Вы находитесь на странице: 1из 57

PF Power Plant

Boiler Efficiency
Boiler Types
Mills
Burners
Total pulverised fuel (PF) related costs provide the means to directly compare alternative coals. These costs

 Annual fuel costs - consist of the FOB mine fuel cost and fuel transportation cost.
 Maintenance costs- associated with maintenance of the steam generator system, air heaters,
system, FGD system, waste conditioning system, and coal handling system.
 Plant operating costs - consist of scrubber additive costs, fixative costs, and waste disposal co
 Replacement power costs - due to unit derates, unavailability and changes in auxiliary power r
For further details on the impact of coal quality on power costs see Conroy & Bennett

Boiler Efficiency
The theoretically achievable maximum efficiency of thermal power generation is limited by a simple relation

The higher steam temperatures of Super-Critical (540°C to 600°C)& Ultra-Super Critical (650°C to 720°C)
now being built. Research into suitable materials for ultra-super-critical boilers is currently been done in Ch

Boiler Types
Three different types of pulverised fuel-firing systems are used in large boilers:

 Horizontal or slightly angled swirl burners located in the front and/or rear walls of the furnace. In a sw
flame.
 Tangential corner-fired burners introduce the fuel and primary air at a tangent to an imaginary circle
 Vertical or downshot firing burners are located in the roof of the combustion chamber with the flame p

Mills
In a PF boiler system, coal is pulverised to typically 70–75%, passing 75 micron and entrained in preheated
Grindability Index (HGI) is used to predict mill performance.

The three types of coal pulveriser are generally identified by the speed of their rotation:

 Low speed mills are of the ball/tube design with a large steel cylinder and a charge of hardened balls
 Medium speed pulverisers are typically vertical spindle mills that grind the coal between rollers or ba
 High speed mills have a high-speed rotor, which impacts on and breaks the coal.
The table below shows the preferred coal properties for each type of pulveriser.

Pulveriser Type Low Medium High


speed speed speed
Tube Vertical Impact
Example mill spindle mill
mill
Coal feed top mm 25 40 32
size
(as) 0–10 0–20 0–25
Coal moisture
%
(as)
Coal ash 1–50 1–30 1–15
%
Coal quartz (as) 0–10 0–3 0–1
%
content
Hardgrove 30–50 40–60 60–100
Grindability
Index 80–100

Abrasion mg/kg 50–100 10–60 5–30


index
Vertical spindle mills are commonly used in large-scale power plants and to pulverise coal for injection into
Burners
Conventional Low NOx burners

Low NOX burners are modified swirl burners that create a fuel-rich combustion zone followed by a leaner bu

Advanced Low NOx burners

These burners create a hot flame envelope around the fuel rich zone near the burner outlet. This flame env
burners have improved NOx reduction and better flame stability characteristics than the conventional Low N

Top
Fluidised Bed
A major factor promoting development of atmospheric fluidised bed combustion systems
(AFBC) is their ability to fire high ash coals, retain sulphur within the bed through the
addition of limestone and low NOX emissions. The main types of fluidised bed
combustors (FBC) used for large-scale plants are known as bubbling beds and circulating
beds.

Operating conditions for fluidised beds are shown in the following table.

Bubbling Circulating
beds beds
Top feed
50 10
size (mm)
Bed particle
0.1–4.0 0.1–0.8
size (mm)
Fluidising
velocity 1–3 4–12
(m/s)
Bed
750–
temperature 750–1000
1000
)
In a bubbling fluidised bed the gas velocity is increased until the whole bed becomes a
turbulent mass of solids and bubbles, but there is no carry-over of bed material with the
combustion gases.

When the gas velocity is increased further, some bed material is entrained in the gas stream
and leaves the combustor. The bed material is separated in an external cyclone and most of
the solids are returned to the combustor. These fluidised beds are known as circulating or
fast fluidising systems.

Pressurised fluidised bed combustors (PFBC) are based on fluidised bubbling technology.
This technology gives the same fuel flexibility and reduced emissions as AFBC, but with
enhanced thermal efficiency and greater heat input for the same bed area. The thermal
efficiency enhancement results mainly from expanding hot flue gas, at pressures in the
range of 10–16 bar, through a turbine. The greater heat input possible with PFBC means a
smaller unit, in terms of physical size, can be built for a given generating capacity, reducing
construction time and costs. Hot gas cleaning is needed before the turbine and is a major
development issue.
Integrated Gasification Combine Cycle (IGCC)
Interest in the integrated gasification combined cycle (IGCC) process is based principally on
its potential environmental benefits. When the gasifier is supplied by oxygen instead of air,
IGCC plant offer the potential to capture relatively pure carbon dioxide (without nitrogen
dilution) that could then be sequestered to virtually eliminate greenhouse gas emissions.
There are several gasification processes in operation as well as in development. These can
be broadly divided into three main categories: fixed bed, moving bed (e.g. fluidised bed)
and entrained flow processes. The table below summaries the typical properties for several
gasifiers.

Lurgi Winkler Entrained – Phase Gasifiers


Properties
Moving Bed Fluidised Bed

Koppers -
Texaco
Totzek

Non-caking to Non-caking to n/a Preferably non-


Caking
weakly medium caking
properties
caking coal caking coal

Particle Particles 0–8mm Grinding to Grinding to 90%


between 6 and 90% < 0.1mm < 0.5mm
size
40mm (bituminous 50% < 0.09mm
required; no coal) or
fines
80–85% <
0.1mm (lignite)

Ash % db Up to 40% Up to 50% Up to 40% Up to 40%

Ash fusion Deformation Deformation Flow Flow


temperature temperature temperature temperature
below below
above markedly
temperature in above 1500°C 1300°C
gasification
combustion
zone temperature
Moisture Up to 40% Up to 8% 1–2% Equilibrium
(bituminous moisture less
coal) or
than 4%
8–10% (lignite)
before
grinding

Sulphur n/a n/a n/a n/a

Shatter High High n/a n/a


index

HGI n/a Greater than 40


Greater than 40

Lignite - HV Lignite - HV
Rank All All
bituminous bituminous

Volatile n/a n/a High High


matter

In IGCC power plants, the coal is gasified and the off-take gas is combusted in gas turbines
to generate electricity. The hot combustion gases are then passed through boilers to
produce steam for further power generation or other industrial use. Thermal efficiencies of
45% higher heat value (HHV) or greater have been achieved in such plant. Several full-
scale demonstration plants are operating commercially in the USA and in Europe.
Integrated Blast Furnace
Cokemaking
PCI
Sinter Plant
The main use of coal within the integrated steelworks is the production of coke. The biggest portion of blast
production is represented by merchant coke production plants.

An increasing amount of coal is used for pulverised coal injection and in the production of sinter.

Cokemaking
Blending of coals with different plastic properties is extensively carried out prior to cokemaking. The blend
1100°C.

Coke is the critical fuel of integrated iron and steel plants It must possess certain properties in order that b
shapes give high voidage (hence permeability) to coke in layers. It must also have a high internal porosity

The concept of a coke plant with direct combustion of the raw gas would omit the entire gas treatment plan
operation in the USA, Australia and India. One plant in the USA is a heat- recovery coke plant that produce
some examples of stamp charging of slot ovens.

Coke production is mainly carried out in conventional slot-ovens.

In a conventional slot oven coking converts the coal to a higher carbon content solid which contains all feed
the steelworks as a fuel. Conventional coke ovens (slot type ovens) with raw gas recovery have reached di
At the same time, the emissions from batteries and gas treatment facilities have been reduced to the lowes
electricity. The energy balance of a conventional coke oven is shown below. The yield, size distribution and
The need to improve environmental controls for existing cokemaking facilities and to find more cost- effecti

 The European Jumbo Coking Reactor has reconfigured batteries for larger individual batch process ove
in utilizing the technology is minimal.
 Non-recovery cokemaking is a proven technology derived from the Jewell- Thompson beehive oven de
products of combustion to high-pressure steam for plant use and for conversion into electricity.
 The Coal Technology Corporation is using a formcoke process that produces coke briquettes from non
 The Japanese SCOPE21 project, still in its early stages of development, is using a formcoke process th
project is being developed as part of an eight-year research program.
Nicol & Durie review major changes in cokemaking and blast furnace steelmaking.

The Chair of Iron and Steel Metallurgy of the Department of Ferrous Metallurgy (IEHK) at the RWTH Aachen

PCI
The current and future needs of blast furnace operators are to maintain a stable and productive blast furna
the impact of quality of the injected coal has increased, there has been a shift from high volatile thermal co

The main criteria used to measure the performance of a coal for injection are:

 Economic Benefit: The main economic benefit is the replacement of high cost coking coals.
 Milling and Handleability: The main operating costs, other than coal costs, are related to the mi
 Blast Furnace Operation: The injected coal quality can influence the quality of the hot metal, pr
The better replacement ratio and better milling performance of low volatile coals makes them the preferred

Blast furnace stability impacts on the productivity of a blast furnace. Many aspects of blast furnace operati
issues relating to coke fines and slag viscosity have a greater impact on blast furnace performance. Curren
accumulate in the deadman having an adverse effect on lower zone permeability.

The generation of coke fines is a function of coke quality and the blast through the tuyeres. Due to increase

In the current times of high steel demand all steelworks operators a striving for high productivity. High pro

 improvement of the permeability in the upper and the lower zones of the BF and
 reduction of the specific gas consumption while acting on the blast conditioning and by decreasing the
Permeability with a BF is strongly influenced by the PCI rate and the properties of the injected coal, as show

Sinter Plant
The sintering process agglomerates the fine material into a clinker-like aggregate, with a size range that is
blended fines, recirculated sinter fines, crushed coke breeze or anthracite (to assist ignition and the propag
Sintering is carried out on a continuous travelling grate ("the strand") which travel over a series of wind bo
ignited under a gas-fired radiant-hood ignition furnace, producing a flame front, which is drawn downwards
the penultimate wind box.

Temperature in the ignition hood ranges from about 1,150°C where ignition begins to about 800 C at the ex
tonnes/m2/day of grate area are expected, and rates in excess of 49 tonnes/m2/day have been attained. T
Direct Reduction & Smelting Processes
COREX
HI-Smelt
Fastmet
A summary of the new technologies that will help meet the expected demand for steel is given in attached
uses coke ovens, sinter/pelletising plants and blast furnaces.

Direct Reduction

Increased use of EAF for steel production will increase demand for scrap. This will lead to a shortage which
from them by a reductant.

HBI is a briquette made by a densification process whereby the direct reduced iron (DRI) feed material is a

Direct Smelting

A number of Direct Smelting processes for the production of hot metal from iron ore or DRI are being devel


o the ability to utilise cheaper and more abundant raw materials such as non- coking coals and non
o smaller economic plant sizes and,
o reduced capital costs and environmental problems through the elimination of coke ovens and sint


The benefits are:


o competitive hot metal,
o flexibility of operation, and
o the product can be fed to an EAF for steelmaking.


The metal produced from such processes is a premium grade product which has a high value to an EAF stee

At this stage the only direct smelting process that is in commercial operation is the COREX ®
process. Howe

COREX
The predicted increase in steel demand in Asia will provide opportunities for the establishment of newer steel making techn
Furnace (EAF), and direct smelting processes. The smelting reduction processes do not require the use of coking coal or sc

A number of Direct Smelting processes for the direct production of hot metal from iron ore or DRI are being developed aro

 the ability to utilise cheaper and more abundant raw materials such as non-coking coals and non-agglomerated ores,
 smaller economic plant sizes and,
 reduced capital costs and environmental problems through the elimination of coke ovens and sinter/pellet plants.
The benefits are:

 competitive hot metal,


 flexibility of operation, and
 the product can be fed to an EAF for steelmaking.
The metal produced from such processes is a premium grade product which has a high value in use to an EAF steelmaker c

At this stage the only direct smelting process that is in commercial operation is the COREX process with plants operating i
generated from the gasification (and partial combustion) of coal. The reducing gas produced in this second stage is used in

The coal used in the COREXprocess must supply heat for the devolatilisation of the coal, the gasification of the char and t
from 55 to 70%. Many coals from the Bowen Basin within Queensland are suited to the COREXprocess and several mines

This is to ensure good gas flow within the melter gasifier and to reduce the carry over of coal and/or char. The amount of fi

There are many alternatives to achieve minimum fines content depending on the facilities at the loading port, discharge por

HI-Smelt
The HIsmelt Process has its origins in the early1980's when Rio Tinto Limited identified the potential to ada
tonne K-OBM converter. After successfully proving the concept, a small scale pilot plant was constructed an

Whilst the process scale-up was successfully demonstrated, the complexity of the engineering, poor plant a
these issues. The horizontal Smelt Reduction Vessel (SRV) was replaced with a vertical vessel. This allowed
thereby simplifying the engineering and overcoming operational issues encountered with the submerged bo

The plant achieved very high plant availability whilst still maintaining high process performance despite the

Fastmet
The Fastmet process is a solid reductant based direct reduced iron process using fine ore mixed with a carb
of virgin metal. Kobe Steel commissioned a demonstration plant in 1995.

In this process, a combination of pulverized coal and iron ore fines are made into pellets.

The pellets are fed into a doughnut-shaped rotary hearth furnace and heated. At 1,350 degrees Celsius, the
converted to briquettes for merchant shipment. Kobe Steel, Ltd. and Midrex Direct Reduction Corporation re
research on the new molten iron making process in three stages. In Phase 1, running from July 1996 to Ma
bench-scale testing, investing some 500 million yen. Should the results of the first two phases be promising

The products produced by the rotary hearth processes FASTMET, INMETCO, and IDI contain large quantitie
because they are either too costly or they degrade the coal's properties. Improved methods for separation
Cement
A wide range of coals are used in cement plants. As is the case with many coal applications,
the plant can be designed in part to suit the coal. Although coal the quality criteria are not
very exacting, it is important that variability in quality be small, as the composition of the
raw feed must normally be tailored to the coal ash chemistry. As the rotary kiln is by the
most common cement plant in operation the following discussion focuses on this type of
plant.

In direct firing systems the coal is milled on-line with the pulverized coal being directly fed
to the kiln burner. The quantity of primary air used with direct firing systems is normally
that required to dry the coal and sweep the mill, and is more than required for transporting
the pulverised coal or producing a satisfactory flame.

Direct firing systems are not particularly suited to high moisture coals because:

The moisture which is evaporated during milling is fed into the kiln with the primary air.
This has the effect of lowering the flame temperature, and process efficiency.

Additional mill air may be required so that the coal can be effectively dried without having
excessive air preheat. This additional air also enters the kiln as primary air. Excessive
primary air also has a similar effect to moisture.

In indirect firing systems the coal is milled off-line and is stored in a bin from which it is fed
to the kiln burner. High moisture in coals can be associated with low rank and the
propensity for spontaneous combustion. This could mean problems when storing the PF in
badly designed bins of an indirect fired system. If coals which are subject to spontaneous
combustion are to be used successfully, they may need to be sold selectively to plants with
a proven track record in this area.

Milling behaviour of coals for cement kilns has special significance because kiln operators
generally place some emphasis on tailoring the coal fineness to suit its reactivity. The
accepted measure of coal particle size in the cement industry is the percentage greater than
90ìm. It is widely held that low volatile matter in the coal can be compensated by finer
grinding. In order for this option to be open the mills must have the spare capacity to
achieve finer grinding, or the coals must have a high Hardgrove Grindability Index (HGI).
The trade-off between HGI and VM is therefore important. Fortunately, low VM bituminous
coals tend to have high HGI values.

The main consideration for coal storage is the propensity for spontaneous combustion and
explosion of stored PF. If fresh PF remains in one spot, particularly in the presence of hot
moist air, self heating and ignition can occur rapidly. This causes any dispersion of coal dust
in air, within certain concentration limits, to ignite and explode.

In order to heat the clinker to the required temperature of around 1500 °C, it is necessary to
have a flame temperature of around 1700°C. This is achieved by providing preheat to the
secondary air and limiting the quantity of primary air. As noted above, for high moisture
coals in direct fired systems there may be an excess of moist primary air and the required
temperature may not be obtained. However for indirect firing systems using dry primary air,
a satisfactory flame temperature can be achieved even with low energy coals.

The correct distribution of temperature along the kiln requires that the flame front be
located close to the discharge end of the kiln, close to the burner. Coals which are
unreactive and slow to ignite may therefore cause problems in some kilns, and in these
cases medium to high volatile coals would normally be specified. The use of swirl burners
with low volatile coals is not always successful, as the diverging jet may impinge on the kiln
walls just as the coal is igniting. The reducing conditions that this causes in the clinker
affects product quality, and damages the refractory lining of the kiln.

While it is normal to have some build up of deposits in the clinkering zone of a cement kiln,
excessive deposits in this area, called "clinker ring", can hinder the movement of solids
through the kiln and in extreme cases require plant stoppages. A second problem is the
formation of deposits in the suspension preheaters. These problems are commonly
associated with the presence of chlorine, sodium, potassium, and sulphur. When the levels
of these elements are considered, the combined effects of coal ash and raw feed
composition must be considered. The major source of sodium and potassium would
normally be the raw feed, however coals with unusually high sulphur or chlorine could
contribute to deposit formation and would not be favoured as cement kiln coals.
Plastic Properties
Plasticity refers to the melting and bonding behavior of the coal and

 is an indication of the initial softening, chemical reaction, gas liberation and


resolidification process within the coke oven.
 is an important requirement in the coke blend and is required for end product coke
strength
 the fluidity of the plastic stage is a major factor in determining what proportions of a
coal is used in a proportions of a coal is used in a blend
During the heating of coal an unstable intermediate phase, called metaplast, is formed after
the moisture is driven from the coal. The metaplast is responsible for the plastic behaviour
of coal. On further heating a cracking process takes place in which tar is vaporized and non-
aromatic groups are split off. This cracking process is accompanied by recondensation and
formation of semicoke.

When a coal/blend is coked in slot-type ovens, two principal layers of plastic coal are formed
parallel to the oven walls. They are linked near the sole and the top of the charge by two
secondary plastic layers forming an envelope of plastic coal. As carbonization proceeds, the
plastic layers move progressively inward eventually meeting at the oven center. It is within
these plastic layers that the processes which result in particulate coal being converted into
porous, fused semi-coke take place. The semi-coke undergoes further devolatisation and
contracts which results in fissures in the final coke.

The only small-scale methods which have stood the test of time and have been accepted as
standard plasticity tests are the crucible swelling number, Gray-King coke type, dilatation
characteristics, Gleseler plasticity and, in some countries, the Rogas index and the
Sapozhnikov test. All of these are essentially empirical in nature and many are subjective,
at least to some degree. The Gieseler test is the only one which attempts to measure the
actual extent of the plasticity of fluidity attained. The Gieseler test is used to characterise
coals with regard to thermoplasticity and is sometimes an important tool used for coal
blending for commercial coke manufacture. Several authors have published methods to
predict coke strength using the plastic temperature range and/or maximum fluidity as
determined by the Gieseler test. The maximum fluidity determined by the Gieseler is very
sensitive to weathering (oxidation) of the coal.

Proton Magnetic Resonance Thermal Analysis (PMRTA) measures hydrogen protons and
therefore acts as a probe to the structure of, and of motions, molecular units within the
molecular lattice as the coal is heated to over 600 °C. A PMRTA determined parameter Fmax
is a measure of plasticity, the maximum extent to which a material has fused corresponds
to Fmax=100% and a solid material has a Fmax=0%. Fmax has been shown to be better
parameter than any Gieseler results in predicting changes in coke strength with weathering.

Steel and coworkers[ 1] combined high-temperature rheometry and 1H NMR to assess the
microstructural changes taking place during carbonization. They showed a relationship
exists between the logarithm of the material’s complex viscosity and the fraction of
hydrogen present in rigid structures for the resolidification region in which the material is
liquid-like with small amounts of dispersed solid.

Recent research conducted by ALS Coal has shown that the plastic layer behaves like a
viscoelastic solid and the research results were found to agree with simple theories of how
the viscoelastic properties may change with blend composition if each blend component is
treated as an individual viscoelastic solid. These findings have begun to identify of the
mechanisms that lead to coke strength, notably when coke strength is non-additive with
blend composition.

1 Karen M. Steel, Miguel C. Diaz, John W. Patrick, Colin E. Snape, "UNDERSTANDING THE
MICROSTRUCTURE OF COAL DURING CARBONIZATION USING RHEOMETRY AND 1H NMR",
2005 ICCS&T Okinawa - October 2005
Coke Yield
The coke yield and by product yield from any cokemaking process is a function primarily of
the coals from which they are produced and the carbonising conditions. The figure below
shows that about 64% of the as received coal is used in the blast furnace (BF).

As shown by this figure below there are two factors that influence the yield of blast furnace
(BF) coke:

 Yield of dry wharf coke, this depends mostly on the volatile matter of the feed
coal and can be estimated from -

 Yield of lumped BF - After the wharf coke is stabilised (usually by dropping from
a fixed height) it is sized to remove the coke breeze. The coke breeze is used
mainly as a fuel in the sinter plant. The yield of lumped BF coke after sizing
is dependent on the fission formation during the coking process and coke
morphology .
Low rank, high volatile coking coals yield predominately a very porous highly reactive
isotropic carbon with only a small proportion of fine textured anisotropic domains. Thus, the
proportion of edge carbon remains high and hence the reactivity is also high. The situation
improves gradually as the rank increases and in blends with prime coking coal, the
hydrogen donor capacity is high enough to effect some improvement in coke texture
relative to that expected from some lower rank components in a blend if there were no fluid
phase interactions. There are no significant interactions between higher rank, low volatile,
coals with prime coking coals in a blend with respect to texture development. The former
contribute to wall thickness and strength, as well as the coke yield and can also decrease
the pore size if the volatile yield is greater than about 12% and the there is some fluidity.
Coke Reactivity & Strength after Reaction
Coke strength is an indicator of physical strength of a coke made from a particular coal.
Coke needs to be strong to support the iron ore and coke mix above it in the blast furnace.

Reactivity and Strength After Reaction

The test is based on a procedure developed by Nippon Steel Corp in the 1970's as an
attempt to get an indication of coke performance and is used widely throughout the world
evaluate coke hot strength. The test studies the effects of one of the key coke reactions in
the blast furnace on the strength of the coke. In the test a 200g sample of 21mm +19mm
square hole coke (usually prepared by crushing 10kg of +25mm coke) is heated at 1100 C
under 1atm pressure of carbon dioxide for 2 hours. The coke is cooled under nitrogen and
the weight loss during reaction is determined. The percentage weight loss is known as the
reactivity (CRI). The reacted coke is placed in an I drum (no lifters) and subjected to 600
revolutions. The percent of material removed from the drum that is +10mm square hole is
known as the coke strength after reaction (CSR). The test procedure is shown the figure
below.

Generally a high CSR indicates a strong coke.

Reifenstein (2003) examined the standard, with particular reference to a set of variables
which were possible sources of variation within the coke reactivity test.
Coke tests
Mean Coke Size – the arithmetic mean size of the coke, determined by hand sizing the
coke over a specified series of screens. Generally, the larger the coke the better. A narrow
size distribution is also desirable.

Japanese Drum Test – a 10kg representative sample of the +50mm square hole coke is
placed in the specified tumble drum and rotated for 30 revolutions, removed, screened and
replaced in the drum and subjected to a further 150 revolutions. The drum contains lifters
that raise the coke and allow it fall so that it undergoes a large number of impacts with the
drum walls. The test is based on the Japanese Industrial Standard JIS K2151. A number of
indices can be measured. The two reported in this study are

 DI3015– the percentage material remaining +15mm square hole after 30 revolutions
 DI15015– the percentage material remaining +15mm square hole after 150 revolutions.
For these indices, the larger the number, the stronger the coke. This test is commonly used
in Japan to evaluate coke strength.

Combined Half Micum / Irsid Test – a 25kg representative sample of the +20mm round
hole coke is placed in the specified tumble drum and rotated for 100 revolutions. The coke
is removed, screened and replaced in the drum and subjected to a further 400 revolutions in
the drum. The test is based on the ISO 556 standard.


o M40 – the percentage material remaining +40mm round hole after 100
revolutions
o M10 – the percentage material –10mm round hole after 100 revolutions
o I20 – the percentage material remaining +20mm round hole after 500 revolutions
o I10 – the percentage material –10mm round hole after 500 revolutions


Larger values of M40 and I20 and smaller values of M10 and I10 generally indicate strong
coke. This test is commonly used in Europe to evaluate coke strength.

ASTM Tumbler Test – a 10kg representative sample of the –75mm +50mm square hole
coke is placed in the specified tumble drum and rotated for 1400 revolutions. The test is
based on the American standard ASTM D 3402.

 Stability – the percentage of coke +25mm after 1400 revolution.


 Hardness – the percentage of coke +6.3mm after 1400 revolutions
Generally the higher both indices the stronger the coke. The test is commonly used in North
America to evaluate coke strength.
Carbon Burnout
Carbon burnout is controlled by the char morphology which strongly depends on the rank
and petrographic composition of the coal and the thermal and gaseous environment under
which the coal devolatilises. Minerals within the coal can have a slight influence by
catalyzing the oxidation reaction.

The prediction of carbon burnout based on empirical curves derived by Blake and Robin
(1982) has been widely used in the power industry. This burnout model requires only one
boiler design parameter, Furnace Heat Release rate (FHR) (fuel burn rate divided by furnace
volume), and the operational parameter of excess air level used in the boiler. The coal
parameters are the dry ash free volatile matter content and the fineness (percent passing
75 um) of PF.

The figure below shows the predicted and actual burnout performance versus the volatile
content of a wide range of coals at both full scale and pilot scale. The predicted full scale
burnout is based on the design of a typical Asian power plant using imported coals while the
actual full scale data is from a wide range of power plants. This is one reason for the
spread in this actual data.

The pilot scale data are from projects conducted in ALS's 150 kW Boiler Simulation Furnace
(BSF) all these results were for coals fired at a fineness of about 70%. The predicted pilot
scale curve uses the FHR that best fitted the pilot scale results without adjusting for carbon
loss and fineness, this FHR is close to the calculated FHR for the pilot furnace. For lower
volatile coals, the poorer burnout will reduce the calculated FHR of the BSF by as much as
20%. When the FHR is adjusted for burnout there is better agreement between actual and
predicted burnout for the pilot scale results.

As indicated in this figure the volatile matter content of a coal can be used as a general
guide to carbon burnout. But, as shown by differences between pilot scale predicted and
actual data for some coals, other factors, such as maceral composition of the coal, also
influence burnout. Su and others (2001b) showed that a maceral index can give a slightly
better fit to pilot scale data, though their correlation does not allow the scaling- up of pilot
data to predict full scale performance.

Other empirical approaches to the prediction of burnout have been published. Wu and
others (2006) have applied petrographic techniques to estimate char morphology which
allows prediction of burnout. This has been successfully used by a UK power plant for the
selection and quality control of imported coals. Niksa has incorporated the work of Hurt in
modeling char conversion with his FlashChain devolatisation model to predict burnout in PF
power plants and model the performance of gasifiers ( Niksa & Hurt 2005).
Volatile Evolution
The rate and amount of volatile release during the heating of coal is strongly influenced by
the rank of the coal and the heating environment. Volatile evolution directly impacts
on flame stability, char morphology and NOx formation in combustion systems and
the plastic properties of coal during cokemaking.

At high heating rates of PF flame or in a blast furnace tuyere the volatile yield is greater
than that measured by a proximate analysis. The "Q factor" is that ratio of the expected
high temperature yield compared to the proximate volatile matter. There is very little data
published on the higher temperature volatile yield of Australian coals, the data of Wall and
others (1992), shown in the figure below, does indicate high Q factors for high rank
coals. The data from Ashman and others (1999) and Haywood and others (1995) show
similar trends to the data of Wall and others, though the Q factors calculated from this data
for low volatile coals were not as great. CoalTech has compared over 11 different methods
for determining high temperature volatile yield.

Lately there has been considerable literature on the use of phenomenological coal models
to predict devolatilisation, volatile nitrogen release and char formations. Generalised
devolatilisation models (network models) are based on chemical/physical descriptions of the
structure and processes of the coal particle as the particle heats up and pyrolyses. The
three main coal devolatilisation models that include nitrogen release are:


o FG-DVC [1] functional group-depolymerization vaporization cross- linking
o FLASHCHAIN [2]
o CPD [3,4,5] chemical percolation- devolatilisation
The reported predictions of these devolatilisation models are shown below.

Solomon and Fletcher reviewed the predictive ability of these network models. Brewster
and others (1995) found that FG- DVC gave improved predictions of mass loss compared to
the usual two step model for coal devolatilisation. FLASHCHAIN is being incorporated into
EPRI’s software package Coal Quality Impact Model to improve the NOx predictions.

The CPD model differs from other network models in that only one empirical parameter is
used to fit the devolatilisation of all coals, all other coal-dependent structural coefficients are
taken directly from 13C NMR measurements. Recently, Perry (1999) expanded CPD model
to include nitrogen release as tar and light gases (CPD- NLG). To extend the use of CPD to
when 13C NMR measurements were not available Genetti (1999) produced equations of best
fit to allow coal proximate and ultimate analysis to be used to determine the NMR based
inputs to the model.

Some work has been carried out by CoalTech to evaluate the CPD-NLG model, this has been
limited to evaluating 8 coals using the NMR inputs calculated based on Genetti’s work.
Reasonable agreement was found with Entrained flow reactor data, but for wire- mesh data
(higher temperatures) the CPD-NLG model predicts a smaller increase in volatile yield than
the actual data. Typical yield curves for char (fchar), tar (ftar), light gases (fgas) and total
volatiles (ftot) with the fraction conversion of nitrogen to the major species are given in
Figure below.
To ascertain the influence of peak temperature on the CPD-NLG model predictions two final
temperatures were used, 1500 C and 2000 C, with a heat up time of 300 ms. For a high
volatile coal, the volatile yield did not increase significantly (Q factor of 1.01), the volatile
nitrogen increased by about 5% and the char nitrogen decreased by 5%. For a medium
volatile coal (24 % daf), the volatile yield only increased by 2% (Q factor of 1.45), the
volatile nitrogen increased also by 2% and the char nitrogen decreased by 2%.

[1] Solomon P., et el 1993, Fuel 72:469.

[2] Niksa S., 1996, “Assess coal quality impacts on your personal computer”, 1996
International AFRF Symposium.

[3] Fletcher T., 1992, Energy & Fuels, 6:414.

[4] Fletcher T., 1999, “User’s manual for the CPD Model”, Brigham Young University, ,
1999.

[5] Fletcher T., Kerstein A.R., Pugmire R.J., Solum M., Grant D.M., 1999, “A chemical
percolation model for devolatilization : summary”, Brigham Young University, , 1999.

[6] Jones J.M., Patterson P.M., Pourkashanian M., Williams A., Arenillas A., Rubiera F., Pis
J.J., 1999b, “Modelling NOx formation in coal particle combustion at high temperature: an
investigation of the devolatilisation kinetic factors”, Fuel 78, 1999.

[7] Niksa S., Muzio L., Fang T., Hurt R., Sun J., Mehta A., Stalling J., 1999, “Assess coal
quality impacts on NOx and LOI with EPRI’s NOx-LOI predictor”, Coal Quality Evaluation
Tools, EPRI 1999.
Flame Stability
The main impact of firing a coal or a blend with poor flame stability performance would be
inability to maintain a stable flame at low load without the need for auxiliary fuel support. A
flame will be inherently unstable if the energy release from the volatile matter combustion is
insufficient to raise the char particle temperature to a level at which char combustion is self
sustaining under conditions of heat loss (including radiation to and from the flame) and
mixing intensity (including hot gas recirculation) for a given burner/boiler arrangement.

The pilot scale flame standoff data for the unblended and blended coals at different loads is
given in figure below. This figure shows that the results at the lower load were best ranked
by the curve for proximate volatile matter content, while the results for the higher load
were best ranked by the curve for high temperature volatile yield. The exception is coal 79
where even at the lower load there was sufficient heating of the coal by the hot refractory
within the furnace to cause heterogeneous ignition of the char. See blending for further
information on blending of thermal coals.

The ranking of the flame stability performance of coals fired in advanced Low NOx burners is
best done by determining the heating value of the volatiles released at the temperature
within the flame envelope created by these burners.
Milling
The Hardgrove Grindability Index (HGI) is included in most thermal/PCI coal specifications
as the indicator of mill performance. While HGI is adequate to characterise the grinding of
most coals, some Australian coals can be disadvantaged as the HGI will indicate higher
power requirements, lower throughputs and /or a coarser size distribution than actual.

The evaluation of a coal's behaviour in the thermal or PCI coal market requires knowledge
of the size distribution of the organic and inorganic components of the coal to enable the
determination of performance parameters such as combustibility, slagging and fouling and
handleability.

In a recent CoalTech report, the mill performance of maceral groups or microlithotypes was
shown to be additive for most coals. That is, each maceral group or microlithotype behaved
independently and a size fraction of the product PF was the mass weighted sum of the
petrographic components of that size . It was possible to determine the size distribution of
the product PF for a wide range of milling conditions based solely on petrographic analysis.
As microlithotypes were not determined directly they were estimated from the maceral
analysis. The size distribution of individual maceral groups or microlithotypes can also be
estimated. Size distribution based on petrographic analysis proved to be a better estimate
than that obtained based on the HGI.

Mill power can also be estimated from petrographic analysis, but the HGI is a better
predictor of mill power.

Based on these findings an improved mill performance nomogram was proposed which
extends the typical mill curve to account for the different maceral composition of coals.

Coal Blends

In anACARP project it was found that for most coals, a good estimate of a blend's size
distribution can be made assuming that the size distribution of the individual coals, milled
under the same conditions, are added together in the proportions of the blend. The
exception is when a very soft coal (HGI 90) is blended with a very hard coal (HGI 35). In
this case preferential milling (more reporting to the smaller size fractions) of the softer coal
occurs.

All coals studied in this project show some sign of preferential grinding of the softer maceral
group when the coal was milled individually or in a blend. This preferential grinding of
macerals is due to differing strengths of the macerals which dictates how the size reduction
of the maceral varies with energy used in the breakage of the particle.

Breakage characteristic curves (change in size reduction per unit of energy) for vitrinite and
inertinite were determined from the milling data of the coals and blends. For these curves
the mill specific power was proportioned to the maceral groups based on the petrographic
analysis and blend composition. These curves have similar trends as those found for the
breakage of lithotypes. Generally, these trends are for the particle size to initially decrease
rapidly then approach a constant size with increasing breakage energy. The results indicate
that the breakage characteristic curves of maceral groups in individual coals do not change
when they are blended with other coals.

It is only when the reduction in breakage energy proportioned to a maceral group of a coal
in the blend moves to the steeper region of its breakage characteristic curve that the
preferential milling of a coal in a blend is observed in the size distribution of the blend. This
would also explain the non- linearity of Hardgrove Grindability Index (HGI) determined on
some blends when compared to their component coals as HGI is a measure of size reduction
for a fixed energy.

The results show that the breakage of a coal particle can have three mechanisms, these
are:

 Vitrinite and Inertinite Breakage: The breakage of both vitrinite and inertinite
consumes energy in the milling process. .
 Vitrinite Dominated Breakage: The breakage energy of the coal is dominated by the
breakage of the vitrinite.
 Inertinite Dominated Breakage: The breakage energy of the coal is dominated by the
breakage of inertinite.
The results also explain why some coals, those with an inertinite dominated breakage
mechanism, do not follow the generally trend between HGI and the maximum vitrinite
reflectance.

It was shown that relationships between mill specific power and HGI, Rosin Rammler
parameters and vitrinite reflectance and the breakage characteristic curve of vitrinite and
inertinite allows one to determine the mill performance of a coal or a blend. Currently it is
not possible to estimate the breakage characteristic curve from petrographic analysis.
Further milling testing, under fixed mill conditions, for wider range of coals will assist
identifying methods to predict the breakage characteristic curves for the vitrinite and
inertinite maceral groups.
NOx Formation
The recent economic problems of Asian countries have slowed the anticipated growth in coal
fired generating capacity. This growth has been estimated to be 345 GW of new capacity in
APEC countries by 2010. The anticipated growth in coal fired generating capacity within
Asia has encouraged the development of power plant technology to fire a wider range of
fuels, to increase efficiency and lower capital costs. In particular, the potential market for
new power plants in China has led to newer burner technologies that achieves stable flames
and reduced NOx emissions while firing coals that have lower volatiles than coals normally
traded in the international coal market.

Increased global concerns on NOx emissions are reflected in the limits being set by
regulating authorities of countries importing coal for electricity generation. This makes NOx
emissions an important consideration in the design of new coal fired power plant.

CoalTech' ACARP report reviewed the influence of coal properties on NOx formation. Coal
properties, such as volatile and nitrogen content, impact the levels of NOx emissions, but
there is no simple relationship between these coal properties and NOx emissions that fits all
operational power plant. The impact of coal properties is lessened through the use of low
NOx burners, over-fired-air, fuel reburn and the post boiler technologies of selective
catalytic (SCR) and non-catalytic reduction (SNCR), though there may be cost
disadvantages due to poorer burnout and higher operating costs.

NOx from Coal Combustion

 About 95% NO, ~5% NO2, and < 1% of N2O,


 Largely depend on combustion intensity
 Uncontrolled NOx level of 175~3200ppm
NOx Sources

 Thermal NOx
o Formed by attack of O atom on N2 in combustion air
o About 20% of total NOx emission from PC burners
o Mainly affected by flame T & O2 concentration, with the former most
important.
 Fuel NOx
o Formed by pyrolysis & oxidation of N compounds in coals
o About 80% of total NOx emission from PC burners
o Fuel N vs. fuel NOx relation is complex and unclear:

 Coals with high N do not necessarily produce
more NOx;
 Increased N in coal will lead to a decreased
conversion rate;
 N in char and in volatiles have different
conversion ways to NOx

o

o

o

o
 O2 concentration plays an important role in fuel N conversion into
NOx



o Prompt NOx
 Formed by capture of N2 by hydrocarbon radicals. Small contribution in
coal combustion.


It is generally accepted that the level of NOxemissions resulting from the combustion of
pulverised coal is influenced by three factors:

 Coal properties.
 Plant design, including furnace design and use of post combustion de-NOx plants.
 Boiler operation, including load, excess air level, in-service mills/burner groups etc.
In this context, it is recognised that NOx generation is a very complex process, and is
dependent on localised temperatures, stoichiometry and chemical reactions, principally in
the near-burner region.

To assist in the understanding of the interactions within this process several empirical
models (fitting an arbitrary equation to data), simple reaction engineering models
(combinations of CSTR and PFR reactors) using simplified chemical kinetics and complex
computational fluid dynamics (CFD) models for the prediction of NOx emissions have been
developed by organisations around the world. All methods have their limitations and
advantages.

To rank NOx emissions of different coals from a power plant operating under the same
conditions based on coal properties requires only two parameters - the volatile yield at high
temperature and the nitrogen content, as shown in the figure below. The product of these
parameters represents the volatile nitrogen release that can be converted to nitrogen within
the flame or conversely the char nitrogen that can react to form NOx. These properties,
together with the known NOx emissions of one coal, allow good estimates to be made on the
likely NOx emissions of other coals under the same operating conditions. But these rankings
or NOx emission estimates are most likely to change with changes in plant operation.
For a power plant, given sufficient NOx emission data on a wide range of coals and
operating conditions, it is possible to formulate an empirical model to accurately predict
NOxemissions. This is the bases of several commercial NOx control software packages.

Currently, there is no simple means to accurately predict NOx when only limited plant
operating or design data is available. Boiler manufacturers continue to use in-house
empirical correlations, not CFD models, to predict NOxformation.

The volatile yield and the early volatile nitrogen release of coals are important to the
reduction of NOx within the flame. The understanding of these high temperature processes
is increasing with the research conducted in Australia and overseas, but generalized models
that can be used with all coals has yet to be developed.

For details on the prediction of NOx see "Predicting NOx".


SOx
SOx emission levels can be calculated from coal sulphur content. Ash chemistry and ash
content have some influence, as some absorption of SO2 takes place in the ash. The level of
absorption increases with increasing alkalinity levels in ash (ie. high levels of CaO, MgO,
etc.) and increasing ash content, however it is unlikely that absorption levels would ever
exceed 10-15% of the total SO2 produced. SO2 emissions are additive and therefore can be
calculated from the sulphur content of the component coals with any SO2 absorption onto
the ash estimated from the ash chemistry of the component coals.

SOx emissions may be reduced by direct injection of a sorbent directly into the furnace
chamber and/or by flue gas desulphurisation (FGD) systems attached to the back end of the
boiler. All FGD systems involve intimately mixing the SOx emissions with an absorbing
chemically reactive material, which converts them into a solid or liquid product. This product
may have a marketable value.
Particulates
Fly ash is usually collected by electrostatic precipitators or fabric filters, which operate at
efficiencies of greater than 99%. In electrostatic precipitators the electrical resistivity of the
fly-ash is an important factor affecting the performance of the precipitator. Resistivities
greater than 1012W.cm are considered undesirable, since high electrical resistance can lead
to back ionisation, which reduces collection efficiency. The collection efficiency of high
resistivity fly ashes may be improved by dosing the flue gas with additives such asSO3 or
NH3,or by using intermittent or pulse-energising techniques to control the voltage in the
precipitator.

The precipitator performance of blends generally is near to or slightly better than the
component coal with the best precipitator performance.

Fabric filters are made of a woven, knitted or felted textile in the shape of cylindrical bags
through which the flue gas is passed. The flyash is collected on the bag and then removed
by shaking, reversed gas flow, sonic horns or air pulses. The properties of the flyash that
affect the performance of fabric filters relate to how the flyash compacts on the bag. This
influences the pressure drop through the bag and the flyash adhesive strength, which
influences the force required to remove the dust layer.

ACARP has published a summary of on the precipitator performance of Australian


bituminous coals. Hall [1] reviewed the effects of process conditions on electrostatic
precipitator (ESP) performances, in this review Hall listed the most important factors
affecting precipitator performance as:

 Ash Resistivity - as controlled by ash chemistry, gas temperature and density, gas
composition - especially H20, S03/H2S04 or the presence of other specific conditioning
agents. Resistivity basically controls the allowable ESP operating current densities,
hence operating voltages, useful power input and size of ESP equipment required.
 Inlet Particle Size Distribution and Concentration - as influenced by boiler type, fuel
properties and process factors. Senior and others (1993) showed that the particle size
distribution greatly influences precipitator performance.
 Electrical Energization - the heart of the electrostatic precipitation process comprising
the electric field strengths for particle charging and collection by means of electric
forces applied directly to the particles, per se. Said electric fields and copious supply of
negative gas ions for particle charging are provided by a high voltage corona
discharge maintained between suitable electrodes, e.g. parallel plate ducts with
coplanar discharge electrodes centred therein. Ion mobilities, voltage/current
characteristics and electric fields are influenced by gas and ash properties. The
electrical energization system and conditions most strongly influence achievable
effective particle migration velocities.
 Specific Collecting Area (SCA) - ESP total collecting area per unit gas flow
rate. Important influences on SCA include efficiency requirements, electrode
geometry, ESP duct width and electric energy density.
 Dust Loss Factors - as influenced by particle adhesive and cohesive properties, gas
distribution quality, gas sneakage, local air inleakage, gas turbulence conditions, gas
velocity erosion and rap reentrainment, excessive ESP sparking, saltation, and long
term reliability factors.
Burnout is also important as the presence of unburnt carbon can alter the electrical
properties of the fly ash, and adversely affect dust emission levels.
Blending can have a large influence on ESP performance as demonstrated during trial burns
at a Queensland power station using blends of Meandu coal with Jeebropilly and Wilkie
Creek coals [2]. Meandu coal, the coal normally fired, has approximately 28% ash, which is
comprised of 71% quartz. This makes the ash quite difficult to precipitate efficiently. In
these trials it was found that the particulate emissions of the blends decreased to less than
a quarter of the usual emissions. The main reason for this decrease was attributed to a
decrease in ash resistivity due to greater amounts of magnesium and sodium ions in the ash
of the blended coals. Spero [3] also noted those from the Walloon Coal Measures had
superior ESP performance.

ESP efficiency can be calculated using the modified Deutsch equation which is given below.
The effective migration velocity (ωk) in this equation is the parameter that is normally
related to the ash properties of a coal. The Specific Collection Area (SCA) is the size of the
precipitator.

The general application of correlations for the prediction of effective migration velocity,
based on ash analysis alone, needs to be treated with the same degree of caution as the
wide application of predictive slagging indices. This is because it is the surface composition
of the fly ash that is important not the bulk composition. The figure below shows the
comparison between actual pilot scale performance and that predicted by a correlation base
of ash analysis (Bennett, 1996). The superior ESP performances of the coal from the
Walloon Coal Measures (Coal 72) and blends using this coal are shown in this figure.
[1] Hall, H.J., 1990. “The effects of fireside process conditions on electrostatic precipitator
performance in the electric utility industry”, Proceedings: Eight Particulate Control
Symposium, San Diego, California, Nov., 1990.

[2] Whelband, B., 1999, “Influence of coal blending on electrostatic precipitator


performance at Tarong Power Station”, Department of Chemical Engineering, The University
of Queensland, Thesis & Seminars, 1999.

[3] Spero,. C., 1997, “Walloon coals: their properties and power station performance”,
Report to: Queensland Department of Mines and Energy QTHERM Program, Austa Energy
Report RTM 97/003. October 1997.

[4] Bennett P. Conroy A., 1996, “Coal Quality Impact Model”, ACARP Final Report C3091,
October 1996.
Slagging & Fouling
Two main types of deposits in boilers.

Slagging

 Deposits within furnace, in areas directly exposed to flame radiation such as furnace
walls and some widely spaced pendant super-heaters.
 Take place in the hottest parts of boiler
Fouling

 Deposits in areas NOT directly exposed to flame radiation such as the more closely
spaced tubes in convection sections of boiler
 Take place as flue gas & suspended fly ash cool down
Effects of deposition on boiler performance

 Reduction of heat transfer from combustion gas to water-steam


 Lead to an increase in gas temperature
 Lead to a further increase in deposition rate
 Result in continually changing conditions in boiler.
The formation mechanisms of slagging and fouling is very different for both forms of boiler
deposits. While indices calculated based on bulk ash properties can be useful for the
estimation of a coal propensity to slag or foul, these indices very rarely simulated the
mechanisms involved. For an accurate estimation of where and the extend of slagging and
fouling requires knowledge of the distribution and size of the minerals species in a coal
particle and free minerals coupled with detailed knowledge of the gas flows from the
burners through the boiler.

Common slagging and fouling indices can be used to indicate the like slagging and fouling
performance of coals.
Blending
The understanding of the combustion performance of blends is important in determining the
economical and environmental benefit of blending. The combustion performance of a blend
dictates the level of carbon in the ash, with high carbon-in-ash equating to energy losses
and increased ash disposal costs. Emissions of NOx, SOx, CO2 and particulates are also
influenced by blend composition.

This brief review is based on CoalTech's report examining the performance of thermal coal
blends for the Queensland Government.

The combustion performance of a coal blend is more complex than that of a single coal
because it is not only dependant on the combustion performance of each component coal
but also on possible interactions between coals. This interaction between coals first occurs:

 in the milling of the blend if there are large differences in the hardness of each coal
then there is potential for large differences in the size distribution of the component
coals which will influence flame stability, burnout and possible NO x formation.
 within the burner flame the volatile release from the different blend components will
impact flame temperature and therefore can influence flame stability, char reactivity
and NOx formation.
CoalTech's milling and combustion model shows how size distribution of some component
coals can impact on burnout performance. For blends with similar milling behaviour the
burnout of blends is additive and can be determined from the burnout performance of the
component coals. When the milling performance of component coals differs significantly
then there will be preferential grinding of the softer coal leading to a finer size distribution
for that coal. Usually the softer coal is of higher rank, lower volatile content and poorer
burnout performance. The preferential grinding of the softer, lower volatile coals does allow,
depending on the coal, up to 30% of the lower volatile coal to be used in blends that will
have acceptable burnouts similar to the higher volatile coal.

The model also indicates that when these higher rank coals are used in proportions greater
than 50% the burnout can be similar or worse than the higher rank coal if it was burnt
unblended.

As the temperature at which the blend burns increases, due to higher load, burner and
boiler design ,the impact of differential grinding decreases and burnout of the blend
approaches a linear relationship to the burnout of the component coals.

The simple model developed does not allow for any influence of different maceral types.
There is some evidence that fragmentation will increase the burnout of some lower volatile
coals with high vitrinite content. This fragmentation can be increased when the heating rate
within the flame is increased by the addition of a higher volatile coal and there is sufficient
oxygen in the flame to oxidise the char surface.

Flame stability of a blend is greatly enhanced when only small amounts of a Surat Basin
coal is included in a blend. This is due to the high hydrogen content of this per- hydrous
coal which contributes to a high energy release within the flame.

Based on the findings in this report, lower volatile Bowen Basin coals that are now excluded
from the international thermal coal market could be used in thermal coal blends in
proportions of up to 30% without unduly impacting on combustion performance.
The early release of coal nitrogen within the fuel rich region of a flame is the key to the
minimising of NOx emissions. Second generation low-NOx burners generate rapid heating
and high temperatures at the centre of the fuel-rich flame, which promotes the
devolatilisation of the coal and early release of fuel nitrogen leading to lower NO x emissions,
better flame stability and carbon burnout.

Coals with high volatile content and low nitrogen content produce very low NO x emissions.
The addition of these coals to a thermal coal blend will significantly reduce NO x emissions. It
has been demonstrated in full-scale burner tests that the medium and low volatile coals of
the Bowen Basin can be fired in second generation low-NOx burners with acceptable
NOxemissions.

The general application of correlations for the prediction of slagging and


fouling or electrostatic precipitator behaviour for a wide range of coals, based on ash
analysis, needs to be treated with a degree of caution. This is because it is the amount, size
and type of minerals in the coal and surface composition of the fly ash that are important
not the bulk composition. A wide range of slagging and fouling indices are given in this
report, their use will only indicate potential problems.

The correlation used in the CoalTech report on blending between ash properties and the
performance of an ESP does give a reasonable estimate for the outlet dust loading for most
single coals. But this correlation failed to predict the very good ESP performance of Surat
Basin coals which resulted in low particulate emissions. When Surat Basin coals are used in
blends with a coal of poor ESP performance there is an improvement that is far greater than
if ESP performance was additive.
Economic Benefit
Replacement of Coke
Impact of Ash
Market Influences on Price
An Economic Injection Rate
The main cost benefit of PCI is the replacement of high cost coking coal. The amount of coke replaced by a
sensible heat of the combustion gases and ash.

There is an economic injection rate, namely the injection rate at lowest total coal costs, for every BF. This r

The value-in-use benefit of PCI depends on its replacement ratio and its impact on the energy balance of th

Replacement of Coke
Any coal can be injected with the main requirements, from the viewpoint of chemical analysis being low sul
coal. See attached summary for details of various methods for calculating the replacement ratio of an injec

In Figure 1 different methods for calculating replacement ratio have been plotted against the dry ash free c
between coal properties. The figure does show that all methods used to determine replacement ratio give a

Figure 1 Variation of replacement ratio with rank

Impact of Ash
Mr T. Fukishma of F-TeCon Pty Ltd carried out the modeling (Bennett & Fukushima, 2003) to investigate th
focusing on the impact on operating costs. For this study the blast furnace process parameters shaft efficie
different amounts of ash – 8, 9, and 10 % ad. These calculations indicated that the change in coke require
For an injection rate of 150 kg/tHM, an increase of 1% in the ash of the injected coal would result in a 1.2 k
chemistry), which results in a net increase in BF energy requirements and other raw materials. Therefore, t

The increase in coke rate per 1% increase in ash determined by this study is significantly less than that sug
Ping and Suen [2] state that for each one percent of ash will increase the blast furnace fuel rate by two per

Brouwer and Toxopeus [3] in summarising the PCI operating results at Hoogovens IJmuiden blast furnace d
coke rate.

There is a value-in-use penalty for increased ash, but for the majority of coals traded as PCI coals the ash d

[1] Poveromo, J., 2004, Blast furnace fuel injection trends, Met Coke World Summit, Chicago, October 2004

[2] Xi Ping, P., Suen H., 2003, PCI Application in blast furnace operation and market trend in China, McClos

[3] Brouwer, R.C., Toxopeus, H.L., 1991, Massive coal injection at Hoogovens IJmuiden BFs, Revue de Meta

Market Influences on Price


Two market forces control the price of PCI coal, the price of hard coking coal and the price of thermal coal.
9%. Figure 2 show how the coal costs of a typical coke oven blend and the thermal coal price influences th

Figure 2 A guide to the expected price range for low volatile PCI coals

An Economic Injection Rate


As shown in Figure 3, the incremental change in the amount of coke replaced by coal injection decreases as
the data in Figure 1and a simple relationship between PCI rate and required coke quality then the most eco
about 115 kg/tHM, just below the average Japanese injection rate for that year.

Figure 3 World coke and PCI rates. The estimated influence of HV


and LV coals on coke rate is determined based on the replacement
ratio of typical coals.

Figure 4 Influence of injection rate on BF total coal costs


Impact on BF Operation
Unburnt Char
Slag Chemistry
Coke Properties
In a period of high global demand steelworks are focusing on maximum productivity. To obtain high produ
permeability in the lower and upper zones of the blast furnace. At injection rates, greater than 140 kg/tHM

 Reduction in coke/ore ratio,


 The size of the raceway,
 Reduction of permeability of the coke surrounding the raceway,
 Changes in temperature distribution in the raceway,
 Mechanical degradation of coke in the raceway, and
 Decrease in deadman temperature.
All these changes are interdependent and are influenced by the properties and amount of the injected coal,

Figure 1 Factors that influence blast furnace performance in the lower zone

Unburnt Char
The development of the modern injection lances has greatly improved the combustion of coal within the tuy
adjustment of BF operating conditions. The impact of unburnt char on BF permeability is not that significa

The full understanding of the all the physical and chemical mechanisms that influence the devolatilisation, f
Slag Chemistry
In the immediate region surrounding the raceway the ash from the injected coal has a major influence on t
ash/slag chemistry and slag viscosity is growing and will lead to tighter controls on amount and compositio

The influence of ash from injected coal on the physical properties of the dripping slag and the permeability

Figure 1 Ash behaviour in and near the raceway ( after Ichida and others [1])
[1] Ichida, M., Orimoto, T., Tanaka, T., Sakatani, M., Ueno, H., 2002, Behavior of pulverized coal ash and p

Coke Properties
Good coke quality is recognised by all BF operators as a necessary requirement to achieve good productivit
be improved by increasing the reactivity of the coke and this is being actively researched particularly in Jap
Milling & Handleability
The main operating costs, other than coal costs, are related to the milling and distribution of
the coal to the blast furnace. LV coals, compared to HV coals, are generally softer and thus
requiring less energy to grind, but they require a higher energy to dry the coal down to a
moisture level required to eliminate the risk of handling problems.

Moisture in the pulverised coals can lead to handling problems in the PF bins and transport
system. Clays in the coal may worsen the handling problems associated with moisture.
Edinburgh Cohesion Tester
Influence of Moisture
Influence of Size
Influence of Ash Content
Starting in 1996 the Edinburgh Cohesion Tester (ECT) was developed over several years with funding from
assisted with the evaluation of the ECT in the UK. ECT Testing can be carried out in Australia through Coa

As shown by many researchers the unconfined compression strength is a measure of the cohesive strength
consolidation pressure (Figure 1).

Figure 1 Edinburgh Cohesion Tester


Figure 1(a) shows the removable three-piece mould into which coal is placed when a uniaxial consolidation
recorded by the digital display. The ECT uses similar principles as the triaxial and uniaxial testers developed

The ECT has been used by Edinburgh University to undertake extensive studies of many coals to discover k
overcome the main problems encountered in other similar testers. These features include:

 a three piece split mould used to minimise the disturbance while setting up the sample,
 the mould is coated to reduce wall friction during vertical consolidation,
 support of the mould on a spring loaded base to effectively consolidate the coal from top and bottom
 sample height correction to avoid interference with the initial sample (cutting or adding material after
 a digital display with a peak hold to record the maximum vertical force required to fail the sample.
A patent application for the novel features of this tester has been filed by Edinburgh University.

Predictions obtained with the ECT were compared with the results obtained using the Durham Cone [3,4]. C
coal bridging in the receiving bunker or action required to aid discharging (air lances etc). These tests show
handleability problems still exist despite the thirty-year extensive usage of the Durham Cone. The ECT res
improved ranking of the coals and therefore gave a better indication of the possible handling problems of m

Figure 2 shows that the ECT measurements for two problematic consignments, which indicates that handlin
Edinburgh University found that handleability is very sensitive to moisture content in a highly non-linear ma
handleability of a coal is characterised by its Stress-Moisture-Cohesion (SMC) function that describes the ef

Zhong and others [5] devised a mathematical model to predict the resultant cohesion of a coal blend consis
found experimentally. The predicted cohesions of trial blends were found to be in very good agreement with

The speed of testing of the Edinburgh Cohesion Tester permits a substantial increase in the available inform
minutes of operator training.

An agreement between The University Court of the University of Edinburgh and CoalTech Pty Ltd allowed Co
project. This project found that all coals tested had unique SMC functions and that the size distribution (not
gives a better discrimination between coals that may or may not have a handleability problem.

[1] Arnold, B., (2004), Efficient handling of coal for power plants: Development of a coal handleability index

[2] Chambers, J., Liu, Y., Roberts, A., (2001), Effect of various coal constituents on the storage and transp

[3] Ooi, J.Y., Rotter, J.M., Lahlouh, E.H. and Zhong, Z. (1998), Blind trial on coals for rapid handling asses

[4] Zhong, Z., Rotter, J.M., Ooi, J.Y. and Armstrong, B. (2001), Rapid assessment of handling performance

[5] Zhong, Z., Ooi, J.Y., Rotter, J.M. (2005), Predicting the handleability of a coal blend from measuremen

Influence of Moisture
The influence of moisture and stress is characterised by the Stress-Moisture-Cohesion (SMC) function. From
moisture are the most important handling properties of a coal product and the ECT allows these to be meas

In the recent ACARP project 15070 19 Australian coals were tested and it was found that all coals tested ha

Figure 1 SMC function for the coals tested in ACARP project

Influence of Size
Figure 1 shows the how the unconfined strength of one coal behaves for a wide range of top size and shows
coarse coal is removed and when fine coal is added was also found in the ACARP project 15070 .

As shown in Figure 2, all parameters relating to the size distribution of the coal have a significant influence
related.

The Rosin Rammler “n” and the fines content have the best linear fit to maximum unconfined strength with
Figure 1 Influence of top size for one coal (testing carried out by Edinburgh
University)
Figure 2 Influence of size distribution on maximum unconfined strength for 19
coals, ( ACARP project 15070)

Influence of Ash Content


The dependence of cohesion strength on consolidation stress and moisture content is the stress- moisture-c
of ash, in particularly non-swelling and swelling clays, on handleability. Figure 1 shows how ash influence th
showed weak linear correlations of ash (%d) with the unconfined strength on as received sample and with
Figure 1 Influence of ash on unconfined strength
Black and others (2005) [ ACARP Project C11016] conducted ECT testing on 7 pulverised coals to determin
content after milling. Figure 2 shows the SMC function for two coals of different rank (Aus L16 & Aus M17)

At higher moisture contents, one could be expected the cohesion strength of the PF at different moistures w
trends of the other coals. Mikka and Smitham [1] showed that in coarse coals clays could contribute to han
Figure 1 ECT test results showing the influence of PF moisture on unconfined
strength for different consolidation pressures, only coals Aus L16 and Aus M17
where tested at a range of moistures. Data from Black and others (2005).
[1] Mikka, R., Smitham, J., (1985), Coal handleability assessment, Third Australian Coal Preparation Confer
Other Tests
Hall and Cutress [1] examined several methods to measure cohesion and found the triaxial
test the most sensitive to changing material properties. As their aim was to measure flow
rates from rail wagons rather than when blockages occurred, they developed the Durham
Cone as a good practical measure of flow of coal from rail wagons. The handling issue
currently faced by the UK and Australian coal industries is to determine when a coal will not
discharge consistently from a wagon and this is directly related to cohesion.

Mikka and Smitham [2] used the Durham Cone to examine the influences of moisture, fines
and clays on the handleability of coals. Their results on blends made with different fines
content showed that at low levels of fines the handleability was not affected by the moisture
content but as the fines increased the handleability became very sensitive to moisture. They
also showed that the size distribution of the coal influenced how a coal’s handleability was
affected by fines and moisture content. In comparing the prediction of handleability based
on the critical outlet diameter, as determined by the Jenike shear cell, and based on the
Durham Cone. The Durham Cone gave a better prediction of the observed handleability for
two coals.

Chambers and others [3], in ACARP project C6057, examined in detail the effects of
moisture and fines content on the handleability of coal. They investigated the influence of
particle size and size distribution on moisture retention and unconfined yield strength. As
moisture is increased there is an increase in unconfined yield strength to a maximum
strength, then with further increases in moisture there is a decrease in the unconfined
strength. The work was limited to two coals with different levels of fines and moisture
contents. They concluded that the most appropriate test for handleability is the unconfined
stress. They used a prototype apparatus that is mechanically more complex but similar in
principle to the ECT.

An extensive study [4] was conducted by Casella CRE Energy into the handleability of
Australian PCI coals using the Johanson Indicizer System (JIS) and ECT. The Johanson
Indicizer System (JIS) comprises three test machines; the Hang-up Indicizer, the Hopper
Indicizer and the Flow Rate Indicizer. As the JIS requires a sample size of minus 6.7mm,
both the JIS and ECT tests were conducted on this size fraction. The results of the JIS and
ECT gave similar handleability ranking for the seven coals tested over a range of moisture
contents. When the coals were crushed to GCI sizing no unconfined stress was measurable.
An interesting result of this research showed the sensitivity of unconfined stress of some
pulverized coals, as measured by ECT, to low moisture levels. This possibly explains why
some steel plants have experienced blockages in PCI systems.

The Handleability Monitor [5] is a handleability test developed by Nottingham University.


The coal sample is pushed through the venturi section by a hydraulic ram. The handleability
index is the maximum pressure obtained as the sample moves through the venturi.
According to the developers of this tester, the main advantages are it can be used
continuously and it can deal with large particles.

Holuszko and others [6] conducted comparative testing between the Handleability Monitor
and the Durham Cone and found, when tested under the same conditions, the tests were
comparable, though when measuring the influence of moisture the Durham Cone was more
sensitive. They observed the influence of size distribution on the Durham Cone test results
as noted by other authors. They showed the wettability characteristics of the coal influence
handleability. They termed higher rank coals (low equilibrium moisture) to be hydrophobic.
These coals displayed no handling problems up to the equilibrium moisture and then were
more prone to handling problems than the other coals tested.
Brown and Miles [7] reviewed the different tests used to measure handleability. These tests
are; Durham Cone, Jenike shear cell, Handleability Monitor, Edinburgh Cohesion Tester and
a new method - avalanching technique. They concluded that the Durham Cone and the
Handleability Monitor were the only tests suitable for the normal product sizing of -50mm.
No comparative testing was conducted in this review nor did the authors make reference to
any results from any other comparative study on handleability.

Arnold [8] summarised the research in the 1990's to develop a coal Handleability Index (HI)
for power stations. This work was funded by the Electric Power Research Institute (EPRI).
The HI is based on the work of Jenike relating the critical opening of a bin to unconfined
yield strength (a material property) and bin geometry.

Initially the unconfined strength was determined by triaxial device where the consolidation
of a column of coal was achieved by pressurising a rubber boot around a column of coal
before the column was loaded to failure. To simplify the testing a uniaxial device was
designed. In the uniaxial device the coal was placed in a PVC cylinder compressed from the
top and then, after rotating the cylinder, compressed again from the top. The PVC cylinder
was in three longitudinal sections for easy removal from the coal column. To ensure the
lever arm was level, before a load was placed on the coal column, spacers were used to lift
the coal column. The load was applied to the column by placing coal in a pail hung from the
lever arm, the force required to collapse the column was calculated from the weight of coal
placed in the pail. EPRI also evaluated the Durham Cone test and the Jenike Shear Cell test,
standard cell and modified cell for larger top size, and found that the unconfined yield stress
as measured by the triaxial was the most reliable predictor of handleability.

A comparison between the HI determined by the triaxial and the uniaxial methods showed
that there was a very good correlation between the different devices. Testing of the coal
flow from a two-ton bin for a wide range of moisture, ash and fines contents showed the HI
correctly indicated when a no flow condition would occur. Based on the testing of 22 eastern
bituminous US coals a handleability classification system was proposed that relates
handleability to the percent passing 0.5mm and the moisture content.

CSIRO [9] researched the causes of rail discharge problems and identified that the loading
of wagons was as an important contributor to hangup in wagons and the position of first
impact is of principal importance in determining the location of the consolidation in the
wagons. Further research [10] showed that the amount of jackhammering required during
the unloading of a train provided a simple method for quantifying unloading performance,
that is, when a coal would trend to be held-up in the wagons. Information on
jackhammering times was collected on six coal products from two mines over a extended
period of time. This data showed that there were different unloading characteristics for the
different products, rail wagons used and the wagon loading techniques. Auditing the forces
during loading and transport to the port showed that travel forces contributed less to
consolidation of the coal than did loading forces.

To clarify the influences of coal properties and consolidation force a laboratory scale rail
wagon was constructed by CSIRO and a pilot scaled rail wagon by Queensland Rail. Around
40 Australian coals were tested in laboratory scale rig (Fig. 2) to determine the influence of
consolidating pressure on the amount of coal discharged on opening the door of the rig (Fig.
3).

[1] Hall, D., Cutress, J., (1960), The effect of fines content, moisture and added oil on the
handling of small coal, Journal of the Institute of Fuel, V. 33, 1960.
[2] Mikka, R., Smitham, J., (1985), Coal handleability assessment, Third Australian Coal
Preparation Conference, Wollongong, Nov., 1985.

[3] Chambers, J., Liu, Y., Roberts, A., (2001), Effect of various coal constituents on the
storage and transportation of coal, ACARP Project C6057.

[4] Black, N. Sullivan, K., Whitehouse, M., (2005), Improved measurement of handleability
for quality control of Australian coals in PCI and carbonisation applications, ACARP Project
C11016, June, 2005.

[5] Brown, D., Atkin, B., (2000), The industrial evaluation of a coal handleability monitor,
UK Dept. Trade and Industry, Report No. COAL R175, DTI/Pub URN 00/662, March, 2000.

[6] Holuszko, M., Laskowski, J., Brown, D., (2004), Handleability assessment of selected
coals using Durham Cone and Handleability Monitor, Coal Preparation, Vol 24, 2004.

[7] Brown, D., Miles, N., (2004), Assessment of coal handleability, Coal Preparation, Vol 24,
2004.

[8] Arnold, B., (2004), Efficient handling of coal for power plants: Development of a coal
handleability index, Coal Preparation, Vol 24, 2004.

[9] O’Brien, G., O'Brien, M., Firth, B., Nemeth, D., Graham, J., Gnanananthan, S., (2002),
Investigation into problems of discharging Queensland coals from bottom dump rail
wagons, ACARP Project C10061, November, 2002.

[10] O’Brien, G., Patterson, S.,O'Brien, M., Graham, J., (2004), Improved coal wagon
unloading by reducing loading force, ACARP Project C12061, September, 2004.
Impact on the Environment
One of the key messages from the Stern Review on the economics of climate change was -

"An overwhelming body of scientific evidence now clearly indicates that climate change is a
serious and urgent issue. The Earth’s climate is rapidly changing, mainly as a result of
increases in greenhouse gases caused by human activities."

Coal is a globally abundant and a relatively cheap fuel and will continue to play a significant
role in power generation and steel production. But coal use contributes significantly to
CO2 emissions. In the future technologies that enable carbon dioxide management will
become increasingly important.

A MIT study discusses the interrelated technical, economic, environmental and political
challenges facing increased coal-based power generation while managing carbon dioxide
emissions from this sector. Key findings in this study:

 Coal is a low-cost, per BTU, mainstay of both the developed and developing world, and
its use is projected to increase. Because of coal's high carbon content, increasing use
will exacerbate the problem of climate change unless coal plants are deployed with
very high efficiency and large scale Carbon Capture and Sequestration (CCS) is
implemented.
 CCS is the critical enabling technology because it allows significant reduction in
CO2 emissions while allowing coal to meet future energy needs.
 A significant charge on carbon emissions is needed in the relatively near term to
increase the economic attractiveness of new technologies that avoid carbon emissions
and specifically to lead to large-scale CCS in the coming decades. We need large-scale
demonstration projects of the technical, economic and environmental performance of
an integrated CCS system. We should proceed with carbon sequestration projects as
soon as possible. Several integrated large- scale demonstrations with appropriate
measurement, monitoring and verification are needed in the United States over the
next decade with government support. This is important for establishing public
confidence for the very large-scale sequestration program anticipated in the future.
The regulatory regime for large- scale commercial sequestration should be developed
with a greater sense of urgency, with the Executive Office of the President leading an
interagency process.
 The U.S. government should provide assistance only to coal projects with CO2 capture
in order to demonstrate technical, economic and environmental performance.
 Today, IGCC appears to be the economic choice for new coal plants with CCS.
However, this could change with further RD&D, so it is not appropriate to pick a single
technology winner at this time, especially in light of the variability in coal type, access
to sequestration sites, and other factors. The government should provide assistance to
several "first of a kind" coal utilization demonstration plants, but only with carbon
capture.
 Congress should remove any expectation that construction of new coal plants without
CO2 capture will be "grandfathered" and granted emission allowances in the event of
future regulation. This is a perverse incentive to build coal plants without CO 2 capture
today.
 Emissions will be stabilized only through global adherence to CO 2 emission
constraints. China and India are unlikely to adopt carbon constraints unless the U.S.
does so and leads the way in the development of CCS technology.
 Key changes must be made to the current Department of Energy RD&D program to
successfully promote CCS technologies. The program must provide for demonstration
of CCS at scale; a wider range of technologies should be explored; and modeling and
simulation of the comparative performance of integrated technology systems should
be greatly enhanced.
The Australian Coal Association web site details what actions the Australian coal industry is
taking to reduce the environmental impact of coal mining and coal use, see also the World
Coal Institute.

Вам также может понравиться