GB2516728A

GB2516728A - Fuel enrichment process

Info

Publication number: GB2516728A
Application number: GB1408385.1A
Authority: GB
Inventors: Syed Sheraz Daood; William Nimmo
Original assignee: INTERNAT INNOVATIVE TECHNOLOGIES Ltd
Current assignee: INTERNAT INNOVATIVE TECHNOLOGIES Ltd
Priority date: 2013-05-10
Filing date: 2014-05-12
Publication date: 2015-02-04
Anticipated expiration: 2034-05-12
Also published as: US20160084500A1; WO2014181133A2; ZA201509033B; GB201308472D0; EP2994438A2; CN105263880A; GB2516728B; JP2016527328A; KR20160067811A; GB201408385D0; RU2015153105A; WO2014181133A3

Abstract

A process for breaking hydrocarbon bonds during the combustion or heating of a particulate hydrocarbon containing fuel bound carbon and fuel bound nitrogen comprises the steps of: heating the particulate hydrocarbon with a fuel improver in a burner, the fuel improver comprising predominantly iron oxide and silicon dioxide. The proportion of fuel improver to fuel is up to 33% by weight of particulate hydrocarbon. The particulate hydrocarbon has a particle size distribution and the fuel improver has a particle size distribution, and particles of hydrocarbon at the 90th percentile of its particle size distribution are larger than the particles of the fuel improver at the 90th percentile of its particle size distribution, and wherein the particles of fuel improver at the 10th percentile of its particle size distribution are an order of magnitude smaller than the hydrocarbon particles at the 90th percentile its particle size distribution. The fuel is preferably coal and the fuel improver is preferably a mixture of iron oxide and silicon dioxide. A pozzolanic fly ash can be made by burning the fuel and improver, and a cementitious composition made therefrom.

Description

Fuel Enrichment Process

Field of the Invention

The invention relates to a process for reducing emissions of nitrous oxides (NC) from the combustion of hydrocarbon fuels, and more specifically to an improved process for the combustion of fossil fuels which reduces NO emissions and results in an ash by-product with improved loss on ignition and lower carbon content.

Background of the Invention

In recent years. there has been a growing concern about air pollution froni industrial processing units, and in particular from coal tired power stations.

The Large Combustion Plant Directive (TCPD) isa European Union Directive which requires EU Member States to limit emissions from certain combustion plants, including fossil fuelled power stations. [he Directive specifies emission limits for sulphur dioxide and nitrogen oxides. In Europe the enforcement of the ICPD is going to close many conventional coal fired power stations. The noxious oxides of nitrogen, carbon and sulphur are causing fundamental damage to the climate, environment and r human health -(0 In particular there has been substantial interest in finding systems to reduce or minimi?.e the emissions, for which catalysts and additives have been developed.

Solutions to this problem have included coal fired power stations being fitted xvith loxv NOx burners. Whilst reducing NOx and SOx enlissions. these burners also lead to a loss in combustion efficiency which can in turn lead tc high levels of unburnt carbon in the ash, typically in the region of 20% carbon, rendering the ash an undesirable waste product.

The United States of J\merica, Tndia. China and z\ustralia are the major producers of fly ash. In 2009, the USA alone produced 57.2 Million metric ton (Mton) of fly ash of which only 22.4 Mton was used in concrete/cement manufacturing (http://minerals.usgs.gov/ds/2005/140/ coalcomhustionproducts.pdf). Fly ash mainly comprises glassy spheres of oxides of silicon, aluminium together with unburnt carbon and some crystalline matter. The introduction of low NOx burners has led to a gradual rise in Lcss on Ignitition (LOl) values of fly ash. [he unburnt carbon is siilficant in air-entt-ained concrete mixtures because of its tendency of adsorbing air-entt-ained surfactant rendering less protection against freeze-thaw conditions. Similarly excessive carbon affects the optimum density and moisture content for filling applications. In order for fly ash to be used as a cement substitute it must have less than 794p 1.01.

Catalysts and ifiel additives have also been developed in order to t-educe or minimise NOx and SOx emissions.

For example, US200G/0034743 (Radway) describes an additive aimed at reducing Sulphur emissions. NOx emissions and heavy metal deposits in coal-fired boilers. The additive contains predominantly magnesium oxide combined with oxides of calcium, silicon, iron and aluminium.

Ash is a by-product generated in the combustion of coal, Ely ash is generally captured from the chimneys of power stations and bottom ash is removed from the bottom of the furnace. In the UK, just over 1,000,000 tonnes of fly ash is produced annually \\hrldwide a large proportion of ash produced from coal fired power stations is disposed of in landfill or stored in slag heaps. Some countries impose a tax on the disposal of such waste in landfill. The r recycling of ash has beconie an increasing concern In recent years due to increasing landfill costs as well (0 as environmental costs.

A significant portion of this ash is pozzolanic in nature, which means that when combined with calcium hydroxide it exhibits cemenritious properties. In principle fly ash can be used as a replacement fur a proportion of Portland cement content of concrete mixtures. Production of Portland cement itself is energy-intensive and produces a large amount of carbon dioxide, approximately one tonne of carbon dioxide per tonne of Portland cement, so replacement of a proportion of this with an otherwise unused by-product could dramatically reduce carbon emissions.

However, ash comprising a high percentage of unburned carbon is not useable asa Portland cement substitute since the ash then has a tendency to adsorb important cemnentitious chemical admixtures from the concrete during die mixing process. This renders admixeures unavailable to effect their intended purpose. Ash with a carbcun content of 7% cur less is desirable fbr use as a pcrzzolan.

I

Fly ash can be processed to reduce the carbon content to ieveis sufficient for use asa pozzolan Examples of such processes include re-burning the fly ash to reduce the carbon content; electrostatic separation processes which produce lo\v carbon fractions and the chemical treatment of fly ash to minimize the effect of the carbon content by reducing the adsorptive properties of the carbon. All of these processes require at least one additional processing step, adding to the overall cost of producing a useful by-product rather than a waste product.

it would be desirable in prcvide an improved prccess fhr the combustion of coal, reducing emissions of nitrous oxides, whilst also providing a better quality ash with a low carbon content that renders the ash a saleable and marketable by-product, rather than a waste product that would need to he disposed of in a manner to satisfy environnmntal regulations. in addition, it would be desirable to provide an improved process in which the amount of coal burned is reduced xx1ilst not reducing the energy output and preferably increasing the energy output and reducing carbon emissions.

Summary of the Invention r

r One aspect of the invention provides a process for breaking hydrocarbon bonds during the (0 combustion or heating of a particulate hydrocarbon containing fuel hound carbon and fuel hound c'J nitrogen, the process comprising the steps ot: heating die particulate hydrocarbon with a fuel improver in a burner, die fuel improver comprising predominantly iron oxide and silicon dioxide, wherein: the particulate hydrocarbon has a particle size distribution and the friel improver has a particle size distribution, and wherein particles of hydrocarbon at the 90th percentile of its particle size distribution are larger than the particles of the friel improver at the 90th percentile of its particle size distribution, and wherein the particles of fuel improver at the 10th percentile of its particle size distribution are an order of maiirude smaller than the hydrocarbon particles at the 90th percentile its particle size distribution; and \vherein die proportion of fuel improver to fuel is up to 33% by weight of particulate h yd roca rbon.

Preferably, the particle size of the fuel improver at the 90± percentile is 55 micron or less, the particle size of the particulate hydrocarbon at the 90th percentile is 60 micron or greater, and the particle size of the fuel improver at the 10th percentile is less than 6 micron.

Preferably, the particle size of the particulate hydrocarbon in the 100th percentile is between 60 micron anti 700 micron.

Preferably the particle size of the fuel improver at the 90th percentile is less than or e9ual to 35 micron, More preferably, the particle size distribution of the fuel improver at the 90th percentile is less than or ec1ual to 32 micron. Still more preferably, the particle size of the fuel improver at the 90th percentile is less than or equal to 25 micron.

Preferably, the volume weigited mean particle size of the fuel improver particles is 26 micron or less. More preferably, the volume weighted mean particle size of the fuel improver is 17 micron or less.

Preferably, the volume weighted mean particle size of the particulate hydrocarbon is in the range 29 micron to 90 micron.

r Preferably, the surface weighted mean particle size of the fuel improver is 9 micron or less. More r preferably, the surface weighted mean particle size of the fuel improver is 6 micron or less. (0

Preferably, the surface weighted mean particle size of the particulate hydrocarbon is in the range 9 micron to 35 micron.

Preferably, the particle size of the particulate hydrocarbon at the 90th percentile is not greater than 220 micron, More preferably, the particle size of the particulate hydrocarbon at the 90th percentile is not greater than 180 micron.

Preferably, the proportion of fuel improver to fuel is in the range 1% to 33% by weight of particulate hydrocarbon. More preferably, the proportion of fuel improver to ftuel is in the range 5%-12% by weight of particulate hydrocarbon.

Preferably, the fuel improver replaces between 1% and 5% of the particulate hydrocarbon and the proportion of the1 improver is equal to or greater than the amount of parñculate hydrocarbon replaced. More preferably, the fuel improver replaces between 1% and 3% of the particulate hydrocarbon.

Preferably, the parñculate hydrocarbon is coal.

Preferably, the fuel improver includes aluminium oxide and/or calcium oxide. _\Iore preferably, the fuel improver includes by weight up to 7% aluminium oxide and/or up to 10% calcium oxide. Still more preferably, the friel improver includes by weight between 3% and 5.5% aluminium oxide and/or between 2% and 7.5% calcium oxide.

Preferably, the fuel improvcr comprises between 70% and 91% iron oxide and silicon dioxide, combined, by xveight.

Preferably, the fuel improver comprises between 41 and 52°/b iron oxide, by weight.

Preferably, the fuel unprover comprises between 32 and 40% silicon dii xide, by weight.

A second aspect of the invention provides a process fbr producing a po'zzolanic fly ash during the combustion of a particulate hydrocarbon containing fuel bound carbon and Fad bound nitrogen, the process comprising the steps of performing the process for cracking hydrocarbon bonds during the combustion or r heating of a particulate hydrocarbon as hereinhefore defined; and r (0 recovering the fly ash from the burner, wherein the particle size not more than 30% of the recovered fly ash is greater than 45 micron.

Preferably, the particle size of not more than 20% of the recovered fly ash is greater than 45 micron, Preferably, at least 70% by weight of the fuel improver is comprised of silicon dioxide, iron oxide and aluminium oxide. More preferably, at least 75% by weight of the fuel improver is comprised of silicon dioxide, iron oxide, aluminium oxide and calcium oxide.

Preferably, the pozzolanic t'ly ash has a loss on igaltion of 7% or less.

1k third aspect of the invention provides a process for producing a cementitious composition comprising performing the process for producing a pozzolanic fly ash as hereinbefore defmed. and mixing the resulting pozzolanic fly ash with calcium hydroxide.

A fourth aspect of the invention provides a composite cement comprising a mixture of portland cement and a cementitious composition produced by the process for producing a cementitious composition as hereinbefore defined and/or a pozzolanic fly ash produced by the process for producing a pozzolanic fly ash as hereinbefore defined.

Prefet-ably, the pi-opot-tion of portland cement to cernentitious composition produced by the process for producing a cementitious composition as hereinhefore defined and/or pozzolanic fly ash produced by thc process for producing a pozzolanic fly ash as hereinbefbre defined is between 30:70 and 70:30.

A fifth aspect of the invention provides a process for reducing slagging in a boiler and/or fouling of a heat recovery apparatus arising from combustion of a particulate hydrocarbon in a burner, the particulare hydrocarbon containing fuel hound carbon and fuel hound nitrogen, the process comprising the step of: perfurming the process for cracking hydrocarbon bonds during the combustion or heating of a particulate hydrocarbon as hereinbefore defined in a burner associated with a boiler r r and/or hear recovery appararus and controlling the amount of fuel improver relative to the (Q particulate hydrocarbon to provide a basic to acidic ratio and/or a simplified basic to acidic ratio (R(/A)) such thar the slagging and/or fouling indices of the combination of particulate hydrocarbon and fuel improver are below the extremely high range.

Preferably, the slagng index is not eater than 2.

Preferably, the fouling index is not greater than 2.

More preferably, the slagging and/or fouling indices are 0.6 or less.

Preferably, the particulate hydrocarbon is coal.

Preferably, the fuel improver includes Na20 and/or K0 and the particulate hydrocarbon includes Na20 and/or I-C20 and the relative proportion of fuel improver to particulate hydrocarbon is selecred according to the proportions of Na20 and/or 1<20 in the fuel iniprover and the particulate hydrocarbon respectively.

A sixth aspect of the invention provides a process for reducing NOx emissions during the combustion or heating of a particulate hydrocarbon containing fuel bound carbon and fuel bound nitrogen, the process comprising the steps of: performing the process for cracking hydrocarbon bonds during the combustion of a particulate hydt-ocarbon as hereinbefore defined to increase the propot-tion of volatile nitrogen to char nitrogen and volatile carbon to char carbon.

Preferably, combustion of the particulate hydrocarbon produces a flame having a fuel rich zone and wherein the proportion of volatile nitrogen to char nitrogen and/or volatile carbon to char carbon is increased in the fuel rich zone.

Preferably, the process comprises the further step of reacting volatile nitrogen with metal oxides of the fuel improver and volatile carbon oxides.

Brief Description of the Drawings

r In the drawings, which illustrate preferred embodiments of the invention: r Figure Ia is a graph showmg the distribution particle sizes of a sample of water cooled fuel improver after pulverisation using a roller mill; Figure lb is a graph showing the distribution particle sizes of a sample of air coiled fuel improver after pulvensation using a roller niill; Figures 2 a, b and c are graphs showing the distribution of particles sizes of three different coal samples after pulverisation using a roller mill.

Figure 3 shows a series of graphs illustrating the effect of the mass fl-action of the fuel mpmover on NO emissions from various commercial coals under un-staged tiamne conditions of stoichiornetric ratio = 1.20; A: Water Cooled WC) fuel improver with Russian Coal (RC); A': Air Cooled fuel improver (AC) with RC; B: WC fuel improver with Columbian Coal (CC); B' AC fuel improver with CC; C: WC fuel improver with Kellingley Coal (Kg; AC fuel improver with ICC.

Figure 4'a illustrates a proposed schematic mechanism for the interaction of the fuel improver particles with the coal particles; Figure 4b illustrates the NOX reduction chemistry pathway in the presence of fuel improver; Figm-e 4c illustrates a schematic mechanism for the thermal degradation of Fuel-C (coal) to Volatile -C ighter hydrocarbons) in the presence of the fuel improver; Figure 5 illustrates a two stage fixed bed reactor; Figure 6a is a graph showing CO release during the combustion of different mixtures of AC fuel impt-ovet-and coal; Figure Oh is a graph showing hydrocarbon concentration during the conthustion of different mixtures of AC fuel improver and coal; Figure 7 shows a series of graphs illustrating NO emissions from various commercial coals with varying mass fraction of both Air Cooled fuel improver (AC) and Xitter Cooled fuel improver (V/C). A to C: 6.4%, 8.8% and 13% mass fraction of V/C fuel improver with Russian Coal (Rfl; D to F: 5451⁄4, 6.2% and 7.8% mass fraction of \VC, AC and \VC fuel improver with Columbian Coal (CC) respectively; and 0 (0 to I: 8.9%, ¶1.1% and 10.3% mass fraction of AC, WC and AC fuel improver with Kellingley Coal (KC), respectively; Figure 8 shows a series of graphs illustrating the LOT of fly ash plotted against the mass fraction of fuel improver for thrce different commercial coals: A: Russian Coal; B: Columbian Coal; C: Kellingley coal; Figure 9 illustrates a coniparison between the particle size distribution of fly ash resulting from burners burning only RC, ICC or typical UIC coal, with burners burning RC and 4.2% fuel improver and KC with 4.8% fnel improver; Figure 10 illustrates temperature measurements within the burner for different mass fractions of fuel improver with different conunercial coals; Figure 11 isata1)le showing calculated and predicted slagging and fouling indices for various coals and blends of coals and fuel improver; and Figi.we 12 isa graph illustt-ating dust concennation in the exhaust gas downstream of an electrostatic precipitator in the west, centre and cast legs of a boiler exhaust.

Detailed Description of the Preferred Embodiments

The improved combustion ptocess of the invention involves the injection of a fuel improver into the main burner in a carbon-based fuel burner, for example a coal fired power station. Ihe fuel improver is derived from a mixture of metal oxides typically sourced from slags, which are by-products of metal smelting processes, typically in the production of copper and nickel. The tael improver includes a mixture of oxides c,f transitic,n metals and other elements. The fuel improver mainly includes a mixture of iron, aluminium, calcium and silicon oxides. Two different types of fuel improver can be produced: one is air cooled, and the other is water cooled. Table I shows the X-Ray fluorescence (YRE) analysis of two fuel improver samples. lable 2 sho'.vs the X-Ray Diffraction (XRD) analysis of the two samples. r r (0 (4

Iablc 1: XltF analysis of two samples of fuel hnprover.

Water Cooled Fuel, Air Cooled Fuel Improver, Component Improver,(%) (%) Na 0.8-1.3 0.5-0.97 MgC) 1.62-1.98 1.0-1.7 Al2 4.71-5.1 3.1-5.5 so + 3469-385 3214-390 + K2 0.362-0.6 0.35-0.847 LaO 2.5-7.06 2.0-5.37 TO 0.09-0,25 0,1-0,28 be2 42.22-51.9 41.2-50.9 0.208-0.9 0,1-0.721 SC) 0,2-1.05 0,5-0,75 Table 2: XRD analysis of two samples of fuel improver.

Water Cooled Fuel Air Cooled Fuel Improver Component Improver,(/o) (/o) Favlite, Fe 7-15 49-64 Maunetite. Fe. . -0. Trace r4.. Amorphous 81-91 21-25 r r. .. . . The tuel improver composition of the invention typically contains chemical elements and their oxides belonging to periods 3 and 4 roups II-\ of the Periodic Table, As sho\si in Table 1 the fuel improver comprises predominantly iron oxide and silicon dicixide, meaning that the cc,mbined amc,unt of iron oxide and silicon dioxide present in the fuel improver is greater than die amount of any other compound present in the fuel improver.

The Rid improver is preferably pulverised using a mill suitable for producing fine powders from hard materials such as a ball mill Oil a roller mill as described in UK patent numbers 0B2451299, GB2460505 and 0132471934. Preferably the fuel improver is milled such that the particle size at 90nh1 percentile is 55 micron or less Id(0.9) < 551. Table 3 shows the particle size distribution, physical and chemical properties of both types of fuel improver milled in a roller mills, with the particle sii.e distribution measured using various methods.

fable 3: Pardcle size distrihntion, physical and chemical properties of the Fuel Improver.

Water Cooled X/C) Fuel Improver Air Cooled AC) Fuel Improver Malvern Malvern Malvern Malvern Mastersizer Air Jet Mastersizer Mastersizer Air Jet Mastersizer Scirocco Sieving Scirocco flydro Sieving 2000 Flydro 2000 2000MU 2000MU d(O I) (pm 3 600 4 658 5 669 2 500 2 824 3 585 d50) 15000 21040 20527 12000 13610 14314 (mu) d(O 9) (pm) 34000 53484 50 18 260(1(1 34861 34277 \ol Weighted Mean -25.813 24.892 -16.672 17.001 ID[4,3] (LuI) Sui face \\èighted Mean -8.885 10.698 -5.461 7.154 DL3,2j (an) I pecific I_ surface -0.15 0.125 -0.244 0.189 area(m Melting 11301160 1100-1200 r Point( r Bulk Density 1.5-1.7 1.4-1.9 (kg/rn (41 Hardness 7-8 Mhos

I

Figures la and lb are graphs showing the range in diameter of particle sizes of the1 improver after passing through a mill, measured using a Malvern Mastersizer Scirocco 2000.

Figures 2 a, 0 and c are graphs showing the range in diameter of particles sizes of threc different coal saniples after passing through a mill, The particle sizes in these graphs were measured using a Malvern Mastersizer Scirocco 2000. Table 4 shows a summary of the particle size distribution of the three di ffe rent coal sarlrp] Cs.

Table 4: Particle size distribution of coal samples, measured using Malvern Mastersizer Scirocco R 1RC Columbian Coal UK Kellingley Coal ussian oa (CC) (KC) d(01) (jim) 10948 3.914 12.564 thU 50) (pm) 53 557 19 107 62 170 d(0.9) (jim) 179.228 L 69525 205362 Volume weighted 78655 29050 89.725 mean (pm) Ilie fuel improver can also replace a proportion of the carbon-based fuel in the burner in an amount ranging from 1% to 5% by weight depending upon the improvement in loss on ignition (LOl).

Carbon based firing lx)ilers can either produce the same steam load by burning less fuel or can increase steam load by burning die same fuel input, depending upon the amount of hid improver and impniivernent in 1.01.

Experiments were conducted using three commercially available coals as the carbon based fuel.

the coals had low; niedimi and hii ash contents. Co1Luihian coal is a low ash coal, Russian sub r bituminous coal is a medium ash coal and UK Kellingley coal is a high ash coal. The chemical r coniposition of these coals is listed in Table 5.

CD

Table 5: Chemical composition of tested coals.

Ultimate Analysis Columbian Coal UK Kellingley Coal Russian Coal (RC) as received, / (CC) (KC) C 66.29 71.63 59.9 11 4.55 4.82 3.89 N 209 096 1.3 C) (dift 895 1104 6.59 S 0.20 0.67 2.02 623 217 10.1 Ash 11.69 8.71 16.2 Proximate Analysis as received, O/ Volatile Matter 32.90 29.62 28.20 Fixed Carbon 49.20 59.5 45.50 + Ash 11.69 2.17 16.20 Moisture 6.23 8.71 10_to Net Calorific Value, 2800 900 3'0 MJ/kg -.

Experiments were conducted in a 100kW Combustion lest Facility comprising a down-fired pulverised coal fired furnace of 4 metres length with an internal diameter of 400mm. The burner of the furnace was operated at an approximately 10-11.65 kg/hr of coal (depending upon types of coal) input feed rate resulting in a net thermal input of about 75-85 k\V The major flue gas species C02, 0, NOX, and CC) were measured at the exit of the furnace, Gas samples were constantly drawn through a water cooled stainless steel probe to the gas sampling system in order that the correct combustion conditions could he set in the furnace. the extracted flue gas from the probe was transferred via polvtetrafluoroethylene (PTFE) tubing through a series of filters and water traps for cleaning and drying purpose. the flue gas was later cooled to 2 °C by passing it through a chiller. the filters were frequently r replaced along with cleaning of water traps in order to prevent any blockage of the gas sampling system.

r The tlue gas wus passed through a manifold that directed the sample gas to different gas analysers. The (0 oxygen in the flue was measured by passing a part of the sample over the self-indicating silica gnl. The gas sampling probe used to draw in the flue gas from frirnace was attached with compressed nitrogen supply for purging to prevent any blockage during operation. On-line gas r-alysis systems monitor 02, C02, CO, and NC) N0) anti temperatures dowi the furnace are monitored and logged to PC during each test period.

the additive was fed with different types of coal to the furnace in mass fractions from 1,3 tc 13%. Fly ash solids were collected by the fly ash catch pot connected to a cyclone separator. The samples and emissions were collected and measured after attaining steady state condition for each test. Fly ash samples were analysed for loss on ignition (1.01) in a muffle furnace by drying at 105°C for one hour followed by heating at 850°C for 2 hours. In boilers generally the LOt value is equated to un-burnt carbon.

N0 Emissious Fuel bound nitrogen contributes to about 8OYo -95% towards the NO formation in pulverized coal combustion. Fuel bound nitrogen during coal combustion is generally split into volatile-N and char-N, This division preferentially depends upon nitrogen content and volatility of coal along with the combustion conditions such as temperature, residence time, and heating rates. Tn the case of sub bituminous coals, the volatile-N comprising of tarry compounds decay rapidly to hydrogen cyanide (1 TCN) or soot-nitrogen. Whereas in contrast the low rank coals would preferentially release the light nitt-ogen species such as NH3. Combustion of nitrogenous species (NH3 and HCN) present in the released volatiles and oxidation of the char-nitrogen results in the formation of oxides of nitrogen.

However, the HCN or NH3 may also be reduced to N2 after reacting with the available NO. This depends upon the available stoichioinetric ratio near the burner, mixing of the evelved species in the furnace and fuel-N concentration, Figure 3 shows the effect of addition of both water cooled wq and air cooled (AC) fuel improver towards NO emissions. in all of the examples the NO emissions decrease as the n-lass fraction of fuel improver increases from 0 to 12%.

r Figure Ta illustrates a proposed schematic mechanism fur the interaction of the fuel improver with the coal, The process of NO reduction under un-staged combustion observed during fuel improver addition is associated partly with the interaction of the fuel improver particles with the coal matrix and volatiles as they are released from coal particles, resulting in cracking of the heavier hydrocarbons favouring the split of fuel-N into volatile-N. As shoxwi in Figure 4a the coal particles swell during the heating that occurs in the combustion chamber, but the fuel improver particles do not swell.

As the coal particles swell the fuel improver particles enter the coal matrix and enhance volatile hydrocarbon cracking.

For this mechanism to proceed the particle size distribution of the coal particles must include larger particle sizes than the particle size distribution of the fuel improver particles, in addition, the particle sizes of the fuel improver at the ioth percentile of the particle size distribution must he an order of magnitude smaller (10 dines smaller) than the particles sizes of the coal particles at the 90th percentile of the coal particle size distribution. F-hr example, looking at the particle size distributions for Russian coal anti Water Cooled Fuel improver Uaoth measured using the Malvern Mastersizer Scirocco 2000), the largest particles of coal are 631 micron in size, whereas the largest particles of fjel improver are 138 micron in size. Tn addition, thc particlc size of the fuel improver at thc 0 percentile, d (0.1). is 4.7 micron and the particle size of the coal at the 9O' percentile, d(O.9), is 179 micron, This means that the tipper I 094 of the coai particles arc 38 times the size of the lower I 01⁄4 of the fuel itnprovcr particles.

The smaller fuel improver particles are able to react with the larger coal particles as shown in Figure 4a.

this mechaiism favours rhe NO reducrion pathway rowards N2 formation rather than NO by oxidation, since this thrin of tuel-N is easier to control in the fuel -rich zones of the flame, as shown in Figure 4b which illustrates the NO reduction chemistry pathway in the presence of fue' improver.

Figure 4c illustrates how the ftel improver enhances the thermal degradation of Fuel-C (coal) to volatile -C (lighter hydrocarbons). The lighter hydrocarbons are less likely to form Char-N.

Figure 5 illustrates a two stage tixed bed reactor I used to carry OUt a laboratory test. Nitrogen was used as a product carrier gas. A coal sample 2 was pyrolysed in the first reactor 3. The derived gases were rcfbrrned in the sccond reactor 4, where the fucl improver 5 was placcd. Products after the second-stage reaction were condensed by air and dry-ice in a condenser system 6. The non-condensed gases were r collected by a gas sample hag 7 and further analysed by gas chrornatograph (GC). Both stages of the (0 reactor were maintained at 950°C. with two grains each of fuel improver 5 and coal 2 in each of the stages. the gas and oil products wrc collected after an hour and l the products as and oil) were collected for analysis. It ws found that gas production increased in the presence of friel improver (see table 4). Similarly oil yield decreased, indicating the convcrsion of heavier hydrocarbons into more of gaseous fracñons. The colour of the oil produced from coal was dark brown, whilst with fuel improver it changed to c1erner oil. the SLI1nflmrV of the experilnents with and without the friel irnprcvcr is tabulated

in table 6 below

Table 6: Gas/Oil yields and concentradon of gases.

I

Air Cooled Fuel Water Cooled Air Cooled Fuel Coal, d(O 9) < 75 Improver, ci(O.9) Fuel Improver, Improver, d(O.9)<5Oim d(O.9)<35!.tm Ga \leld (\ ° 1437 21 97 1826 19(1(1 Oil Yield (t ° 1250 11 06 905 653 Residue Yield (xt 645(1 6583 6533 65 Mass balance (wt 91 37 9886 9263 91)86 Gaseous Compositions CO 16.50 18.14 16.54 18.80 + H 58.42 56.04 56.74 55.67 CO 3.54 4.88 3.87 4.38 CiT 19.64 19,09 20,80 19.15 (11(1184 0.0(1(15 0.0019 0.0018 (2 00004 00001 00001 0000! Total 1(1(100 10)1(1') 10)) 0)) 10000 r r The results suggest gas production and hydrogen concentration increased in the presence of the fuel improver, it is concluded that the gas yields increased by around 5-7% in presence of both types of fuel improver.

Similarly, as shown in Figures 6a and 6b, the increase in the concentration of CO and hydrocarbon from coal has also been confirmed using thermogravimetric analysis coupled with Fourier transform infrared (ETIR) spectroscopy by blending 5% to 33% weight proportion of AC fuel improver n coal.

The combustion of the coal char and the behaviour of the unhlended fuel improvers were investigated using a Stanton Redcrofr 10782 thermo-gravirnetric analyser (i'GA connected to a Nicolet 2vlaa 560 FTTR spectrometer via a heated interface and heated transfer line. In this study, the FTir spectrometer was operated in scan range of 400-4000cm-I and a spectrum was taken every 45s during the course of the TGA run, with 100 background scans taken prior to the run to correct for ambient moisture and carbon dioxide. The transfer line was maintained at 170°C. while the TGA interface cell was held at 300°C, with a constant purge of nitrogen around the cell to minimise the effect of changes in atmospheric moisture and carbon dioxide levels. The intensity of absorbance in the wavenumber range 2000-2500cm-i, which in this study corresponds to the relative concentration of carbon monoxide and carbon dioxide in the sample gas during char combustion, was plotted against time. In addition, the intensity of wavenumber ranges 2170-2180cm-i (carbon monoxide only) and 2800-3200cm-i (a variety of C-TI bonds, such as hydrocarbons, released during coal pyrolysis) were plotted from the spectral series data.

The onset of CO release is-ata temperature of about 300°C reaching a plateau at under GOC°C.

CO release peaks at about 900°C followed by a reduction then complete burnout when the 02 containing mixture is introduced. As shown in Figure Ga, there appears to be an increase in the CO produced when the fuel improver is present compared to that measured from coal alone. Elution of hydrocarbons during the coal pyrolysis/combustion tests was followed by FTIR and the results ate shown in Figure Gb.

Hydrocarbon release from the coal (under N2) begins at about 300°C and teaches a maximum at about 500°C and is complete by about 900°C. The concentration of hydrocarbons in the presence of the fuel mprover were higher than that from the coal alone, there does not appear to be significant difference in r (J_) and hydrocarbon concentration measured between the amount of tuel Improver in the coal-fuel

CD

improver blends.

Ihis increase in the gas yield supports the hydrocarbon cracking and release producing more of \rcJlatile which in turn facilitates the NO reduction into N2 as described with reference to Fiires 4a and 4h.

Moreover, the presence of iron oxide in the fuel improver would also interact with coal to result in additional NO reduction reactions supplementing the existing pathway towards N2 forn2ation, Fe2O3 can he reduced to Eb in presence of CC), and later on NO can oxidize iron to reproduce *203, Ike

summary of reactions is as follows;

3C0 ± FeOs -* 3C02 ± 2Fe 2Fe + 3 NO -N2 + Fe203 The net algebraic addition of reactions yield CO + NO -CO2 + -N2 Three different types of coals investigated for the study with a view to observing any vat-iation in the behaviour of additivc on NO reduction. the medium ash Russian Coal (RC) and high ash Kellingley Coal (ICC) resulted in slightly highet-t-eduction in NO as compared to Columbian Coal (CC) because of relatively higher volatile matter and lower fixed carbon compared to CC. lie air to fuel ratio in the con'ibustion test facility (CTP) was set at 20% excess-ir levels (stoichiotnetric ratio of 1.20) for un-staged flame firing condition. The optimum range up to 13% by weight of that of coal input was observed for both types of fuel improver. NO reduction of 15% & 16% for 13% & 12°/b mass fractions of WC and AC fuel improver were observed for RC, respectively \Vhereas, 11% & 10% NO reduction was achieved r for 110 bo and 13° ma fi action of \\ C and AC fuel impi oi ci uth CC repecti eh ICC ith \ C and AC r fuel improver co-firing resulted in 14% & 15% reduction in NO for 10% and 13% added mass fractions, C'J respectively.

In general the following mechanisms can be summarised towards reduction of NO emissions using both types of' f'uel improver.

* Tt is associated partly with the interaction of fuel improver particles and coal matrix, resulting in cracking of the heavier hydrocarbons favouring the split of fuel-N into volatile-N. the increase in the gas yield supported by the higher concentration of CO and hydrocarbon would favour the NOX reduction pathway towards N2 formation rather than NO by oxidation, since volatile part of' N is easier to control in the fuel -rich zones of the flame.

* Ihe fuel improver, having higher surface area because of its finer particle size distribution compared to coal, would facilitate the thermal degradation of heavier hydrocarbon into lighter hydrocarbons and these lighter hydrocarbon are less likely to form Char-N.

The presence of iron oxide in the additive would also intet-act with coal to result in additional NOX reduction reactions supplementing the existing pathway towards N2 fbrmation.

Figure 7 represents the effect of change of stoichiometric ratio near the combustion zone on different co-firing blends of Euel inprover with kG, CC and KG. The in-furnace air staged combustion creates fuel rich zones due to the delayed mixing of fuel particles with air resniting in the abatement of NO. The reduced stoichiometric ratios i.e. 0.8. 0.9 in primary combustion zone restrain coal combustion, and large amount of unburned char enters the burnout zone resulting in poor carbon burnout.

The addition of Fuel improver resulted in an additional impact on increase in NO reduction with decreasing air to fuel ratio, WC Fnel improver with kG resulted in a range of 4.6% to 258% reduction in NO) fbr range of 0.9 to 1.21) stoichiometric ratio. Whereas, a range of 4.7% to 23.9% was observed for WC/AC Fuel improver with CL for 0.8 to 1.16 changing air to the1 ratios, AC/WC Fuel improver with TKC fbr 0.9 to 1.33 stoichiometric ratio resulted in 7.3% to 31.1% reduction in NC) with respect to coal staged flame base lines. r

Effect on LOT (1\J Use of the the1 improver results in a substantial tinprovement in the LOT values of the fly ash of all three types of coal tested, as shown in Figure 8. The presence of the fuel improver has increased the hydrocarbon intensity and gas yield conversion from coal, \vhich in turn intensifies the combustion and results in improved TOT. Tn the case of Russian coal, an overall net reduction thr the LOT in the range of 19% to 63°/o for T,3°/o to 13% mass fraction of added fuel improver was achieveef Similarly, a LOT reduction in the range of 20% to 70% was found fbr addition of 2.5% to 11% mass fractions of fuel improver for Columbian coal and a LOT reduction in the range 64% to 70% xvas found for 5% to 13% mass fractions of added fuel improver fbr Kelhinglev coal. The optimum mass fraction of fuel improver tanges from 5% to 12%. resulting in fly ash having less than 7% LOT that can be used in cen2ent mnanufacturilig.

The particle size distribution (PSD) of resultant coal fly ash can potentially fluctuate depending upon the operation of the power station. Typically power stations are operated under a steady load to compensate fbr variation of the resulting fly ash. I Towever. the PSD of the coal fly ash is also important when considering its use in concrete manufacturing General purpose cement utilizes finer ash because finer ash is more reactive. The strength and water content of the resulting concrete is also dictated by the variability in the fineness of die fly ash.

Ely ash fineness is usually measured as the % retained on a 45 micron sieve. British standard ES EN 450 which governs particle sizes of fly ash suitable for mixing with cement states that the fineness of the fly ash must he «= 40% retained on a 45 micron sieve.

The fuel improver has been found to improve die fineness of the resultant fly ash as shown in Figure c. Figure 9 shows a comparison between coal fly ash resulting from burners both with and without fuel improver present. It can be seen from the graph that with addition of fuel improver (added at either 4.2% or 4.8% mass fraction) the fineness of resulting fly ash increased by about 36% to 85% in addition to that of RC and KC fly ash baselines, respectively Using the fuel improver of the invention the fineness of the fly ash is typically «= 20-30% retained on a 45 nucron sieve.

Table 7 below illustrates compression strength tests carried out on cement cubes manufactured from fly ash from a burner burning just RU and a burner burning RU and 4.2% fuel improver. The tests r were performed by(1 major cement manufacturer in the UK. it can he seen from the table that die cube (Q strength results are reasonably close to each other without any major variation in the strength characteristics of the cement mixfttre, the addition of the fuel improver therefore results in an equally comparable strength mortar when prepared by mixing 30% of coal fly ash from a fuel improver/coal blend xvith Portland cement.

If the fly ash is to he used in cement manufacture it is beneficial that the fuel improver includes calcium oxide since the calcium oxide improves the cementitious properties of the tly ash. The fly ash may be added directly to the calciner during cement manufacture, or just mixed in with Portland cement to produce a ready mix/Portland fly ash mixed cement.

Table 7: Compression strength tests on RC fly ash, with and without fuel improver.

Mortar with 30% of RC fly ash Mortar with 30% of RC + 4.2% and 70% Portland cement fuel improver fly ash and 70% Portland cement Age in Days Cornprcsion Densit Compression Densit Strength (N/ g/m Strength J/' (jcg/m mm mm + 1 10.6 2340 10.5 2375 3 26 2352 25 2366 7 33.4 2358 33.7 2374 14 39.9. 2372 38.8 2388 28. 42.4 2357 4(1.5 2377 56 49.6 2359 2398 Effect on Temperamre The addition of fuel improver to the burner has been found to result in increased temperature measurements within the burner. The generation of extra temperature is due to the bunung of the additional carbon of the coal feed, favouring the split of carbon into light volatiles rather than remaining in the char, Figure 10 illustrates the temperature differences calculated at axial distances downwards from r r the burner, with TI being dose to the burner and T7 near the flue section. The values were calculate fbr (0 different mass fractions of fuel improver for all the studied coals against the corresponding coal baseline temperature measurements. As indicated from the general trend fbund in Figure IC, the different mass frachons of fuel improver produced a broad range of 12-30 C increase in temperatures at Ti, These changes in TI values are dependent upon the added mass fraction of fuel improver. The increase in the temperature also supports the improved values of LOi.

Effect on Slaeein and l'bulin Shagging and Fouling characterizes the deposits on the radiant section of the boiler and heat recovery section, respectively. These deposits are fbrnmd through a series of complex mechanisms, forming a variety of compounds which cause corrosion and reduction in heat transfer rates, Slagging and fouling indices are used for the assessment of the propensity of' fuel ashes in form these deposits. These indices have been specifically developed for the assessment of coal ashes only, but these indices are widely used in literature for cc-fired fuels as well. Ihe most conrnmonly used traditional indices used to calculate the ftiel ash deposition tendency are shown in Figure ii. The predicted composition is calculated as a mass average of the metal oxides present in the known feed rate of coal and fuel improver. The actual ash samples collected during these combustion tests were used to measure the ash components and were reported as measured values in the table shown as Figure 11.

The pt-edicted values of metal oxides are in close proximity to the actual measut-ed concentration of metal oxides, the existing difference between the values is expected due to the ± l-2°/o combined variation in the actLral feed rates of coal and fuel improver. I lowever, irrespective of the predicted and calculated indices, there is an insignificant increasing trend in both the predicted and measured values of fouling and slagging indices, when compared with the metal oxide concentration of individual coal fly ash samples. The reported chemical composition of the fly ash samples show a narrow range of variety of alkali oxides between coal fly ash and coal fly ash plus fuel improver samples. A dominance of Si02, A1203 and Fe203 was found in all the fly ash samples. This is partly due to the inherited concentration of these oxides in the actual coal and fuel improver, the percentage of oxides of iron was found to have increased in the fly ash mix, whereas the percentage of alumina concentration decreased, slighfly.

r c;ererallv. a substantial increase in the percentage concentration of NazO and lKO results in higher fouling propensinr in commercial boiler, The %age of K20 is relatively higher h RC and ICC fly ash as compared to the fuel improver, hence addition of the fuel improver delivers an overall positive impact towards lowering fouling propensities. Moreover, the tabulated overall measured concentrations show an insignificant variation concluding trivial effect on the actual boiler furnace wall. Moreover, the measured values of Rm/A) were less than 0.75 indicating that ash flow temperature will he higher resulting in a decrease of slagging tendency.

The addition of the fuel improver delivers an overall positive impact towards lowering fouling propensities for the fuels wInch have relatively higher O, of Na:() and K20. It can also result in increasing the ash flow temperature resulting in decrease of slagging tendency depending on the type of fuel (coal).

the formulae used to calculate the fouling and slagging indices in the table shown in Figure 11 are as follows: B (Fe,O. + laO +AigO ÷Va,O + K,O / . A 5O, +AI,O, +iO, B/ A) is the base/acin ratio: -Eu, the Ebuling index is calculated as follows: Eu = + K,O) Where Fu 0.6 there is a low fouling inclination; where Pu = 0.6-40 there isa high fouling inclination; and whei-e Pu »= 40 thei-e is an exti-ernely high fbuling inclination.

Rs, the slagging (l3abcok) index is calculated as fbllows: Rs = I -IS A) r Where S is the mass percent of total Sulphur in the dry the1.

\Vhere Rs < 0.6 there is a low slagng inclination; where Rs = 0.6-2.0 tjrere is a medium slagging inclination; where Rs = 2.0-2.6 there is a high slagging inclination; and where Rs > 2.6 there is an extremely higli slagging inclination, R(/A;. is a si nplihed basic to acidic ratio calculated as fbllows: B EeO+CaO+A'IgO Slrnplltled -=R = 2 A Si02 +A120, Rh = Pe206 + CaO + MgO + A/a20 + K20 The amount of fuel improver can he controlled in relation to the type of coal being burnt to ensure that the slagging and fouling indices are at the desired levels, and below the extremely high levels.

For coals with poor slagging and fouling indices larger amounts of fuel impover may be added in order to bring the basic to acid ratio and/or the simplified basic to acid ration (Rav.) back into the desired range.

Effect of Fuel Trnprover on Efficiency of an Electrostatic Precipitator Exhaust stacks of particulate hydt-ocarbon fired boilers typically include electtostatic precipitators. it has been found that by using the fUel improver of the invention the efficiency of these electi-ostatic precipitators can be increased resnlting in a greater vicld of fly ash, which is beneficial for two reasons. First, less dust is emitted to the atmosphere. Second, die fly ash resulting from die process of the invention is useful as a po>n'.olan.

Table 8 below and the graph of Figure 12 show the resnlts of dnst collection at east, central and west legs of a coal fired steam prodncing boiler (230MWi1) operating at (205 tons/hour of steam) during a 4 honr trial. The outlet legs of the boiler pass throngh an electrostatic precipitator situated at the outlet of the boiler. The boiler was fired with 23 tons/hour of milled coal. 6.8Yo by weigiit of coal was added to the coal and this is shown by the vertical line intersecting the time axis approximately mid way between 10.36 and 1126. The second vertical line intersecting die time axis approximately mid way between 1445 r and 15.35 marks the point where die introduction of the fuel improver ceased.

(0 hi another shorter test lasting 45 minutes using the same boiler hut operating at (252 tons/hour of steam) and fired with 30 ton/hour of coal, when 9% of water cooled fuel improver was added the reducticn in dust conccntratiou in the exhaust dowustream of the electrostatic precipitatcr measured in the west, centre and east legs was 13%, 4.5% and 20% respecflvely when compared to the coal base line.

Table 8

Dust Concentration post East Leg of Boiler (rng/ Centre Leg of Boiler West Leg of Boiler (mgi eleclrostalic precipilalor ni3) (rngim3) rn3) Coal baseline 241.0535 262.2579 145.9069 Coal + 6,8% Air Cooled Fuel Improver 206.2909 224.5522 122.3604 % reduction Tn dust concentration 14.4211 14.37735 16.13808 It can be seen that the dust concentration post the electrostatic precipitator is iess where the fuel improver is added, even though the combusted mass is increased by 6.8% (the amount of coal entering the boiler's furnace is nor reduced when the fuel improver is added).

this reduction in dust concentration post the electrostatic precipirators is explained by a number of factors attributed to the fuel improver. First, the improvement in 1,01 discussed above means that the fly ash has a lower content of carbon. Electrostatic precipitarors function better where carbon concentration in the fly ash is lower. Second, the fuel improver has electrica1y conductive propertics; i.n parimcular the oxides of iron within the fuel hprovet-and the presence of faylite and magaetite phases of the fuel improver post combustion in the fly ash enable the electrostatic precipitators to remove more material as fly ash rather than dust entrained in the flue gas.

References to relative amounts in compositions are percentages by weight The process of the invention provides for improved combustion efficiencies improved slagng and/or fouling in addition to reducing emissions and improves the po'zY.olanic pnperties of fly ash. r r (0 (4