Increase Mercury Oxidation And Scrubber Capture Biology Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The taconite industry located in Minnesota produces taconite pellets which are currently among the main source of iron for the iron and steel industries all over the world. Minnesota's taconite industry is located on the Mesabi Iron Range in the north east of the state of Minnesota. Minnesota's existing taconite processing plants were built during the 1950's to 1970's well before mercury was recognized as a global pollutant. (1)

Taconite pellet production process is divided into three major steps. The first step is mining of the ore from the open pits. In the second step, beneficiation is performed on the ore to increase the iron content and to improve the ore's physical structure. The third step involves agglomeration which includes pelletization (indurating) processes to oxidize the iron present in the ore. In this part the iron-rich concentrated ore is mixed with water and a binder and then the concentrate is rolled into green balls inside rotating cylinders. These green balls are then fed to the induration furnaces (Straight Grate or Grate Kiln) with the help of a moving shaft in which they are heated up to 2300 - 2500 °F. (2) Taconite ore has significant quantities of mercury deposited on it. Due to extensive research in this area, it was only recently recognized that mercury present in taconite concentrate is released during induration to process gases. Also, the majority of this mercury is not captured by the plant's wet scrubbers, but released to the atmosphere. (2)

Previous studies related to mercury release suggest that taconite processing in Minnesota releases approximately 350 to 400 kg (750 - 900 lbs) of mercury to the atmosphere each year. (3)(3, 3) Even though this amount is small compared to overall global emission rates, it is Minnesota's second largest industrial source of mercury to the atmosphere. Hence, The Environmental Protection Agency (EPA) regulates the mercury emission from taconite plants. There are a number of methods to remove this mercury from a flue gas stream. Among the most common methods to control mercury emissions is to inject halogens into the taconite processing system to increase oxidation of mercury in the induration furnace and also to promote the capture of mercury in the scrubbers. Short term tests were conducted at different taconite facilities by the Minnesota Department of Natural Resources (DNR) to identify potential means to reduce mercury oxidation. These tests included addition of halogen additives such as chlorides and bromides to the green ball feed. (1)

Depending on chloride application rate in a straight-grate facility, NaCl addition decreased total mercury (Hg (T)) emission in stack-gases by 5 to 9%. 6% - 13% reduction is observed with addition of NaCl and salt solutions directly into the pre-heat zone of straight grate furnace. 18% - 32% reductions of Hg (T) emission was observed in grate-kiln facility when NaCl addition rates were kept similar to that of the straight grate facility. Compared to chloride salts, bromide salts proved to be more effective with a reduction of 62-64%. The maximum reduction (80% capture) was observed with proprietary U.S. Environmental Protection Agency (EPA) oxidant when it was added to scrubber solution. Each of these methods provides some level of mercury capture, but not at a high level (>95%).(1)

Even though these additives have proved to be effective in reducing mercury emissions, they also pose a problem on the taconite facilities. There is a possibility of increased corrosion in the system due to the additives and an increase in particulate matter emissions due to additional fine particulate formation. (4) Hence, in this project, our strategy was to minimize the amount of bromine required for mercury oxidation and in turn, try to reduce the imposed problem of corrosion. An alternative oxidizing agent that is not corrosive is used in this project. Also, we have investigated the effectiveness of a special additive in achieving oxidation of the mercury in two steps as follows.

1. Preliminary Testing (Phase I)

In this section, research focuses on preliminary laboratory scale work performed to determine if the technology has a potential of oxidizing mercury significantly when included in the green ball formation process.

The work in phase I investigated:

Optimal additive to green ball ratio

Additive and green ball combination method - through mixing or surface addition

Effectiveness of halogen enhanced carbon against plain carbon

Surface chemistry of green balls during testing.

Green ball samples used during this testing were obtained in January, 2012 for UTac, and October, 2011 and February, 2012 for MinnTac. The green balls were tested over a period of 4 to 5 months, during which the test equipment was optimized continuously.

2. Analysis of Mercury Oxidation Potential of ESORB-HG-11 (Phase II)

This section focuses on laboratory scale work performed to establish the extent of oxidation achievable when ESORB-HG-11 was included in the formulation of green balls obtained from all five taconite facilities. ESORB-HG-11 loadings of 0.1 weight percent (wt%) and 0.5 weight percent(wt%) were used for the duration of the test, based on the optimum loading established during the phase 1 testing. The main goals of these tests were to:

Establish potential oxidation levels achievable by including ESORB-HG-11 in green ball formulations.

Perform chemical analyses on test products to better understand the mechanism of mercury oxidation.

Green balls used for the testing were prepared by the Coleraine Minerals Research Laboratory (CMRL). Preparation was done according to a batch balling procedure established by CMRL and based on the green ball formulations of each respective facility.

CHAPTER II

BACKGROUND

2.1 Taconite Industry Mining and Processing

Taconite is an iron ore which contains 25 to 30 percent iron minerals which is inter- layered with quartz, silica or carbonate. When taconite is heated in the presence of a reductant, it yields metallic iron (Fe). It primarily contains iron oxides, mainly magnetite (Fe3O4 - iron content 72 percent), hematite (Fe2O3 iron content70 percent), and goethite (Fe2O3.H2O- iron content 63 percent).

2.1.1 Mining

In Minnesota's Mesabi Iron Range, taconite is mined from open pits because most commercial ore bodies lie close to the surface of mines and their lateral dimensions are large. Mining activities at these sites involve overburden removal, drilling, blasting, and removal of waste rock and crude taconite from the open-pit. (2) Mining in open pits is mostly done with large powerful shovels and trucks. Shovels at taconite mines are used to dig surface overburden as well as iron ore and waste rock. Rotary drills are used to create holes which are 16 inches in diameter and 45 to 55 feet in depth for explosives to be placed for blasting activities. 0.4 to 1.5 million tons of taconite ore is broken during individual blasts. Trucks then transport the crude iron ore to the primary crushers. At some mining operations, trains are used to transport ore to the crushers. (2, 5, 6)

2.1.2 Beneficiation

The beneficiation process increases the iron content by reducing the impurities in the ore and it also improves the physical structure of the ore. (7) The process includes milling (crushing and grinding), screening, washing, and processes that separate ore minerals from gangue (sand, rock, and other impurities surrounding the iron) by using differences in physical or chemical properties. Figure 1 illustrates the general beneficiation processes. (2, 6)

Figure : Flow sheet of concentrating section for taconite plant. (2)

2.1.3 Crushing and Grinding

Crushing and grinding of the ore is an important step to produce acceptable concentrates from crude taconite ore. In the first step of crushing and grinding, taconite ore is fed to a gyratory crusher. In the crusher, ore is crushed down to a size of approximately 6 inches. Secondary and tertiary fine-crushing stages are used to reduce the material to 3/4 inch. To remove the undersized material, there are a few intermediate vibratory screens between the crushing stages.

After crushing, the crushed ore is sent to rod mills for fine grinding. Product from rod mill will go to ball or pebble mills which are charged with heavy steel rods or balls and taconite ore with water slurry. The discharged taconite slurry from ball mill will be fed to the magnetic separator. (2)

2.1.4 Magnetic Separation

Magnetic separation involves three stages of separation. The first stage is called cobbing, which is followed by cleaning or roughing of the ore and the final step is called as finishing. Each stage works on a finer particle size as compared to the previous ones by removing the oversized particles. Rejected oversized particles are sent to non magnetic tailings or gangue. Generally, 40 percent of the feed is rejected to non-magnetic tailings. Tailings from these two stages are sometimes re-ground or discharged to the tailing basin. (2), (8, 9)

2.1.5 Flotation

In the flotation process, excess water is removed from the iron-bearing slurry through gravity separation in a hydraulic concentrator. This is followed by a chemical flotation unit. In the flotation process, three types are additives are used to increase the iron contents namely frothers, collectors/amines, and anifoams. After this step, the iron-rich concentrates become the raw materials for producing taconite pellets in the agglomerating process.(2, 10)

2.1.6 Agglomeration

Agglomeration is the third and the most important step in taconite pellet production since in this part the iron-rich concentrated ore is mixed with water and a binder (generally some mixture of bentonite, hydrated lime and/or organic material) and then the concentrate is rolled into green balls inside rotating cylinders. These green balls are then fed to the induration furnaces with the help of a moving shaft in which they are heated up to 2300 - 2500 °F. The induration or heating of the green balls can be done in a vertical shaft furnace on a travel grate (straight grate) or by a combination of a travel grate and a rotary kiln (grate-kiln). (2, 4) The finished product is taconite pellets. Figure 2 explains the pelletizing process in detail.

Figure : Flow sheet of pelletizing section for taconite plant (2)

Travel Grate (Straight Grate). As shown in Figure 3, the green balls are fed to the updraft drying section of straight grate with the help of a moving shaft. In drying and preheat sections, the green balls are dried and preheated after which they are fed to the ignition section of the grate, where all the magnetite is oxidized to hematite. Finally, the pellets are cooled by intake air at cooling stages before they are discharged by conveyor belt to storage.

Figure : Straight Grate Furnace

Grate-Kiln. The grate-kiln system combines a travel grate, a rotary kiln, and an annular cooler (see Figure 3). Drying of the green pellets and partial induration occur at the grate while final induration is finished in the rotary kiln. The pellets are heated to a temperature of 2,000°F on the travel grate before being hardened in the rotary kiln furnace. Then the hardened pellets enter the cooling zone of the annular cooler.(2)

Figure : Grate Kiln

2.2 Mercury Release in Taconite Processing

To develop effective control measures for mercury emissions from taconite industry, it is important to understand the release of mercury during taconite processing. Previous study at taconite plants proved that the major source of mercury is taconie ore and not the fuel (coal) used in the induration furnaces. (1, 11, 12) Mercury release and transport during taconite processing involves a relatively complex series of reactions, whereby some of the mercury released at high temperatures in the furnace is recaptured by magnetite and/or magnetite solid-solutions with maghemite (magnetite/maghemite solid-solutions). In all plants, however, there is also mercury captured by scrubber systems that is dissolved in solution, indicating potential importance of a molecular reaction between mercury and gaseous species, most likely Cl. To simplify the release process, we write four reactions:

2Fe3O4(ss) + ½ O2(g) = 3Fe2O3(ss) (I)

Magnetite Maghemite

2Fe3O4 + ½ O2(g) = 3Fe2O3 (II)

Magnetite Hematite

HgO(ss) = Hg0(g) + 1/2O2(g) (III)

HgO(ss) + 2HCl(g) = HgCl2(g) + H2O(g) (IV)

Reactions (I) and (II) represent the relative formation of magnetite/maghemite solid-solutions and hematite, while Reactions (III) and (IV) represent release of mercury in reduced and oxidized form, respectively. In reaction (I), Magnetite is getting oxidized to give maghemite solid solution; in which maghemite interacts with mercury in flue gases, while magnetite does not. The minerals have the same structure and form a solid solution but little is known about how mercury reacts with magnetite solid-solutions.

For reaction (II), when magnetite is converted to hematite in induration furnaces, Mercury is released. Hematite does not interact with mercury in flue gases. Reaction (III) is important because Hg0(g) is insoluble in water and cannot be caught by wet scrubbers. HgO(ss) represents mercury associated with magnetite and magnetite/maghemite solid-solutions. Reaction (IV) determines the formation of HgCl2(g) from HgO(ss) in which HgCl2(g) is soluble in water and the Hg2+ base atom can adsorb to solids. Oxidized mercury is more easily captured by wet scrubbers than is Hg0(g).(12)

The oxidation reaction of magnetite holds utmost important in mercury release since it determines the nature and composition of the dust in process gases. This dust will ultimately help to trap the oxidized mercury in process gases. Zygarlicke (2003) and Galbreath et al. (2005) have demonstrated that magnetite and hematite does not participate in gaseous mercury reactions. During the formation of maghemite, oxygen is added to the spinel-type crystal lattice without any modification.(13),(14)

Data presented by Berndt et al. (2005) from the onsite testing demonstrate that magnetite/maghemite composition is close to the original composition of magnetite. Hence, there is a high probability that magnetite/maghemite solid solution interacts with mercury even for a low level of maghemitization. To understand this behavior, let's have a closer look at the mineral reactions in preheat and firing zone. (Refer to figure 5.) When oxygen atom comes in contact with the magnetite surface, it reacts with the electrons from Fe+2 and forms Fe+3 and O-2 ions. This will effect in extending the mineral lattice and a cation vacancy will develop.

This oxidation reaction takes place outside inwards. Hence, the full oxidation of the interior portion depends on extend of diffusion. There are a number of factors which affect this diffusion which includes - oxygen availability, temperature, humidity, nucleation effects, as well as crystal orientation. (15). According to the literature, conversion of magnetite to magnetite/maghemite solid-solutions takes place starting from 400 to 500 °C in a very short span of time. Hence, only the outer most surface gets converted to magnetite/maghemite solid-solutions. In the kiln, around 1200 to 1300 °C complete conversion of magnetite to hematite takes place. Hematite is not a significant oxidant for Hg0 in flue gases. Hence reaction (II) might limit the mercury oxidation and capture process.

Figure : Mineral Reactions in Preheat and Firing Zones

Figure : Mercury Release in Preheat and Firing Zones

In reaction 3, oxidized mercury reduces to a volatile form of Hg0 (g). Previous studies have proved that Hg0 (g) is dispersed throughout the green ball composition. Also, it is proved that the elemental mercury also exists on the surface of magnetite/maghemite solid-solutions in the cooler regions where the hematite formation reaction is not begun. Hence, it is safe to say that mercury evolves from the surface of magnetite/maghemite solid-solutions to the process gases which have a further impact onto the scrubber systems. Reaction (IV) is a hypothetical mechanism to generate HgCl20(g). It is a molecule which is easily absorbed by scrubbers in taconite facilities. Relative rates of reaction (III) and reaction (IV) will determine the overall emission or capture of mercury in stack gases.(1, 11, 12) Chemistry of mercury reactions gets affected with the presence of HCl(g) in process gases, which will favor reaction(IV) over (III) and thus will give good capture efficiencies. Hence, combining all the four reactions, we get:

Hg0(g) + 3Fe2O3(ss) + 2HCl(g) = 2Fe3O4(ss) + HgCl2(g) + H2O(g) .

Maghemite Magnetite

This reaction shows that if components on the left hand side of the reaction are elevated (maghemite and HCl), it will favor generation of HgCl2(g) relative to Hg0(g) and, thus, greater capture efficiency.

It's important to note that, previous studies have shown that addition of halogens to taconite plants increase corrosion of the equipment and hence, this research will focus on finding a low corrosion method to achieve mercury oxidation.

2.3 Mercury Cycle

Mercury (Hg) is a naturally occurring chemical element that is found in air, water, and soil. It exists in several forms: Mercury circulates through the environment in different chemical forms and different physical states. Mercury can exist in the environment in three forms: elemental mercury (Hg0), oxidized mercury (mercurous [Hg22+] or mercuric [Hg2+]), and particulate-bound mercury (HgP). (16),(6),(17) Much of the mercury released into the atmosphere is in the form of elemental mercury. Elemental mercury can persist in the atmosphere for up to two years and travel thousands of miles, thus creating a global issue. Most of the oxidized and particulate-bound mercury will deposit in nearby water and soils, thus creating a local or regional issue. Bacteria can convert all forms of mercury to organic mercury, namely methyl mercury (CH3Hg+), most efficiently in the aquatic food chain. Once methyl mercury enters water, it can bio-accumulate in fish and other aquatic animals. Humans are primarily exposed to mercury through the consumption of fish and other aquatic animals that come from contaminated lakes and streams.(16)

2.4 Health Effects

Mercury is a neurotoxin and long term exposure can lead to permanent damage of the brain, kidneys, and developing fetuses. (6, 18) In 2000, National Research Council declared that the EPA reference dose of 0.1 ug/kg of mercury intake is scientifically justifiable and this limit protect against the neurological effects of mercury exposure. (19) In the past, children born from women exposed to higher amounts of mercury during pregnancy have shown a variety of neurological abnormalities. Effect of methyl mercury exposure was noticed in Minamata, Japan, where 1000 deaths occurred and an additional 17,000 people were affected by methyl mercury exposure. Adverse effects on children born to these women included cerebellar symptoms, dysarthria, mental retardation, retention of primitive reflexes, hyperkinesia, hypersalivation, strabismus, and pyramidal symptoms. Another incident in Iraq where women mistakenly consumed bread made from methyl mercury treated wheat resulted in 500 deaths. Children born to these women showed delays in speech and motor development, mental retardation, reflex abnormalities, and seizures.(18, 20)

Dietary intake of methyl mercury was also associated to increased risk of coronary heart diseases and cardiovascular diseases. In Amazonian women, a significant decrease in vision, manual dexterity, and muscular strength was found with an increase in hair mercury levels. (21) A recent study found that exposure to methyl mercury near electric generating facilities is correlated with mental retardation in thousands of American babies each year. Environmentally released mercury has also been shown to increase rates of special education services and autism. (17)

2.5 Existing and Prospective Federal Regulation

Mercury pollution poses a problem to human health and environmental risks. Although, mercury is naturally present in the environment after industrial revolution, human activities have increased the amount of mercury cycling among land, ocean and atmosphere. Mercury is generally emitted in elemental form and gets converted to methylmercury in aquatic system and enters the food chain. Mercury deposited in fish tissue in now the leading cause of advisories issued for fish consumption in Minnesota lakes. In 1997, US legislation has mandated emission regulations for coal-fired power plants, previously identified as the largest anthropogenic emitter of mercury to the atmosphere. Decreased emission from this source and other sources in Minnesota led to reductions goals for mercury emission from 1990 levels by 60% in 2000 and 70% in 2005. However, the decreases in percentage of mercury emissions from coal fired power plants have resulted in an increase in the proportion for industries where control measures are either not available or difficult to implement. Taconite is one such industry whose share increased from 16% in 1995 to 20% in 2000. Important timeline for reduction of mercury emission from these taconite plants include 2007 and 2010 when mercury limits were set for Great Lakes Basin and Minnesota, respectively. (3)

In 1999, due to LaMP reduction schedule and the requirements of Clean Water Act, Minnesota's legislature developed a plan for attaining the reduction requirements. The proceedings required the taconite industry to reduce emissions to 210 lbs by 2025 which accounts for 75 percent reduction in total. In an attempt to achieve these reduction targets, Minnesota Department of Natural Resources (DNR) and others have funded the research for identifying the control technologies capable of achieving 75% reduction.

Referring to Title 40, Part 63, 63.8980 for taconite industry from EPA website:

(a) For existing ore pretreatment processes, you must emit no more than 127 pounds of mercury per million tons of ore processed.

(b) For existing carbon processes with mercury retorts, you must emit no more than 2.2 pounds of mercury per ton of concentrate processed.

(c) For existing carbon processes without mercury retorts, you must emit no more than 0.17 pounds of mercury per ton of concentrate processed.

(d) For existing non-carbon concentrate processes, you must emit no more than 0.2 pounds of mercury per ton of concentrate processed.

(e) For new ore pretreatment processes, you must emit no more than 84 pounds of mercury per million tons of ore processed.

(f) For new carbon processes with mercury retorts, you must emit no more than 0.8 pounds of mercury per ton of concentrate processed.

(g) For new carbon processes without mercury retorts, you must emit no more than 0.14 pounds of mercury per ton of concentrate processed.

(h) For new non-carbon concentrate processes, you must emit no more than 0.1 pounds of mercury per ton of concentrate processed.

(i) The standards set forth in this section apply at all times.

2.6 Control Technologies

Existing technologies employed at taconite facilities capture a small percentage of mercury. Most of the taconite plants have wet scrubbers which are effective in capturing oxidized mercury (Hg2+) but not elemental mercury (Hg0). Wet scrubbers capture 10 to 40 percent of mercury in the taconite facilities. (13, 22) A study conducted by Minnesota Department of Natural Resources (DNR) evaluated potential approaches for mercury reduction as follows:

Injection of mercury sorbents into the gas stream.

Use of fixed bed sorbent reactors to oxidize a higher percentage of the mercury

Use of chemical oxidants to the gas stream, such as chloride and bromide salts or hydrogen peroxide.

Use of halogenated oxidants in conjunction with activated carbon injection. (23)

In previous studies, they have found a significant number of possible approaches to control mercury emissions but there is no single best technology that can be broadly applied to taconite industries. Hence, a standard technology would be very difficult to implement worldwide. On the basis of current developments, the remediation costs for mercury ranges from $2500 to $1.1 million per kg of mercury isolated from the environment, generally making mercury control a better option. (24) Hence, policy makers and industry show growing interest in multi-pollutant removal to achieve environmental quality and reduction cost.

In the literature, a variety of potential mercury oxidation catalysts have been investigated which includes gold, palladium, Iron oxides etc. Gold was found to be extremely useful since it absorbs mercury and chlorine and does not adsorb other species like nitric oxide, sulfur oxide and water.(25) (26) Palladium was found to be a good oxidizer since it has oxidized >95 percent of elemental mercury in pilot scale tests. Iron oxides (Fe2O3and Fe3O4) have been shown to promote mercury oxidation (27)(28). Al2O3 and TiO2 have been shown to oxidize 50-60 percent of mercury in pilot scale tests. Other metal catalysts shown to promote mercury oxidation include iridium (29), MnO2 (30), and CuO (30).

Olson et al. studied the detailed chemistry mechanism of mercury oxidation and it's binding on activated carbon in the stream of flue gas. Olson has also studied the effect of carbon sorbents and their performance with different sorbent properties, process conditions, and other flue gas constituents.(31) In coal fired power plants, carbon has proved to be a good additive to control mercury emissions.(32) Although, the conditions in a coal fired power plant and a taconite plant are entirely different, mercury behavior and oxidation properties remain same. Similarly, as compared to other catalysts used in oxidation reaction carbon is inexpensive and will not interfere with the taconite industry process. Hence, we conclude that carbon will be a good additive for mercury oxidation and capture.

2.7 Previous work in the Taconite Industry

The Department of Natural Resources (DNR) conducted a study to understand the mercury release in taconite processing and also summarized the result of the research. DNR's study on scrubber waters in taconite plants showed that mercury is present in two forms viz. dissolved and particulate bound. By applying the technology which captures mercury in process gases, we could reduce the dissolved mercury percent but particulate bound mercury values vary over time.(3) Berndt et al. found that there is a correlation between capture rate of mercury in wet scrubbers to the rate at which HCl and scrubber dust were generated during induration.(1, 11, 12) Thermal mercury release experiments conducted by Benner and Galbreath, spectroscopic measurements for heated taconite pellets suggest that mercury release during taconite induration is rather a complex process. (33),(13)

Also, it is a proved fact that structural conversion of magnetite to hematite generates maghemite solid solution and is closely tied to release of mercury. Mercury gets released in either as Hg0 or HgCl2, depending on availability of HCl in the process gas environment. This in turn affects its capture since oxidized form gets captured easily. Literature shows that 10 - 15 percent of mercury was captured in a straight grate kiln while approximately 30 percent was captured in grate kiln. (22) Berndt et al. also found that mercury gets adsorbs onto non-magnetic surfaces easily. (1, 11, 12)Hence, we need to promote the oxidation of mercury which in turn will get captured in scrubber waters and can be discarded with mercury adsorption and magnetic separation processes. In the literature, there are two ways in promote mercury oxidation viz addition of halogens (Cl, Cl2, Br2, HCl) and addition of oxidizing compounds to scrubber waters. (24, 33, 34) Other methods like ozone or activated carbon injection to process gases may also have application to the taconite industry. (16) Although, mercury oxidation and capture studies have been conducted at similar types of facilities such as waste incinerators, gold mining facilities, coal-fired power plant, the taconite industry in unique in a sense that induration includes formation of iron oxides which takes part in mercury release reactions and also plays a vital role in mercury transport.(16)

2.8 Surface Chemistry (SEM, XRD etc.)

CHAPTER III

PRELIMINARY TESTING (PHASE I)

Phase I testing is divided into two major sections as follows.

I. Carbon addition in the flue gas

The preliminary testing was carried out in a laboratory scale reactor to optimize the equipment required for carbon addition to the flue gas.

II. Carbon addition to the green balls

In this section, research was focused on the preliminary laboratory scale work. This was done to determine if the technology has a potential of oxidizing mercury significantly when included in the green ball formation process.

The work in phase I investigated:

Optimal additive to green ball ratio

Additive and green ball combination method - through mixing or surface addition

Effectiveness of halogen enhanced carbon against plain carbon

Surface chemistry of green balls during testing.

Green ball samples used during this testing were obtained in January, 2012 for UTac, and October, 2011 and February, 2012 for MinnTac. The green balls were tested over a period of 4 to 5 months, during which the test equipment was optimized continuously.

3.1 Carbon addition in the flue gas

3.1.1 Experimental Setup

Experimental setup for carbon addition in the flue gas includes setting up a fixed bed laboratory-scale reactor with simulated flue gas introduction and continuous exhaust mercury measurement. The bed will include pellets and carbon catalyst. The fixed bed reactor is sized to be 2" diameter and the bed depth will be varied (8 to 16") depending on the type of kiln. This setup will help us to evaluate the behavior of activated carbon in the pellet bed. A schematic of the fixed bed reactor system is shown in Figure 1. It consists of a fixed bed reactor which is 2" in diameter. We are using nitrogen gas instead of simulated flue gas. This helps us to understand activated carbon behavior under different conditions inside the reactor. We are using a PVC pipe of 2" diameter as our reactor for visibility purposes. Activated carbon is added into the carbon fluidized bed assembly through which it is carried out to the reactor with the help of gas flow through the assembly.

Setup.gif

Figure : Experimental Setup for carbon addition into the flue gas

3.2 Experimental Observations

3.2.1 Dust Collection for dry pellets

Experimental Procedure

Weigh 1 kg of taconite pellets.

Weigh the filter. (Filter weight before = "A")

Attach a filter at the reactor outlet for collecting the dust.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow. (Gas flow rate = 110L/min)

Collect the dust particles at the outlet.

Let the gas flow for 15 minutes.

Remove the filter at the outlet and weigh the filter with dust. (B)

Attach a new filter at the outlet.

Calculate the weight of dust particles collected in first 15 minutes. (B-A)

Repeat the experiments for every 15 minutes till the dust accumulation on the filter becomes negligible.

Table : Dust collection for dry pellets

Time

Dust weight at the outlet(g)

Cumulative Dust Collection(g)

15 minutes

0.0856

0.08562

30 minutes

0.0456

0.13122

45 minutes

0.0225

0.15372

60 minutes

0.0195

0.17322

75 minutes

0.0050

0.17822

90 minutes

0.0030

0.18122

Figure : Dust collection at the outlet v/s Time graph

Figure : Cumulative dust collection at the outlet v/s Time graph

3.2.2 Dust collection for wet pellets

Experimental Procedure

Weigh 1 kg of fresh fired pellets.

Weigh the filter.(A)

Add 100 ml of distilled water into a squeeze bottle.

Coat the pellets with water.

Remove the excess water from the pellets by drying them in the oven at 105 C.

Put the dry pellets into the reactor.

Attach the filter at the outlet.

Switch on the gas flow. (Flow rate= 110L/min)

Let the gas flow for 15 minutes.

Remove the filter at the outlet and weigh the filter with dust.(B)

Attach a new filter at the outlet.

Calculate the weight of dust particles collected in first 15 minutes. (B-A)

Repeat the experiments for every 15 minutes till the dust accumulation on the filter becomes negligible.

Table : Dust collection for wet pellets

Time

Dust weight at the outlet(g)

Cumulative Dust Collection(g)

15 minutes

0.0654

0.0654

30 minutes

0.0324

0.0978

45 minutes

0.0224

0.1202

60 minutes

0.0089

0.1291

75 minutes

0.0050

0.1341

90 minutes

0.0020

0.1361

Figure : Dust collection at the outlet v/s Time graph

Figure : Dust collection at the outlet v/s Time graph

3.2.3 Experiments with dry pellets

Experimental Procedure

Weigh 1 kg of taconite pellets.

Weigh 1 gram of activated carbon.

Weigh the filter. (Filter weight before = "A")

Attach a filter at the reactor outlet for collecting the activated carbon.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow for 5 minutes. (Gas flow rate = 110L/min)

Collect the dust particles at the outlet.

Add activated carbon.

Slowly increase the flow rate through activated carbon bed so as to obtain fluidized bed of carbon.

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Obtain the percentage recovery of activated carbon. ( (B-A)/1*100)

Table : Percentage recovery of carbon with dry pellets

Experimental Run

% Recovery

Run 1

60

Run 2

61

Run 3

82

Run 4

85

Run 5

86

Average

75

Figure : Percent recovery of activated carbon for each run with dry pellets

3.2.4 Experiments without pellets

Experimental Procedure

Weigh 1 gram of activated carbon.

Weigh the filter. (Filter weight before = "A")

Attach filter at the reactor outlet for collecting the activated carbon.

Attach the reactor to the holder.

Check all the connections.

Switch on the gas flow for 5 minutes. (Gas flow rate = 110L/min)

Collect the dust particles at the outlet.

Add activated carbon.

Slowly increase the flow rate through activated carbon bed so as to obtain fluidized bed of carbon.

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Obtain the percentage recovery of activated carbon. ( (B-A)/1*100)

Table : Percentage recovery of carbon without pellets

Experimental Run

% Recovery

Run 1

82

Run 2

78

Run 3

70

Run 4

78

Run 5

94

Average

81

Figure : Percent recovery of activated carbon for each run without pellets

3.2.6 Experiments with wet pellets

Experimental Procedure

Weigh 1 kg of taconite pellets.

Weigh the beaker. (Bw)

Add dry pellets to this beaker.

Measure 100 ml of distilled water

Add this distilled water to pellets.

Weigh the beaker containing pellets and distilled water.(Pw)

Keep the beaker into oven to evaporate excess water. Maintain oven temperature around 105 C.

After every 15 minutes, weigh the pellets.

Continue evaporating the water until weight of wet pellets comes to 1.1 kg.

Weigh 1 gram of activated carbon.

Weigh the filter. (Filter weight before = "A")

Attach a filter at the reactor outlet for collecting the activated carbon.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow. Let the gas flow for 5 minutes. (Gas flow rate = 110L/min)

Collect the dust particles at the outlet.

Add activated carbon.

Slowly increase the flow rate through activated carbon bed so as to obtain fluidized bed of carbon.

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Calculate the percentage recovery of activated carbon. ( (B-A)/1*100)

Table : Percentage recovery of carbon for experiments with wet pellets

Experimental Run

% Recovery

Run 1

7

Run 2

11

Run 3

13

Average

10

Figure : Percent recovery of activated carbon for each run with wet pellets

3.2.7 Experiments for different carbon sizes

Experimental Procedure

Weigh 1 kg of taconite pellets.

Weigh 1 gram of activated carbon.

Weigh the filter. (Filter weight before = "A")

Attach filter at the reactor outlet for collecting the activated carbon.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow. Let it flow for 5 minutes. (Gas flow rate = 110L/min)

Collect the dust particles at the outlet.

Add activated carbon.

Slowly increase the flow rate through activated carbon bed so as to obtain fluidized bed of carbon.

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Obtain the percentage recovery of activated carbon. ( (B-A)/1*100)

Repeat the experiment for other sizes of carbon.

Table : Percentage recovery of carbon for experiments with different carbon sizes

Carbon Type

Size

% Recovery

C1

carbon "C" > 1.2 mm

10

C2

1.2 mm >"C"> 0.853 mm

75

C3

0.853mm > "C" > 0.599 mm

69

C4

0.599mm> "C" > 0.251 mm

69

C5

0.251mm> "C" > 0.178 mm

81

Figure : Percent recovery of carbon for different sizes of carbon

3.2.8 Experiments with low flow rate (20 lit/min)

Experimental Procedure

Weigh 1 kg of taconite pellets.

Weigh 1 gram of activated carbon.

Weigh the filter. (Filter weight before = "A")

Attach a filter at the reactor outlet for collecting the activated carbon.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow for 5 minutes. (Gas flow rate = 20 Lit/min)

Collect the dust particles at the outlet.

Add activated carbon.

Slowly increase the flow rate through activated carbon bed so as to obtain fluidized bed of carbon.

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Obtain the percentage recovery of activated carbon. ( (B-A)/1*100)

Table : Percentage recovery of carbon for experiments with low flow rate

Experimental Run

% Recovery

Run 1

20

Run 2

14

Average

17

Figure : Percent recovery of carbon for dry pellets with low flow rate

3.2.9 Experiments for Activated Carbon coated pellets

Experimental Procedure

Weigh 1 gram of activated carbon.

Add 100 ml of distilled water and activated carbon into a squeeze bottle.

Weigh 1 kg of taconite pellets

Coat pellets with activated carbon slurry.

Keep the pellets into oven to evaporate excess water. Maintain the oven temperature at 105 C.

Weigh the filter. (Filter weight before = "A")

Attach a filter at the reactor outlet for collecting the activated carbon.

Put taconite pellets inside the reactor and attach the reactor to the assembly.

Check all the connections from N2 tank.

Switch on the gas flow. (Gas flow rate = 110 Lit/min)

Let the gas flow for 20 minutes.

Weigh the outlet carbon on the filter.(Filter weight after = "B")

Obtain the percentage recovery of activated carbon. ( (B-A)/1*100)

Table : Percentage recovery of carbon for experiments with activated carbon coated pellets

Experimental Run

% Recovery

Run 1

63

Run 2

56

Run 3

58

Average

59

Figure : Percent recovery of carbon for activated carbon coated pellets

3.3 Analysis of Data - Carbon addition to flue gas

Observed values for percent recovery of carbon leads us to following conclusions:

Dust collection values for dry pellets are gradually decreasing with time.

Percentage recovery of carbon decreases as the size of the carbon particle increases.

Carbon sticks to wet pellets and hence percentage recovery for wet pellets is very low.

With lower flow rates (20 L/min), percentage recovery of carbon has a very low value.

We are losing 10-20% carbon during its passage from fluidized bed to outlet. This could be due several reasons. Since the pipe develops static charge, carbon sticks to the walls of the reactor. Similarly, it could get clogged in the valves or pipelines.

Hence, after careful review of the results obtained from the experiments for carbon addition to the flue gas, it was proved that in the given circumstances it is important to evaluate the possibility of addition of carbon to the green ball feed.

3.4 Carbon Addition to Green Ball feed

Previous studies conducted by Minnesota Department of Natural Resources (DNR) suggested that ore is the main source of mercury and mercury is released during taconite processing in the agglomeration step. Most of the mercury release takes place between 200 oC up to 600 oC. (35) This temperature is observed in the pre-heat section of the induration kilns. Hence, in order to oxidize the elemental mercury and in turn to capture it in the scrubbers, it is important to understand and evaluate the possibility of adding oxidizing agent in to the green ball feed. This approach explores the possibility of oxidizing the mercury in the preheat section as soon as or even before the release of mercury to the flue gases.

In order to evaluate this possibility, green balls were obtained from two plant sites namely United Taconite (UTac) and Minnesota Taconite (MinnTac). Green ball samples used from United Taconite were obtained in January, 2012 and those from Minnesota Taconite were obtained in two batches October, 2011 and February, 2012. Green balls were tested over a period of 4 to 5 months in which the test equipment was continuously optimized. Results presented under this section are obtained before and after the equipment was optimized.

The proposed technology involves use of carbon based additive known as ESORB-HG-11 which will be added to green ball feed prior to induration to reduce mercury emissions. This is a proprietary enhanced Powdered Activated Carbon (PAC) which contains trace amounts of halogens and hence it is a low corrosion method to enhance mercury oxidation. In this section, research mainly focuses on the preliminary work done in the laboratory to determine if this technology has a potential to oxidize the mercury significantly when included in the green ball formulation.

The research work done under this section mainly investigated:

Optimal additive to green ball ratio

Additive and green ball combination method - thorough mixing or surface addition

Effectiveness of halogen enhanced carbon against plain carbon

3.4.1 Experimental Setup

The bench scale apparatus is illustrated in Figures 2 and 3. It consists of a tube furnace, reaction vessel, a gas metering system, gas conditioning, mercury pretreatment system, and mercury analyzer.

The procedure for testing involves placing approximately 100 grams of green balls into the reaction vessel and heating the green balls up to 700 °C. During the heating process, air passes through the vessel at 7.5 lpm(during initial testing, flowrate = 5 lpm), and flows through heated PFA tubing to a pretreatment system and then directly to the analyzer for an elemental mercury determination.

Before each run, the Horiba undergoes a calibration or calibration verification. While this goes on, the PFA tubing is disconnected from the impinge train and preheated to 170 0C to prevent condensation or reduction of oxidized mercury in the lines with the help of heating tape. The furnace reactor is also heated to 700 0C and then allowed to cool to 250 0C to drive out residual mercury in the furnace and simulate average temperatures experienced by green balls during induration at a Taconite facility. During testing, once the green balls are added to the reactor the temperature of the reactor is increased to 700 0C with a ramp rate of 20 0C per minute based on calculations from field testing conditions. Note that due to heat losses in the bench scale assembly, the actual ramp rate decreases as the temperature of the reactor bed increases, resulting in a slower overall ramp rate when compared with the field conditions.

Figure : Schematic of testing equipment

Figure : Pictures showing Horiba DM-6B mercury analyzer, Wet-chemistry impinger train, reactor vessel

As shown in Figure , a wet pre-treatment unit was used to condition the flue gas before it enters the Horiba mercury analyzer. It consisted of two parallel sets of impingers (4 impingers in total). One set is used to determine the elemental mercury concentration (Hg0) while the other set is used to determine the total mercury concentration (HgT) in the sample flue gas.The set-up was designed based on a modified wet chemistry PS Analytical pre-treatment conversion system and ASTM D6784-02 (also known as the Ontario Hydro [OH] method).

The first impinger train is for conditioning the elemental mercury stream which consists of two impingers in series: The first impinger contains a 150 ml of 10 weight percent potassium chloride (KCl) and 0.8 weight percent of Sodium Thiosulfate (Na2S2O3) solution that captures the oxidized mercury in order to obtain only elemental mercury concentration, while the second impinger sits in an ice bath and traps all moisture present in the gas sample before analysis by the mercury analyzer.

The second impinger train is for conditioning the total mercury stream. Here, the first impinger contains 150 ml of 0.8 weight percent stannous chloride (SnCl2) solution and 20 weight percent of Sodium Hydroxide (NaOH). The SnCl2 reduces the oxidized mercury in order to obtain a total mercury measurement of the flue gas. The second impinger also sits in an ice bath and traps all moisture present in the gas sample before analysis. The trains were modified from a continuous flow to a batch system. The Horiba mercury analyzer simultaneously and continuously measures both total and elemental mercury. The difference between the total and elemental is assumed to be oxidized mercury. Gas flow rates are measured with rotameters and were validated with mass flow controllers.

Impingers

Total Hg Sample

Hg0 Sample

Impinger Solution 1

Impinger Solution 2

Hg0 analysis

Total Hg analysis

Ice

bath

Ice

bath

Impingers

Figure : Schematic diagram of impinger train

3.4.2 Carbon Addition Method

Two methods for adding trace amounts of carbon to the pellets were used for the phase one testing:

Mixed Addition

Surface Addition

Mixed Addition Procedure

Crush a random sample of green balls from the given sample.

Weigh 100 grams of crushed sample.

Weigh the mass of additive to be tested.

Mix the additive with 100 grams of crushed green ball sample.

Add 1 to 2 ml of water to the given mixture.

Roll the mixture into balls with required size.

Surface Addition Procedure

Crush a random sample of green balls from the given sample.

Weigh 90 grams of crushed sample.

Add 1 to 2 ml of water to the crushed sample.

Roll the mixture into balls with a smaller size, (90 percent of average taconite size was used.)

Weigh 10 grams of crushed sample.

Weigh the mass of additive to be tested.

Mix the additive with 10 grams of crushed green ball sample.

Roll the smaller green balls in this mixture so as to get a coat of additive and green ball on the surface.

During baseline runs, the green balls were also prepared using the same procedure as for the mixed addition; however, no carbon was added to the new green balls. This ensured that any impact the production process had on the mercury emission profile would be inherent to both baseline and mixed/surface tests. Figure 1 is a picture of the final green ball pellets produced by hand. Once produced the green balls were then subjected to heating tests to determine the amount of oxidation occurring as a result of carbon addition.

IMG_4757 copy

Figure : Picture of new green balls produced for phase one testing

3.5 Test Matrix

The test matrix was designed to achieve objectives explained in Section 3.4 of this thesis. In Phase I, green balls from two different plants (MinnTac and UTac) were evaluated for mercury release and effect of additives on level of mercury oxidation was studied. We conducted tests based on the following preliminary test matrix. Slight variations in the testing may result based on intermediate findings.

Table : Minntac Test Matrix

Plant

Flow Rate

Maximum Bed Temperature

Impinger Solution

Additive

Method

Additive Loading (wt%)

MinnTac

7.5

700

KCl +Na2S2O3

None

-

0

MinnTac

7.5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.1

MinnTac

7.5

700

KCl

ESORB-HG-11

Mixed

0.1

MinnTac

7.5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.2

MinnTac

7.5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.3

MinnTac

7.5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.5

MinnTac1

5

700

KCl

None

-

0

MinnTac1

5

700

KCl

ESORB-HG-11

Mixed

0.1

MinnTac1

5

700

KCl

ESORB-HG-11

Mixed

0.2

MinnTac

5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.05

MinnTac

5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.2

MinnTac

5

700

KCl +Na2S2O3

ESORB-HG-11

Mixed

0.3

MinnTac

7.5

700

KCl

Halogenated salt

Mixed

0.01

MinnTac

7.5

700

KCl +Na2S2O3

PAC

Mixed

0.2

MinnTac

5

700

KCl

ESORB-HG-11

Surface

0.2

Table : Utac Test Matrix

Plant

Flow Rate

Maximum Bed Temperature

Impinger Solution

Additive

Method

Additive Loading (wt%)

UTac

5

700

KCl

None

-

0

UTac

5

700

KCl +Na2S2O3

None

-

0

UTac

5

700

KCl

ESORB-HG-11

Mixed

0.1

UTac

5

700

KCl

Halogenated salt

Mixed

0.01

UTac

5

700

KCl

PAC

Mixed

0.1

UTac

5

700

KCl

ESORB-HG-11

Surface

0.1

UTac

5

700

KCl

ESORB-HG-11

Surface

0.2

3.6 Experimental Results and Discussion

Results presented in this section are broadly divided into two sections depending on the plant from which the green balls are obtained. Results presented in this section are mostly after the equipment was modified and are at similar conditions for comparision.

3.6.1 MinnTac Results

In Phase I, Minntac green balls testing had several objectives as follows:

Evaluating effectiveness of different additives for mercury oxidation.

Optimizing the additive loading amount.

Surface v/s Mixed addition results.

I. Effectiveness of different additives for mercury oxidation

Table summarizes the results obtained from three different additives namely Esorb-HG-11, Powdered Activated Carbon (PAC) and halogenated salt with a baseline run. No significant oxidation was observed with PAC or Halogenated salts. In the literature, addition of halogenated salts into the grates during the field testing was proved to effective in Hg oxidation. (1) However, during the lab scale testing of PAC addition into the green ball feed did not show any significant reduction. This can be due to several reasons such as lack of residence time of carbon in the lab scale setup as well as lack of halogens in the flue gas. This could also suggest that Hg oxidation in the full-scale facility is more significant than the levels observed in the lab scale setting. Figure , , , , represents the mercury emission profiles with different additives.

Table : Minntac Test Results with different additives

 

Additive Loading

Loading Ratio

Oxidation

Additive

(wt.%)

(mg/kg)

(%)

None

0

0

7.00

PAC

0.2

2000

1.01

Halogenated Salt

0.2

2000

19.47

Esorb-HG-11

0.2

2000

67.00

Figure : Mercury Emission Profile from MinnTac green balls with no additive (Baseline)

Figure : Mercury Emission Profile from MinnTac green balls with PAC additive

Figure : Mercury Emission Profile from MinnTac green balls with halogenated salts

Figure : Mercury Emission Profile from MinnTac green balls with Esorb-HG-11

II. Additive Loading

Table summarizes the results obtained from Minntac green ball testing with different additive loading. Mixed addition technique was used to form the green balls used in this testing. These tests used two different batches of green balls obtained in October 2011 and February 2012. There was a concern over "aging" of the green ball formulation and hence similar tests were performed on the green balls obtained in February, 2012. The two tests performed gave similar results; hence, data generated in February, 2012 is reported in this section. The results from October, 2011 are reported in Appendix A.

Table : Results from Minntac green ball testing - Additive loading

 

Additive Loading

Loading Ratio

Percent Oxidation

Additive

(wt.%)

(mg/kg)

Runs (%)

Esorb-HG-11

0

0

7

26.7

Esorb-HG-11

0.05

500

39.7

33

Esorb-HG-11

0.1

1000

54.7

35.3

Esorb-HG-11

0.2

2000

46.9

67

Esorb-HG-11

0.3

3000

50.1

46.5

Esorb-HG-11

0.5

5000

50.6

3.6.2 UTac Results

In Phase I, Utac green ball testing was done in two steps namely mixed addition and surface addition. Results for mixed addition are presented in Table and results from surface addition are presented in Table.

Table includes the results of mixed addition with different additives which includes Esorb-HG-11, Powdered Activated Carbon (PAC) and halogenated salts. First run mentioned in the table is a baseline run in which green balls are produced as per the procedure in section 3.4.2 of this thesis without any additive. Comparing the baseline run to with additives runs, it is clear that none of the additives have significant effect on the elemental mercury level. After referring to numerous resources, it was found that UTac green ball formation process is susceptible to aging and hence the testing was postponed till Phase II without any solid conclusion when a fresh batch of green balls will be obtained.

Table : Mixed Addition Results

 

Additive Loading

Loading Ratio

Hg0 Curve Area

HgT Curve Area

Oxidation

Additive

(wt.%)

(mg/kg)

(ng)

(ng)

(%)

None

0

0

1538

1932

20.4

Esorb-HG-11

0.1

1000

914

1331

31.3

PAC

0.1

1000

913

1206

24.3

NH4Br

0.01

100

943

1314

18.2

Table : Surface Addition Results

 

Additive Loading

Loading Ratio

Hg0 Curve Area

HgT Curve Area

Oxidation

Additive

(wt.%)

(mg/kg)

(ng)

(ng)

(%)

Esorb-HG-11

0.1

1000

122

146

16.4

Esorb-HG-11

0.1

1000

218

401

45.6

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.