properties, but it will enhance the electrical properties suitable for certain applications where conductivity is required. (Alam et al., 2006). The project will use this technology to produce conducting polymer composite electrodes. 1.1 Goal The typical electrode is made out of steel, making it heavy and highly susceptible to corrosion failure over time. Corrosion resistant steels are very expensive. Polymers are corrosion resistant and cheap; however, polymer materials are not electrically

10 conductive. The goal of this project is to design, manufacture, and test discharge electrodes using electrically conductive fiberglass composite. The objective is to produce a composite electrode that will be suitable in strength, lighter, adequately conductive, and more corrosion resistant than the current steel electrode. 1.2 Background 1.2.1 Electrostatic Precipitation Electrostatic Precipitators were invented by Frederick Cottrell and have been in operation since 1907. Over this time, ESP design has changed greatly, mainly due to the changes in emission regulations for industrial manufacturing plants. In addition, the industry has been constantly changing due to new techniques and products that are constantly being produced (Mastropietro, 1998). Processes which clean industrial manufacturing exhaust gases can be grouped into two categories, mechanical and electrical. Mechanical cleaning generally refers to processes using gravitational effects, filtering, centrifugal, cyclonic separation, or similar methods to remove particulate from the contaminated air stream (White, 1963). Electrical gas cleaning processes are referred to as electrostatic precipitation (ESP). In ESP, the separation forces are directly focused on the contaminant particles instead of the gas stream as a whole. Use of this method puts no limit to the degree to which an air stream can be cleaned. Tailored designs can capture up to and over 99% of particulate entrained in a given air stream. High capture efficiency, low gas flow resistance, and the capability of ‘cleaning’ large gas flow volumes at high temperature and high corrosivity

11 has made ESP the most widely accepted process for particulate removal in this industry to date. (White, 1963) The ESP fundamental process can be described in three basic steps (White, 1963): •

Application of charge to particles suspended in gases.

•

Capture of these particles by electrodes in an electric field.

•

Cleaning of collection electrodes to remove the particulate.

In a typical electrostatic precipitator, exhaust gases carrying particulates pass through ducts which have charged discharge electrodes and oppositely charged collection surfaces. A highly charged discharge electrode with a sharp point will emit what is known as a corona discharge. In the immediate area of the tip of the electrode, the air will become ionized, making it partially conductive. These charged particles will migrate naturally to an oppositely charged collection surface where they will become neutralized. This is the process by which the passing particulate in a particulate laden stream of gas is ‘filtered’ of its contaminants. The collected particulate remains on the collection surface until mechanical removal such as vibration or rinsing is performed (Corona discharge, 2007). Figures 1.1 and 1.2 graphically illustrate how the particulate laden gas enters the ESP, becomes ionized by the corona discharge, and is driven to the nearby collection surface. Figure 1.3 depicts one example of a full scale ESP operation using vertical collections plates attached to mechanical rappers for particle removal.

12

Figure 1.1: Graphical representation of ESP looking down from above (Dayley et al., 2003).

Figure 1.2: Graphical representation of ESP looking in from the side (Dayley et al., 2003).

13

Figure 1.3: Full scale ESP example (Environmental Works, 2006).

ESP ‘science’ is proven mostly in the field, as ESP cost and size prevents smaller scale testing facilities to adequately model the wide range of environments. Typical startup ESP facilities see anomalies that must be corrected in practice as the theory used to design them normally does not foresee all encountered situations (Mastropietro, 1998). The trend over the lifetime of operation of ESP’s shows that ESP’s have grown in size to accommodate the regulations. This points to the fact that a larger ESP (of the same construction) will capture more particulate than a smaller one. A larger ESP has longer flow lengths which allows the polluted air steam a longer exposure to the ESP resulting in greater capture. Earlier regulations focused on particulate capture

14 efficiencies, but new regulations are based totally on particulate emissions. Capture efficiencies are generally not even measured (Mastropietro, 1998). 1.2.2 ESP Electrodes This section will discuss two types of electrodes in the ESP, the discharge electrode and the collection electrode (sometimes referred to as the collection plate). The purpose of the discharge electrode is to produce the corona which ionizes the surrounding air, thus charging and driving particulate to the collection plate. Discharge electrodes have changed over the years, starting out as a thin wire design. The typical problem with this design was wire breakage, resulting in inadequate reliability.

More

recent electrodes fall into several construction types including stringing wires, thin metal strips, roll formed masts, and the pipe and spike arrangements. Figure 1.4 shows the pipe and spike discharge electrode, along with a plate type collection surface (Mastropietro, 1998).

Figure 1.4: Pipe and spike discharge electrodes. (Hamon Thermal Transfer, 2006).

15

Figure 1.5: Collection plates. (Hamon Thermal Transfer, 2006).

Southern Environmental (SEI) is in the business of rebuilding and new construction of wet and dry electrostatic precipitators and scrubbers. Most aspects of their systems are patented, tested components. SEI also manufactures a patented design discharge electrode, referred to as the RS Discharge Elex Electrode shown in figures 1.5 and 1.6. The design consists of a formed carbon steel part having cantilevered dual emitter arms to uniformly distribute the corona, and prevent dust build up (Southern Environmental, 2006d).

Figure 1.6: Elex discharge electrode detail.

16

Figure 1.7: Elex discharge electrode picture. The electrodes are then independently hung from their own suspension system. The goal of a good electrode system is to have the corona discharge equally dispersed throughout the entire effective length of the precipitator column. Optimum corona dispersion results in uniformly charged particles giving the best collection performance. In gas passages, the electrodes are typically spaced 8” – 16” apart (Southern Environmental, 2006d). Other ESP applications installed by SEI incorporate an electrode that is more of a rod and star assembly. These applications tend to be used in high corrosive environments and require stainless steel alloys such as Hastelloy 2000. In these cases, a conductive composite rod electrode would be of tremendous cost advantage, lighter, and perform for a longer period of time. Figure 1.7 depicts a small scale rod and star discharge electrode arrangement, made for testing in this project.

17

Figure 1.8: Small scale rod and star electrode.

In the typical ESP’s, the discharge electrodes, collections plates, and all other devices associated with the ESP process are subject to particulate build up on all available surfaces. A build up can result in loss of efficiency due to air flow disturbances and lowered electrical clearances between the collection plates and electrodes. In order for dust collection removal, the plates are ‘rapped’ or vibrated by an electromagnetic or pneumatic mechanical rapping device. They are designed to remove the collected particulate periodically while the ESP is in operation. The rapper is a solenoid controlled, electromagnetic device which operates a plunger up and down by applying and releasing a magnetic charge. The plunger is dropped onto a rapper bar connected to a bank of collection plates, effectively dislodging the collected particulate. (Southern Environmental, 2006b). Results from ESPs show that with typical air velocities of 0.6 – 1.2 M/sec, and collection plate spacing of 300 – 400 mm works well. Velocities higher than that need to have collections plates spaced closer together to adequately charge and collect the faster moving particles (Mastropietro, 1998).

18 1.2.3 Wet and Dry ESP ESP’s can be made to operate either wet or dry depending on the particulate that is to be collected, and the overall efficiency goals to be met. Wet ESP’s run a liquid over the collection plates to remove particulate. There are several distinctions that can be made concerning wet ESP’s relative to dry ESP’s. Wet ESP’s use a metal or fabric type collection surface. They typically result in higher capture of sub-micron particulate due to the better adhesion of the particles to the collection surfaces’ cleaning liquid, and virtually no re-entrainment. Wet ESP’s are used when fine particles need to be removed from the exhaust air stream, typically with aerodynamic diameters less than 2.5 µm, also stated as PM2.5. In a dry ESP, fine particulate gets lodged on the collection surface, and cannot be ‘rapped’ free fully, therefore lowering the corona discharge and resulting in lower efficiencies. Wet ESP’s remove the particulate by running or spraying water on the collection surfaces. Both running and spraying of water can be problematic for metal plates. Spraying leads to an easy path for the high voltage to short to the nearest ground, and running water leaves parts of the plate dry for particle build up. Corrosion is also a major factor in wet ESP, and typically results in high cost steel alloys being used, making them cost prohibitive. One of the latest advances in wet ESP is the development of a membrane collection surface utilizing capillary action as the collection surface cleaning process. This process eliminates the spraying and running liquid problems discussed. In addition, certain woven polymer fiber arrangements that were tested eliminated the corrosion problem, lowered the associated cost for the collection plate, and proved

19 sufficient for the capillary action and removal of particles. Both manufacturing costs, and raw materials cost attribute to a lower cost of the membranes (Bayless et al., 2004). WESP metal collection plates (and electrodes for that matter) are manufactured from high cost high alloy stainless steel due to the corrosive environments with chlorine and acidic aerosols. The introduction of the membrane collection surface produced tremendous cost savings, on the basis of the price comparison of the membrane material and the costly alloys that are currently used. Therefore, electrodes manufactured using a conductive composite, if one were available, would reduce cost and increase the working life. 1.2.4 Conductive Polymers Electrically conductive polymer composites can be very useful in working environments that are too corrosive for typical metals. An example that has been discussed is the electrostatic precipitator, which is widely used by the power industry to remove particulates and mist from flue gases. The gases can be highly corrosive, which leads to the use of expensive metal alloys such as hastelloy. The use of metal alloys also requires expensive structural support for the structures. These problems can be alleviated by the use of a conducting polymer composite (Alam et al., 2006). To make a conducting polymer, carbon nanofibers can be used as an additive in small volume fractions. When the volume fraction of the nanofiber is small, the nanofiber may not provide much enhancement of the mechanical strength; but the electrically conductivity can be improved significantly. However, producing a composite with nanofibers requires careful processing steps because the stiff nanofibers must be

20 dispersed without significant breakage. If the fibers are broken due to the mixing action, the reduction in fiber length can significantly reduce the conductivity because the shorter fibers may not provide a continuous path for the flow of electrons. (Alam et al., 2006) In a project funded by Edison Material Technology Center (EMTEC) partner company (Iten Industries of Ashtabula, Ohio) applied this technology into their polymer processing line, which resulted in the making of conductive rods for the backbone of several discharge electrodes for the experiment (figure 2.5). They made several samples, varying the amount of carbon to test for the minimum volume fraction required to achieve the necessary conductivity. Their electrodes were then tested in the ESP laboratory at Ohio University. 1.3 Summary Chapter 1 focuses on the general design and background of the electrostatic precipitation concept as applied to the removal of particulate contained in exhaust gases created by industrial processes. The discharge electrode, being a major system component in ESPs has also been discussed. An electrically conductive composite polymer material for use as the discharge electrode was introduced and described in this chapter. In chapter 2, the experimental apparatus and its construction has been laid out. Chapter 3 explains the testing performed, and results obtained. Finally, chapter 4 summarizes the results, and provides direction for possible future work on the topic.

21

Chapter 2: Experiment For this project, it was necessary to design and construct an ESP chamber to conduct tests on sample discharge electrodes using electrically conductive composite materials for the rod or backbone of the electrode, and in some cases for the discharge tip of the electrode itself. 2.1 Design and Fabrication of the ESP Test Chamber Initial results in a pre-existing ESP setup showed promising collection data for the composite electrodes. During initial testing, it was found that dust humidity levels were a major factor in collection amounts for the apparatus being used. Successful testing was conducted on less humid days. As testing continued, it was determined that more initial environmental characteristics needed to be recorded in order to validate one test to the next. The input location for the fly ash feeder could not be held constant, the air flow was not being measured, electrodes and collection plates could not be easily removed, and the apparatus was extremely leaky. As a result, a new apparatus was built which addressed all of these issues. The new chamber consisted of an interior rectangular duct shape, 10 inches wide by 8 inches high, and 8 feet long. Construction materials for the duct were one quarter inch acrylic plate as duct walls, acrylonitrile-butadiene-styrene (ABS) angle brackets for support, and a five minute set epoxy for assembly. In figure 2.1, the middle section of the testing chamber is shown. Figure 2.2 shows the fly ash feed tube arrangement at the leading end of the chamber. Figure 2.3 depicts the setup with one collection plate

22 removed, and figure 2.4 is a close up of the discharge electrode being tested, with one collection plate removed.

Figure 2.1: ESP top assembly.

Figure 2.2: ESP end view.

23

Figure 2.3: ESP inside detail.

Figure 2.4: ESP electrode detail.

In figure 2.5, typical test electrodes constructed out of rods made at Iten Industries are shown. Pictures of other test electrodes are represented in figure 2.6. The test chamber arrangement contained three electrodes, two grounded collection plates, and one grounded wire mesh plate. All electrodes were of the rod and star design shown in figure 1.7. Two electrodes were used as pre chargers to produce initial charging of the passing particulate. The third electrode was the sample discharge electrode being tested. The electrodes, pre-chargers, collection plates, and grounded wire

24 mesh were all attached to removable top plates in the middle of the 8’ long duct. An isometric representation is shown in figure 2.7, as well as an end view detail in figure 2.8.

Figure 2.5: Typical test electrodes Produced by Iten Industries & Ohio University for this Project.

Figure 2.6: Typical test electrodes produced by Ohio University for this project.

25

Figure 2.7: ESP chamber drawing, isometric view.

Figure 2.8: ESP chamber end view.

A dimensioned drawing of the interior layout can be seen in figure 2.9, and the electrode design is detailed in figure 2.10.

26

Figure 2.9: ESP chamber component layout, top view.

Figure 2.10: Detailed drawing of the Electrode.

The specific equipment that was used for the testing is listed below: •

Fly ash feeder:

Company: Schenck Accurate. Model: mod102M (2003)

27 •

Voltage Regulator:

Company: NWL Transformers. Model: Switchmodel Resonant T/R #:102462 Rev B.

•

Blower:

Company: Nailor Industries. Model: 6” flow duct.

Other equipment consisted of an air compressor to supply feed air to the fly ash feeder, variable speed exhaust fan for air flow control, and various multi-meters for VI curve measurement. 2.2 Electrode Fabrication Iten Industries (Ashtabula, Ohio) produced sample hollow rods containing small volume fraction amounts of conductive carbon nanofibers produced by Applied Sciences, Inc., in Cedarville, Ohio. The rods were shipped to Ohio University and used to construct some of the sample discharge electrodes. The rods were first cut to a six inch length. Next, three stainless steel ninja stars were pressed onto the outside of the rod, and fixed using a conductive epoxy. To attach the rod to the test chamber, and to attach the high voltage supply cable, a nut welded to a washer was fixed to the top of the rod by conductive epoxy; therefore, creating a conductive path from the top of the rod to each of the stars. Iten produced several rods containing different amounts of nanofiber to test the lowest threshold of conductivity required to allow the ESP to function. Several other electrodes were constructed using solid carbon rods as backbones, and different conductive polymer materials as discharge tips to test various other combinations for use in ESP’s. All electrodes were assembled similar to the above example. The sample descriptions are provided below in detail.

28 •

Sample 001 was constructed entirely from 316 SS, rod and stars, welded together. This sample is the base electrode against which all sample composite electrodes will be tested. In addition, two of these were also used as the pre-chargers.

Figure 2.11: Sample 001 Stainless Steel •

Sample 002 was constructed from a pultruded conductive hollow rod made by Iten Industries. This rod had a resistivity of 80.12 Ohm-cm. Stainless steel 316 stars were fixed to the rod by light press fit, and conductive epoxy. A steel washer with a welded nut was also attached via conductive epoxy for mounting the sample to the ESP setup.

Figure 2.12: Sample 002 Iten Rod •

Sample 003 was constructed from a pultruded conductive hollow rod made by Iten Industries. This rod had a resistivity of 16.02 Ohm-cm. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. A steel washer with

29 a welded nut was also attached via conductive epoxy for mounting the sample to the ESP setup.

Figure 2.13: Sample 003 Iten Rod •

Sample 004 was constructed from a pultruded conductive hollow rod made by Iten Industries. This rod had a resistivity of 2.76 Ohm-cm. Stainless steel 316 stars were fixed to the rod by light press fit, and conductive epoxy. A steel washer with a welded nut was also attached via conductive epoxy for mounting the sample to the ESP setup.

Figure 2.14: Sample 004 Iten Rod •

Sample 005 was constructed from a pultruded conductive hollow rod made by Iten Industries. This rod had a resistivity of 96.37 Ohm-cm. Stainless steel 316 stars were fixed to the rod by light press fit, and conductive epoxy. A steel washer with a welded nut was also attached via conductive epoxy for mounting the sample to the ESP setup.

30

Figure 2.15: Sample 005 Iten Rod •

Sample 006 was constructed from a pultruded conductive hollow rod made by Iten Industries. This rod had a resistivity of 1176.1 Ohm-cm. Stainless steel 316 stars were fixed to the rod by light press fit, and conductive epoxy. A steel washer with a welded nut was also attached via conductive epoxy for mounting the sample to the ESP setup.

Figure 2.16: Sample 006 Iten Rod •

Sample 007 was constructed from a solid carbon rod with carbon fiber matt used as discharge tips. The fiber matt had no resin, and it was sandwiched between polypropylene discs to hold its shape. Is was assembled using white PVC rods as spacers, and fixed with conductive epoxy.

31

Figure 2.17: Sample 007 sandwich carbon matt

•

Sample 008 was constructed from a solid carbon rod and carbon plate disks. The disks were cut with a hole saw out of a carbon plate obtained from the Composite Store (item #C4063, Tehachapi, California). It was assembled using white PVC rods as spacers, and fixed with conductive epoxy.

Figure 2.18: Sample 008 Carbon disks •

Sample 009 was constructed from a solid carbon rod with carbon plate laser cut into star configuration. The carbon plate was obtained from the Composite Store (item #C4063, Tehachapi, California). It was assembled using white PVC rods as spacers, and fixed with conductive epoxy.

Figure 2.19: Sample 009 Carbon plate star

32 •

Sample 010 was constructed from a solid carbon rod with carbon matt sandwiched between white PVC spacers, and fixed with conductive epoxy.

Figure 2.20: Sample 010 Carbon matt

ESP collection efficiency is typically found using the exponential Deutsch-Anderson Equation, given the particulates relative size and its density within the flow. The equation is given as: Eff = 1 – Exp [(-2ceL)/(Ub)] where U is the volumetric flow rate, ce is the electrical migration, L is the length of the precipitator, and b is the collection plate spacing (Friedlander, 2000). Specific parameters which affect collection efficiencies are defined below: a. Air velocity: The speed at which the particles are passed through the ESP, in conjunction with the aspect ratio of the electrodes and collection plates have a significant effect on collection efficiency. Today’s ESP designs tend to stay in the velocity range of 0.8 m/s – 1.2 m/s which lends to optimal performance. Problems are not seen in this velocity range when using 300-400mm plate spacing (Mastropietro, 1998). The air velocity was held constant for this testing at as close to 1.05 m/s as possible with equipment used.

33 b. Particulate feed rate: Using the Schenck Accurate feeder, and regulated compressed air at 20 psi, the fly ash was held at a constant 70% rate on the feeder’s control c. Applied Power: Using the voltage regulator from NWL Transformer, the DC voltage and current were held at their peak performance value during the collection tests. At the time of spark over, the regulator automatically ramps the voltage and amperage back up to the previous set value. Starting voltage and current values were recorded for each sample. d. Electrode resistance: The tests were performed on discharge electrodes of varying resistances. The resistance has never been a factor, as current industry electrodes are made from steel alloys which have conductivity far exceeding the requirements. However, the conductive composites have much lower conductivities which may reduce the performance of the electrode. e. Ambient Humidity: The humidity on any given day impacted the collection amounts greatly. The equipment and funding was not available to control this variable. To increase collection amounts, the fly ash was heated to lower humidity content. Note that comparison tests were run very close together to eliminate this variable.

34

Chapter 3: Testing 3.1 Introduction Particle collection experiments were first conducted for ten minutes for each electrode sample. Due to the large amount collected in the experiment, and particulate being lost, collection time was lowered to five minute runs for the setup. The particulate used was fly ash collected from the Gavin power plant in Cheshire, Ohio. It should be noted that the amount of humidity in the fly ash has a very high impact on the mass of particulate collected. The fly ash in the Gavin plant is at an elevated temperature during collection, keeping it relatively low in humidity level. For testing, the fly ash was heated to reduce moisture. During testing, it was found tests that were run on different days did not correlate to the exact same test that was run earlier; therefore, it was necessary to do all comparison testing as close to the same time as possible. The steel comparison electrode (sample 001) was run for a five minute test under the same conditions as the composite electrode to be sure there was an accurate, one to one comparison. This was done just prior to each test on the sample composite electrode. For more accurate tests, the lab environmental conditions would need to be controlled (temperature and humidity), as well as the condition of the fly ash with an emphasis put on humidity control. Testing was performed on the samples, comparing the voltage versus amperage curves and the particulate collection to that of the steel discharge electrode.

35 3.2 Results To summarize, each of the composite discharge electrodes which were tested against the base stainless steel discharge electrode did at least as good, or better in their respective collection efficiencies which is shown in figure 3.1.

Collection Comparison

Collection Amount (% over base)

160% Sample 001

140%

Sample 002

120%

Sample 003

100%

Sample 004

80%

Sample 005

60%

Sample 006 Sample 007

40%

Sample 008

20%

Sample 009

0%

Sample 010 Samples

Figure 3.1: Collection amount of each composite electrode against the SS base electrode (sample 001) which is taken as 100%.

Each composite electrode was tested against the base electrode one at a time, in as close to the same time as possible to ensure identical test conditions were present. The following pages discuss and show each individual comparison test which was conducted for the experiment. First, the base stainless steel electrode was put into the test chamber and the voltage was increased incrementally until it was large enough to create an electrical arc or

36 short between the electrode tip, and the closest grounded collection plate. The data in table 3.1 show the correlation between the voltage and amperage that the sample can conduct before arcing, or what is called “spark over”. The correlation is referred to as the “VI curve”. Each composite electrode which follows has the same method VI curve data as done here.

Table 3.1: Sample 001 VI curve test data Sample

001

Description Comparison electrode made entirely from 316 SS, rod and stars. Welded together.

VI Curve: Sample 001 VI Curve: Sample 008 1.6

Current (mA)

Current (mA)

VI 2.0 Curve Data Current Voltage 1.5 mA KVDC 1.0 0.12 10.08 0.12 10.92 0.5 0.16 12.60 0.24 18.48 0.0 0.98 0.0 40.32 1.42 47.88 1.78 50.40

1.4 1.2 1 0.8 0.6 0.4 0.2 0

10.0

0

5 20.0

10

15

30.0

20

25

40.0

30

Voltage (KVDC) Voltage (KVDC)

35

50.040

45

60.0 50

37 Table 3.2 shows the test data collected for sample 002. The resistivity of the Iten rod was 80.12 ohm-cm, and the collection efficiency over that of the base stainless steel electrode was 2.1%.

Table 3.2: Sample 002 VI curve and collection comparison data Sample

002

Description:

Resistivity =

80.1 ohm-cm

Pultruded conductive hollow rod made by Iten Industries. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. Mounting steel washer with welded nut also attached via conductive epoxy.

VI Curve: Sample 002

Current (mA)

VI 1.6 Curve Data Current Voltage 1.4 mA1.2 KVDC 1.0 0.12 9.24 0.8 0.12 9.24 0.6 0.16 10.92 0.4 0.2 0.24 16.80 0.0 0.94 38.64 0.0 5.0 1.40 47.04

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Voltage (KVDC)

mA @ Start 0.8 0.8

Collection Data Sample Sample

001 002

KVDC @ Start 37.0 36.1

Collection (g) 4.7 4.8

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 002

0.0

0.5

1.0

1.5

2.0

2.5 Amount (grams)

3.0

3.5

4.0

4.5

5.0

38 Table 3.3 shows the test data collected for sample 003. The resistivity of the Iten rod was 16.02 ohm-cm, and the collection efficiency over that of the base stainless steel electrode was 7.4%.

Table 3.3: Sample 003 VI curve and collection comparison data Sample

003

Description:

Resistivity =

16 ohm-cm

Pultruded conductive hollow rod made by Iten Industries. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. Mounting steel washer with welded nut also attached via conductive epoxy.

VI Curve: Sample 003

Current (mA)

VI 1.6 Curve Data Current Voltage 1.4 mA1.2 KVDC 1.0 0.12 10.08 0.8 0.12 10.08 0.6 0.16 11.76 0.4 0.2 0.22 17.64 0.0 0.94 39.48 1.40 0.0 47.045.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Voltage (KVDC)

Collection Data Sample Sample

mA @ Start 0.8 0.8

001 003

KVDC @ Start 37.0 36.1

Collection (g) 4.3 4.6

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 003

0.0

0.5

1.0

1.5

2.0

2.5 Amount (grams)

3.0

3.5

4.0

4.5

5.0

39 Table 3.4 shows the test data collected for sample 004. The resistivity of the Iten rod was 2.76 ohm-cm, and the collection efficiency over that of the base stainless steel electrode was 19.8%.

Table 3.4: Sample 004 VI curve and collection comparison data Sample

004

Description:

Resistivity =

2.8 ohm-cm

Pultruded conductive hollow rod made by Iten Industries. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. Mounting steel washer with welded nut also attached via conductive epoxy.

VI Curve: Sample 004

Current (mA)

VI Curve Data 1.6 Current Voltage 1.4 mA1.2 KVDC 1.0 0.12 9.24 0.8 0.12 9.24 0.6 0.16 10.92 0.4 0.2 0.22 15.96 0.0 0.92 38.64 5.0 1.40 0.0 43.68

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Voltage (KVDC)

Collection Data Sample Sample

mA @ Start 0.8 0.8

001 004

KVDC @ Start 37.8 32.8

Collection (g) 3.9 4.8

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 004

0.0

0.5

1.0

1.5

2.0

2.5 Amount (grams)

3.0

3.5

4.0

4.5

5.0

40 Table 3.5 shows the test data collected for sample 005. The resistivity of the Iten rod was 96.37 ohm-cm, and the collection efficiency over that of the base stainless steel electrode was 30.5%.

Table 3.5: Sample 005 VI curve and collection comparison data Sample

005

Description:

Resistivity =

96.4 ohm-cm

Pultruded conductive hollow rod made by Iten Industries. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. Mounting steel washer with welded nut also attached via conductive epoxy.

VI Curve: Sample 005

Current (mA)

VI Curve Data 1.6 Current Voltage 1.4 mA1.2 KVDC 1.0 0.12 11.59 0.8 0.20 16.46 0.6 0.4 0.74 37.30 0.2 1.14 46.45 0.0 1.40 0.0 49.56

10.0

20.0

30.0

40.0

50.0

60.0

Voltage (KVDC)

mA @ Start 0.8 0.8

Collection Data Sample Sample

001 005

KVDC @ Start 37.8 38.5

Collection (g) 2.6 3.7

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 005

0.0

0.5

1.0

1.5

2.0

2.5 Amount (grams)

3.0

3.5

4.0

4.5

5.0

41 Table 3.6 shows the test data collected for sample 006. The resistivity of the Iten rod was 1176.1 ohm-cm, and the collection efficiency was equal to that of the base stainless steel electrode.

Table 3.6: Sample 006 VI curve and collection comparison data Sample

006

Description:

Resistivity =

1176.1

ohm-cm

Pultruded conductive hollow rod made by Iten Industries. Stainless steel 316 stars fixed to the rod by light press fit, and conductive epoxy. Mounting steel washer with welded nut also attached via conductive epoxy.

VI Curve: Sample 006

Current (mA)

VI 1.4 Curve Data Current Voltage 1.2 mA1.0 KVDC 0.12 10.08 0.8 0.6 0.12 11.76 0.4 0.22 17.30 0.2 0.83 37.80 0.0 1.26 43.68 0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Voltage (KVDC)

Collection Data Sample Sample

mA @ Start 0.8 0.8

001 006

KVDC @ Start 38.6 38.1

Collection (g) 3.5 3.5

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 006

0.0

0.5

1.0

1.5

2.0

2.5 Amount (grams)

3.0

3.5

4.0

4.5

5.0

42 Table 3.7 shows the test data collected for sample 007. The collection efficiency over that of the base stainless steel electrode was 29.5%.

Table 3.7: Sample 007 VI curve and collection comparison data Sample

007

Description: Solid carbon rod with carbon fiber matt (no resin) sandwiched between polypropylene rings used as discharge tips. Assembled using white PVC rods as spacers, and fixed with conductive epoxy.

VI Curve: Sample 007

Current (mA)

VI 2.0 Curve Data Current Voltage mA1.5 KVDC 0.12 9.24 1.0 0.12 9.24 0.14 10.08 0.5 0.24 15.12 0.0 1.00 36.96 5.0 1.60 0.0 46.20

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Voltage (KVDC)

Collection Data Sample Sample

001 007

mA @ Start 0.8 0.9

KVDC @ Start 37.0 33.6

Collection (g) 5.6 8.0

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 007

0.0

1.0

2.0

3.0

4.0 Amount (grams)

5.0

6.0

7.0

8.0

43 Table 3.8 shows the test data collected for sample 008. The collection efficiency over that of the base stainless steel electrode was 3.8%.

Table 3.8: Sample 008 VI curve and collection comparison data Sample

008

Description: Solid carbon rod with carbon plate disk. Assembled using white PVC rods as spacers, and fixed with conductive epoxy.

VI Curve: Sample 008

Current (mA)

VI 2.5 Curve Data Current Voltage 2.0 mA KVDC 1.5 0.12 10.08 0.12 10.08 1.0 0.14 10.92 0.5 0.24 16.80 0.0 0.96 41.16 0.0 1.38 47.88 2.00 53.76 Collection Data Sample Sample

001 008

10.0

20.0

30.0

40.0

50.0

60.0

Voltage (KVDC)

mA @ Start 0.8 0.9

KVDC @ Start 37.0 37.8

Collection (g) 6.4 6.7

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 008

0.0

1.0

2.0

3.0

4.0 Amount (grams)

5.0

6.0

7.0

8.0

44 Table 3.9 shows the test data collected for sample 009. The collection efficiency over that of the base stainless steel electrode was 38.0%.

Table 3.9: Sample 009 VI curve and collection comparison data Sample

009

Description: Solid carbon rod with carbon plate laser cut into star configuration. Assembled using white PVC rods as spacers, and fixed with conductive epoxy.

VI Curve: Sample 009

Current (mA)

VI Curve Data 2.0 Current Voltage mA1.5 KVDC 0.12 9.24 1.0 0.12 9.24 0.16 10.92 0.5 0.24 15.96 0.0 0.97 38.64 0.0 1.44 46.20 1.78 50.40 Collection Data Sample Sample

001 009

10.0

20.0

30.0

40.0

50.0

60.0

Voltage (KVDC)

mA @ Start 0.8 0.8

KVDC @ Start 37.0 34.4

Collection (g) 4.7 7.6

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 009

0.0

1.0

2.0

3.0

4.0 Amount (grams)

5.0

6.0

7.0

8.0

45 Table 3.10 shows the test data collected for sample 010. The collection efficiency over that of the base stainless steel electrode was 32.5%.

Table 3.10: Sample 010 VI curve and collection comparison data Sample

010

Description: Solid carbon rod with carbon matt sandwiched between white PVC spacers, and fixed with conductive epoxy.

VI Curve: Sample 010

Current (mA)

VI 1.2 Curve Data Current Voltage 1.0 mA KVDC 0.8 0.12 8.40 0.6 0.12 9.24 0.4 0.14 10.08 0.2 0.24 14.28 1.02 32.76 0.0 1.040.0 33.60 5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

Voltage (KVDC)

mA @ Start 0.8 0.9

Collection Data Sample Sample

001 010

KVDC @ Start 37.0 30.2

Collection (g) 5.6 8.3

Note: Collection amount +/- 0.3 g

Collection Comparison

Sample 001 (Base) Sample 010

0.0

1.0

2.0

3.0

4.0

5.0

Amount (grams)

6.0

7.0

8.0

46

Figure 3.2 shows and overlay of the VI curves for each of the composite electrodes, including the base stainless steel electrode for comparison.

Voltage vs Amps (curve) Sample 001 Sample 002

2.5

Sample 003 Sample 004

2.0

Amp

Sample 005 1.5

Sample 006 Sample 007

1.0

Sample 008 Sample 009

0.5

Sample 010

0.0 0.0

10.0

20.0

30.0

40.0

50.0

60.0

KVDC

Figure 3.2: VI Curves for all electrodes tested

The data reflects that within the given experimental parameters done here, each of the sample electrodes performed at least as well as or better than the standard stainless steel discharge electrode. Each of the electrode samples were run more than three trials each. Due to the many number of previous trial runs necessary to correct the testing method, only the last trail was recorded in the given experimental data. The collection amount over the three trails had an average variance of +/- 0.3 grams. The measured data for these experiments did not reflect any direct correlation between the VI curve, and the collection efficiency. Collection efficiency will be affected by the VI characteristics, but

47 this information was not a request by the funding partners and therefore sufficient data was not collected to determine the exact correlation between the collection efficiency and the VI characteristics of the samples. To compare data with theoretical collection efficiency, more precise loading data (particulate concentration) would need to be known. This data was also not requested, and not within budgetary limits; and therefore not done.

48

Chapter 4: Discussion and Future Work 4.1 Introduction Testing with the new apparatus shows that even testing on less humid days was not fully representative of previous tests. The fly ash humidity had a significant effect on the collection amount, and was verified by repeating the same tests on different days. Holding all other conditions the same, different amounts were collected from the same electrodes on different dates. The characteristics of the dust vary, and have a large impact on the results; therefore, final testing was done as a comparison test only, and were conducted at as close to the same time as possible, to eliminate environmental differences between tests and differences in the fly ash being used day to day. 4.2 Conclusion This testing verifies that conductive polymer composites could be used to replace current steel discharge electrodes in a commercial electrostatic precipitator application. Composite electrodes that were adequately conductive, and more corrosion resistant than the current steel electrode were designed and tested. 4.3 Future Work The results here should only be considered as a strong positive basis for the concept of introducing conductive composites for use as the construction material for discharge electrodes in commercial ESP’s. The next step should be constructing a small scale number of commercial size electrodes for implementation into a pilot level ESP for long term study.

49 The results for carbon fiber composite discs and carbon fiber mats as discharge electrodes were particularly impressive. These materials should be tested further.

50

References Alam, M.K., Chatterjee, A., and Morosko, J. Extruded CNF composites with HDPE and Glass Fibers. Submitted for publication. 2006. Bayless, David J., Alam, Khairul M., Radcliff, Roger., Caine, John. (2004) Membranebased Wet Electrostatic Precipitation. Fuel Processing Technology, 85. 781-798. Corona discharge. (n.d.). Wikipedia. Retrieved January 16, 2007, from Answers.com Web site: http://www.answers.com/topic/corona-discharge Dayley, Matt and Holbert, Keith. (2003) Electrical Precipitators for Power Plants. Retrieved December 14, 2006, from http://www.eas.asu.edu/~holbert/wise/electrostaticprecip.html. Environmental Works. (2006). In Encyclopedia Britannica. Retrieved December 13, 2006, from http://www.britannica.com/eb/article-214285. Friedlander, S.K. (2001) Smoke, Dust and Haze. New York. John Wiley & Sons. 83. Hamon Thermal Transfer. (2006). Precipitator Components. Retrieved December 14, 2006, from http://www.hamon-thermaltransfer.com/Products_EspComponents.asp. Mastropietro, R.A. (1998). Practical Problems with Electrostatic Precipitators can provide Significant Contributions to Science. Presented at the 7th International Conference on Electrostatic Precipitation. September 20-25, 1998, Kyongju, Korea. Retrieved August 1, 2006, from http://hamonresearchcottrell.com/HRCTechnicalLibrary/Practical%20Problems%20with%20Elect rostatic%20Precipitators%20can%20prov.pdf.

51 Southern Environmental. (2006a). Air Pollution Control Systems. Retrieved July 31, 2006, from http://www.southernenvironmental.com/_pdf/APCS.pdf. Southern Environmental. (2006b). Electrostatic Precipitators. Retrieved July 31, 2006, from http://www.southernenvironmental.com/_pdf/ESP.pdf. Southern Environmental. (2006c). Membrane WESP. Retrieved July 31, 2006, from http://www.southernenvironmental.com/_pdf/04_MACT_MEMBRANE_REPORT.p df. Southern Environmental. (2006d). SEI-Elex RS Discharge Electrodes. Retrieved July 31, 2006, from http://www.southernenvironmental.com/_pdf/ELEX.pdf. White, Harry James. (1963). Industrial Electrostatic Precipitation. Reading, Massachusetts. Addison – Wesley Publishing Co., Inc. 01-87.

52

Appendix A

VI Curve Data Sample 001 Current Voltage mA KVDC 0.12 10.08 0.12 10.92 0.16 12.60 0.24 18.48 0.98 40.32 1.42 47.88 1.78 50.40

VI Curve Data Sample 002 Current Voltage mA KVDC 0.12 9.24 0.12 9.24 0.16 10.92 0.24 16.80 0.94 38.64 1.40 47.04

VI Curve Data Sample 003 Current Voltage mA KVDC 0.12 10.08 0.12 10.08 0.16 11.76 0.22 17.64 0.94 39.48 1.40 47.04

53

VI Curve Data Sample 004 Current Voltage mA KVDC 0.12 9.24 0.12 9.24 0.16 10.92 0.22 15.96 0.92 38.64 1.40 43.68

VI Curve Data Sample 005 Current Voltage mA KVDC 0.12 11.59 0.20 16.46 0.74 37.30 1.14 46.45 1.40 49.56

VI Curve Data Sample 006 Current Voltage mA KVDC 0.12 10.08 0.12 11.76 0.22 17.30 0.83 37.80 1.26 43.68

VI Curve Data Sample 007 Current Voltage mA KVDC 0.12 9.24 0.12 9.24 0.14 10.08 0.24 15.12 1.00 36.96 1.60 46.20

54

VI Curve Data Sample 008 Current Voltage mA KVDC 0.12 10.08 0.12 10.08 0.14 10.92 0.24 16.80 0.96 41.16 1.38 47.88 2.00 53.76

VI Curve Data Sample 009 Current Voltage mA KVDC 0.12 9.24 0.12 9.24 0.16 10.92 0.24 15.96 0.97 38.64 1.44 46.20 1.78 50.40

VI Curve Data Sample 010 Current Voltage mA KVDC 0.12 8.40 0.12 9.24 0.14 10.08 0.24 14.28 1.02 32.76 1.04 33.60