SESSION TITLE: COMBUSTION TECHNOLOGIES – ADVANCED SYSTEMS AND

PROCESS MODELING STUDIES - 2

ABSTRACT TITLE: OPTIMIZATION OF BOILER OPERATING SETTINGS AND

SOOTBLOWING SCHEDULE TO REDUCE SLAGGING IN
COAL-FIRED BOILERS1

Harun Bilirgen, telephone: (610) 758-6289, fax: (610) 758-5959, [email protected]

Carlos E. Romero, telephone: (610) 758-4092, fax: (610) 758-5959, [email protected]
Nenad Sarunac, telephone: (610) 758-5780, fax: (610) 758-5959, [email protected]

Energy Research Center

Lehigh University
117 ATLSS Drive
Bethlehem, Pennsylvania 18015
ABSTRACT

Uncontrolled deposits on heat transfer surfaces in coal-fired utility boilers can

interfere with the normal operation of the unit, reduce output or efficiency and cause
forced outages. The primary cause of excessive slagging is a direct result of operating
units at furnace exit gas temperatures (FEGT’s) higher than the ash fusion temperature
of the coal. FEGT is a major parameter affecting the form of the fly ash entering the
convective sections and must be reduced below the ash fusion temperature before
passing over the slope area in a boiler. However, boiler operating conditions can have
a considerable effect on FEGT and, consequently, on the slagging potential. The
amount of excess air, boiler load, the distribution of both fuel and air among the various
burners, and variability of the coal feed (mill performance, mill bias and mill out-of-
service conditions) are key factors affecting FEGT. In addition, sootblowing selectivity
and frequency are also critical parameters to prevent severe slagging problems from
occurring.

The objective of this study was to find optimized operating conditions that can
help to mitigate the slagging potential in a coal-fired unit. The investigation of operating
conditions included optimization of boiler operating parameters such as secondary air
register positions, excess O2 and mill biasing, as well as optimization of sootblowing
frequency and selection.

This paper describes the development of boiler operating settings and

sootblowing optimization for slagging control in a 700 MW wall-fired unit. As a result of
this study, boiler control settings for achieving the lowest FEGT and a time-based
sootblowing schedule were developed. The developed boiler settings and sootblowing
schedule were implemented by the utility and the results were summarized in this
paper.

1
Presented at the Twentieth Annual International Pittsburgh Coal Conference, Pittsburgh, Pennsylvania, September
15-19, 2003.

1
INTRODUCTION

Ash deposition that occurs in the section of the boiler where radiant heat transfer
is dominant is referred to as slagging, which is typically caused by the formation of
molten ash particles deposited on the upper furnace tube surfaces (1). The liquid form
of the ash deposits re-solidifies when hits the tube wall due to the quenching effect of
the tube surface (normally at a lower temperature than the furnace exit gas
temperature, FEGT). Once an initial deposit layer has formed, its surface tends to
become sticky, and most of the incident particles will be held on the surface, causing a
build-up.

Slagging also interferes with heat transfer and reduces local heat transfer to the
waterwalls. As a result of reduced heat transfer to the waterwalls, flue gas
temperatures at the exit of the furnace raise above the ash fusion temperature of fly
ash, worsening the slagging problem in the slope area and on the screen tubes in the
convection section of the boiler. Many slagging deposits form from coals that have
pyrite (FeS2) as a major mineral component (2). Iron content in the form of pyrite tends
to react with alumino-silicates (clays) to form low melting point compounds or low
viscosity particles which stick to the tubes. In addition, Calcium Oxide, (CaO), is known
to combine with the aluminosilicate fly ash particles, which is considered to contain
sufficient fluxing oxides to give rise to sticky ash particles and consolidation of the
deposit by viscous flow sintering (3).

The primary cause of excessive slagging is a direct result of operating coal-fired

units at FEGT’s higher than the ash fusion temperature of the coal. Boiler operating
conditions can have a considerable effect on FEGT and, consequently, on the slagging
potential. The amount of excess air, boiler load, the distribution of both fuel and air
among the various burners, and variability of the coal feed (mill performance, mill bias
and mill out-of-service conditions) are key factors affecting the FEGT. In addition,
sootblowing selectivity and frequency is also critical parameters to prevent severe
slagging problems from occurring.

2
 In this study, slagging potential in a 700 MW gross, opposed wall-fired unit (Unit A)
was investigated. On this unit, slagging is predominant in the upper waterwalls and
pendant surfaces, becoming severe in the nose region of the furnace. Typically, ash
plugs the lower region of the screen tubes and lays down on the nose. Furnace wall
blowers and several groups of retractable sootblowers were found to be partially
effective for controlling slagging, providing slag has been loosened and the slag layer is
not too thick. Sootblower groups located at the slope of the arch are blown more
frequently when slagging is observed to become severe. When this happens, however,
it may be too late to reverse the slagging problem. Several costly practices are used to
help control/recuperate from severe slagging, including dropping load overnight to shed
slag, on-line water lancing to remove slag and derating of the unit (typically by 150
MW).

The objective of this study was to find operating conditions that can help to
mitigate the slagging problem at the unit (Unit A). The operating conditions investigated
in this study included manipulation of boiler operating parameters such as secondary air
register positions, excess O2 and mill biasing, as well as an optimization of sootblowing
frequency and selectivity.

PLANT BACKGROUND

The Unit is a 700 MW gross, opposed wall-fired Foster Wheeler boiler. Figure 1
illustrates a cross-section of the boiler. The steam generator is a once through,
supercritical pressure, single reheat unit firing pulverized coal in a balanced draft
furnace. Full load steam capacity is 5,035,000 lb/hr at 3,810 psig superheater outlet
pressure with 1,010 oF superheat temperature and 4,500,000 lb/hr from the reheater
outlet at 890 psig with 1,010 oF reheat steam temperature.

The Unit is equipped with six Foster Wheeler (FW) D9F ball mill pulverizers that
supply coal to 24 burners arranged in three elevations on the front and rear walls, with

3
four burners per row. All mills were retrofit with Foster Wheeler fourth-generation low-
NOx burners and over-fire ports (OFA) in 1999. The OFA ports are always kept closed
due to possible adverse impacts on gypsum production via high fly ash LOI levels. Two
forced draft fans deliver the total combustion air. The boiler and auxiliary equipment
were designed to achieve full load capability for a mill out-of-service (O/S) condition.

Figure 1: Schematic of the Unit A

4
 Unit A fires relatively high sulfur, high volatile Eastern bituminous coals. High-
sulfur coals are fired in this unit because it is equipped with flue-gas-desulfurization
scrubbers that make gypsum, CaSO4.

FUEL ANALYSIS RESULTS

As a first step, series of coal and ash analyses was performed to identify
coal/ash characteristics that may be associated with the boiler slagging problem. The
recommended analyses included coal ultimate/proximate analyses, ash chemical
analysis and fusion temperature determination, and computer controlled scanning
electron microscopy (CCSEM) mineralogical analysis. Coal/ash chemical analyses
were also performed for another unit (sister unit, Unit B), which has an identical design
to the Unit under investigation (Unit A). The Unit B, however, does not suffer from acute
slagging problems. The coal/ash chemical analyses results for both units are presented
in Table 1. The coal analyses indicated that both fuels are relatively high sulfur, high
volatile, moderate ash coals. The Unit A coal is somewhat higher in sulfur on a lb per
MBtu basis (3.11 lb/MBtu versus 2.64 lb/MBtu). The Unit B coal is slightly higher in ash
content on a lb per MBtu basis. As mentioned in the previous section, Unit A requires
high sulfur coals because of gypsum production (CaSO4). However, if the sulfur in the
coal is present with iron as pyrite (FeS2), this may lead to slagging problems.
The ash chemistry analyses indicated that the Unit A coal is significantly higher in
iron oxide content than the Unit B coal (approximately 23 versus 15 percent). The
slagging potential of bituminous coals is generally known to increase with higher iron
content (4). Traditionally, the form of iron in coal producing ash rich in Fe 2O3 is known
to be primarily pyrite. The mineral analyses of both fuels also indicated significant
amounts of calcium oxide, CaO. CaO is known to combine with the aluminosilicate fly
ash particles, which is considered to contain sufficient fluxing oxides to give rise to
sticky ash particles and consolidation of the deposit by viscous flow sintering (5).

5
 Table 1: Summary of Coal/Ash Analyses

Coal Name Unit A Unit B

Proximate/Ultimate Analysis (%)
H2O 6.73 6.26
Ash (db) 10.38 14.20
VM 38.66 37.65
S 3.82 3.30
FC 44.52 45.77
Heating Value (BTU/lb)
HHV 12,260 12,500
Ash Analysis
SiO2 45.85 50.41
Al2O3 19.95 20.46
TiO2 1.08 1.05
Fe2O3 23.57 15.07
CaO 2.60 3.76
MgO 0.81 1.35
Na2O 0.53 1.03
K2O 1.71 2.45
SO3 2.22 3.75
P2O5 0.32 0.32
Ash Fusion Temperatures (°F)
Reducing atmosphere
Initial deformation (IDT) 2,003 2,129
Softening (ST) 2,101 2,176
Hemispherical (HT) 2,176 2,346
Fluid (FT) 2,262 2,372

CCSEM mineralogical analysis results for both fuels are presented in Table 2.
The analyses show that mineral compositions of both coals contain a high percent of
pyrite. The concentration of pyrite in the Unit A coal is higher than that of the Unit B
coal (67 versus 54 percent). The high concentration of iron-containing minerals in the
Unit A coal is expected to cause slagging problems due to the formation of low melting
iron compounds and also due to partially oxidized pyrite (if burned under low excess air
conditions) (5). Large partially oxidized pyrite particles are known to be sticky and to be
related to slagging. Table 3 lists the melting point of several iron compounds that
typically form in boiler furnaces. Table 3 suggests that, if the pyrite is oxidized

6
completely to Fe2O3, it has less chance of becoming fluid in the furnace and sticking to
heat transfer surfaces.

Table 2: Summary of CCSEM Mineralogical Analyses

Weight Percent

Grain size, m
Pyrite +
Quartz Kaolinite Calcite Pyrrhotite +
Oxidized Pyrrhotite
1.0 to 2.2 1.5 0.4 0 1.8
2.2 to 4.6 0.8 0.1 0 5.1
4.6 to 10.0 1.8 0.3 0.3 12.2
Unit A 10.0 to 22.0 0.5 0 1.4 20.3
22.0 to 46.0 0.2 0 0 12.8
46.0 to 400 0.2 0 0 15.4
Total 5.0 0.8 1.7 67.6
1.0 to 2.2 1.4 0.6 0 1.5
2.2 to 4.6 2.9 1.4 0 3.6
4.6 to 10.0 2 0.6 0 4.9
Unit B 10.0 to 22.0 0.8 0 0 7
22.0 to 46.0 0.5 0 0.1 12.2
46.0 to 400 0.1 0 0 24.9
Total 7.7 2.6 0.1 54.1

Table 3: Melting Point of Several Iron Compounds

Compound Formula Melting Point, °F

Partially Oxidized Pyrite FeS-FeO 1,700
Iron Silicate FeSiO3 2,095
Iron Oxide Fe2O3 2,850
Iron/Calcium Aluminosilicates Fe, Ca, Al, SiOx 2,300

CCSEM analysis also indicated that both coals contain relatively coarse pyrite
(>10 microns). Coarse pyrite promotes slagging, since the core of the particle may not
oxidize to a higher melting point oxide (Fe2O3). Poor mill performance may also lead to
slagging when firing coals containing high levels of pyrite. Coarse pyrite exiting the
mills will not oxidize as easily to a higher melting point state. In addition, unburned

7
carbon from oversize coal particles can cause locally reducing and high temperature
regions, when attached to the ash deposits.

The ash fusion temperature (AFT) is one of the most used parameters in
assessing propensity of coal ash to slag in a furnace. The AFT measures the softening
and melting behavior of ash at temperatures above 1,500°F. Figure 2 illustrates the
comparisons of AFT between the coals fired on Units A and B. The initial deformation
temperature (IDT) is the temperature at which the rounding of the tip of an ash cone is
noted, which has been accepted as the temperature where the ash first softens and
may become sticky. Gas temperatures entering tube banks are designed to be below
the IDT to avoid slagging and fouling problems.

According to AFT measurements in a reducing environment, the IDT of Unit A

ash is 126°F lower than that of the Unit B ash. Furnace exit gas temperature (FEGT)
measurements at Unit A have shown that typical temperatures (at full load conditions)
around the nose region are between 2,300 and 2400°F (FEGT measured by Infra-view
pyrometers by installed for this study), well above the ash fusion temperatures in a
reducing atmosphere. This temperature range covers the calculated average ash
softening temperature that would occur in an oxidizing atmosphere, 2,360°F. Maximum
FEGT measurements have been observed to be as high as 2,700°F. FEGT depends
upon the level of excess O2 in the furnace, mill loading configuration and burner
secondary air register settings. Periods of severe slagging would be expected when the
FEGT exceeds the fluid temperature of the ash.

8
 2,400
Pleasants
Unit A
Harrison
2,300

Unit B
Temperature, F
o

2,200

2,100

2,000

1,900
Initial deformation Softening (ST) Hemispherical Fluid (FT)
(IDT) (HT)

Figure 3: Ash Fusion Temperatures of Coals from Both Units

RESULTS

BOILER OPERATING PARAMETERS

The objective of testing for boiler operating parameters was to determine the
effects of excess O2, secondary air register openings, mill biasing, and mill and burner
out-of-service conditions on boiler parameters of interest including FEGT, main and
reheat steam temperatures, NOx and CO emissions, fly ash LOI, and slagging potential.
This was done to find combination of boiler control settings that could be used to
mitigate the slagging situation at Unit A.

Furnace Excess Oxygen

The Unit A furnace has four oxygen probes located at the economizer inlet (two
probes for each side). Discrepancies in excess O2 readings between the East- and the

9
West-side of the flue gas duct were observed, the West-side being almost 1.0 percent
higher than the East-side (2.9 vs. 3.9 percent, respectively).

Figure 3 shows FEGT as a function of average economizer excess O2. These

tests were performed at a unit load of 610 MW net with all six mills in-service. The trends
in Figure 3 indicate that FEGT’s on the both sides decrease by approximately 100 oF for
a 1.0 percent increase in economizer excess O2.

2,700 2,500
Secondary air register opening, 1/2/3/4 [%] = 80 / 90 / 60 / 40
2,650 Net Load [MW] = 610
2,450
2,600
FEGT West = -79.161*O2[%] + 2650.3
2,400

West FEGT [oF]

East FEGT [ oF]

2,550

2,500 2,350
FEGT East = -116.37*O2[%] + 2908.8
2,450
2,300
2,400
2,250
2,350

2,300 2,200
2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00 4.20
Average Economizer O2 [%]

Figure 3: FEGT as a Function of Average Economizer Oxygen (O2)

Figure 4 illustrates NOx emissions as a function of average economizer inlet O 2

concentration. The relationship between NOx emissions and excess O2 is linear and
indicates that NOx emissions increase by approximately 20 percent for a 1.0 percent
increase in average economizer O2. Main steam temperature is controlled by superheat
spray water and sootblowing, while sootblowing is the only means of controlling reheat
steam temperatures. The effect of excess O2 on main and reheat steam temperatures
was found to be of second order. The allowable range for both main and the reheat
steam temperatures is 980 oF to 1,020 oF. While the main steam temperature fluctuated

10
within  4 oF, the reheat steam temperature was reduced by approximately 10 oF for a
1.0 percent increase in excess O2. For the range of O2 levels tested in this study, both
main and reheat steam temperatures remained within the limits.

0.460
Secondary Air Register Opening, 1/2/3/4 [%] = 80 / 90 / 60 / 40
Net Load [MW] = 610
0.440
NOx [lb/MBtu] = 0.0716*O2 [%] + 0.1539
NOx Emission [lb/MBtu]

0.420

0.400

0.380

0.360
2.90 3.10 3.30 3.50 3.70 3.90 4.10 4.30
Average Economizer O2 [%]

Figure 5: NOx Emissions as a Function of Average Economizer Oxygen

(O2)

Secondary Air Register Position

Adjustments to the secondary air registers were done on each burner column,
rather than on each air register individually. Figure 6 illustrates the responses of FEGT
and economizer O2 as a function of secondary air register position. In the case of
Figure 3, a total of six air registers on Column 4 (including front and rear sides at the
West-side) were closed from an average position of 65 percent to 40 percent open. The

11
change in secondary air registers resulted in an increase of approximately 145 oF on the
West-side FEGT, while the FEGT on the East-side was slightly affected by this change.

7.5 2,600

7.0 FEGT East

2,500

6.5
2,400
Economizer Inlet O 2 [%]

6.0
FEGT West
2,300
5.5

FEGT [ oF]
O2 Probes
5.0 EAST WEST 2,200
B Right A Left A Right B Left B Right
4.5
2,100

4.0
A Right 2,000
3.5
B Left 1,900
3.0 A Left

2.5 1,800
9:07 AM 9:14 AM 9:21 AM 9:28 AM 9:36 AM 9:43 AM 9:50 AM 9:57 AM 10:04 AM 10:12 AM
Time
Figure 6: Effect of Secondary Air Register Positions (on Column 4) on
Economizer O2 and FEGT

The secondary air registers (SAR’s) were used to adjust the air flow rates and
the FEGT's between the East- and the West-sides. The adjustments to the secondary
air registers affect windbox pressure and cause a re-distribution of the secondary air
flow among the burners. Therefore, the inter-dependence among all secondary air
register openings was accounted for by a secondary air register bias definition. The
secondary air register bias is defined as:

SAR Bias = [1.5(SAR1-SAR4) + 0.5(SAR2-SAR3)]/[SAR1+ SAR2+ SAR3+ SAR4]

Where:
SAR1 Secondary Air Register Openings of the Burners on Column 1

12
 SAR2 Secondary Air Register Openings of the Burners on Column 2
SAR3 Secondary Air Register Openings of the Burners on Column 3
SAR4 Secondary Air Register Openings of the Burners on Column 4

Figure 7 shows FEGT variations as a function of the SAR bias. For the range of
SAR biases between 0.00 and 0.28, significant changes were observed in FEGT’s.
Opening the East-side air registers (increased positive bias) increased the secondary
air flow rate on this side and, therefore, resulted in reductions in FEGT on the East-side.
At the same time, this change caused a reduction on the secondary air flows available
for the West-side of the furnace. Therefore, while FEGT on the East-side dropped, the
West-side FEGT increased.

2,650 2,450
Net Load = 605 MW
Econ. O2 = 3.5 % (+ - 0.14 %) 2,430

2,630 2 2,410
y = 1005.8x - 534.74x + 2636.6

2,390

West-Side FEGT [ F]
East-Side FEGT [ F]

o
o

2,610 2,370
East
2,350
West
2,590 2
y = 423.9x + 476.18x + 2272.8 2,330

2,310

2,570 2,290

2,270

2,550 2,250
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30
Secondary Air Register Bias

Figure 7: Distribution of FEGT’s as a Function of Secondary Air Register Bias

It was found that secondary air register biasing have a second order impact on
steam temperatures and on superheat spray flow rate. The reheat steam temperature
was reduced by almost 10 oF in comparison to baseline conditions with all mills in-
service. Both main and reheat steam temperatures remained within the limits. The set
point for the main and the reheat steam temperature is 1,000 oF. Variations of  20 oF

13
from the set point are considered as normal operation. The superheat spray flow rates
were between 155 and 165 kpph during the tests. The superheat spray flow rate for
baseline condition is 155 kpph.

Mill Conditions

A series of tests was conducted to determine the effects of mill biasing, and mill
and burner-out-service conditions on FEGT and other boiler parameters. It was found
that mill biasing, within the allowable mill capacity limits have almost no impact on
FEGT and NOx emissions. However, the mill and burner-out-of-service conditions were
found to be effective in manipulating FEGT’s on both sides. Results indicated that
FEGT’s could be lowered by 60 (East-side) to 120 oF (West-side) by taking the top mill
out-of-service and adjusting the secondary air register positions.

The tests for burner out-of-service conditions were also conducted with all-mills
in-service. Two burners at the top elevation were shut down with the SAR position at
the minimum opening (to provide cooling air to the burners to avoid material damage).
Although, no noticeable changes were observed in FEGT’s, this burner out-of-service
condition caused a NOx reduction with respect to baseline conditions of the order of 12
percent (Figure 8). Burners out-of-service conditions have a little effect on main and
reheat steam temperatures (within  2 oF).

14
 0.420

Baseline (All Mills I/S)

SAR
0.400 Off A2 A3
NOx Emission, [lb/MBtu]

A2&A3 25 25
Others 85 50

0.380

SAR SAR
Off A1 A4 Off A1 A4
A1&A4 25 25 A1&A4 50 50
0.360 Others 85 50 Others 85 50

0.340

0.320
1 2 3 4
Test Number

Figure 8: Effect of Burner Out-of-Service on NOx Emission

SOOTBLOWER CHARACTERIZATION AND OPTIMIZATION

Sootblower characterization tests were conducted by activating IR’s and IK’s

(rotating and retractable sootblowers). Data were collected from the plant’s PI system
while the status of the slope and platen superheater tubes were periodically inspected
(visually) during the tests. Plant personnel set up most of the existing sootblower
groups based on advice provided by Foster Wheeler during the initial unit start up in
1978.

Using the data collected during sootblower characterization tests, the effect of
new and existing sootblower groups on FEGT’s, main and reheat steam temperatures,
superheat spray flow rate, and NOx emissions were investigated. Some individual
blowers in the original groups were found to be more effective than others in terms of
reducing FEGT. For this reason, new groups were formed to be able to activate the
most effective sootblowers more frequently, while using the others on an as-needed
basis. The effects of some sootblowers on FEGT’s for a 12-hour time period are
presented in Figure 9. As it can be seen from this Figure, some of the sootblower

15
groups had a large impact on FEGT, causing temperature reductions of the order of 100
o
F, while the other groups had almost no effect on FEGT. This indicates that the
deposition rates at various section of the boiler are different. Therefore, selectivity and
activation frequency of sootblowers (optimization) are very important for controlling
slagging.

2,600
Sootblowers B9 to B1 2
Sootblowers 1 to 8 B19 to B22 Group 9
2,550

East
2,500
FEGT [ oF]

2,450

2,400
West

2,350

2,300 Group 8
Sootblowers A9 to A12
A19 to A22

2,250
AM

AM

PM

PM

PM
PM
AM

PM
AM

PM

PM
12

48

00
36

36

12
24

12

24
:4

:0
7:

4:

6:
3:

7:
9:

1:

2:
8:

10

12
2

02

02
2

02
2
0

02

02
/0

/0

/0

/0
0

8/

8/

8/
8/

8/

8
8/

8/
7/

7/

7/

7/
7/

7/

7/

7/

7/
7/

7/

Figure 9: Effect of Various Sootblower Groups on FEGT’s

Based on the data collected during testing and visual observations, new
sootblowing groups were generated (see Figure 10).

16
 Figure 10: Developed Sootblowing Groups.

SUMMARY AND CONCLUSIONS

Coal and ash analyses helped to identify the nature of the problem. The key is to
maintain FEGT’s on both sides of the boiler relatively below a relative ash fusion
temperature (~2,350 oF) by manipulating boiler control parameters and sootblowing
activation while keeping parameters such as steam temperatures, NO x emissions, and
spray flow rates within prescribed limits. Three different boiler settings (All mills in-
service, and two-one mill out-of-service cases) together with an optimized, time-based
sootblowing schedule were recommended. It was recommended to operate the unit
with top elevation mill, feeding front burners, shut off for the lowest FEGT and NO x
emissions. Two other settings when all mills are in service were also recommended for
slag-free unit operations.

17
 Based on the results described in this study, new boiler control settings and a
sootblowing schedule were recommended for maximum FEGT reduction and slag
control at the upper furnace and slope area. The recommended sootblowing schedule
is shown in Table 3. The proposed sootblowing schedule requires continuous blowing
throughout the day since the unit has limited compressor capacity for sootblowers.

Table 3: Recommended Sootblower Schedule

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Group 1 1 1 1 1 1 1

Group 2 1 1 1

Group 3 1

Group 4 1 1 1 1

Group 5 1 1 1

Group 6 1 1

Group 7 1 1

Group 8 1

Group 9 1 1 1 1 1 1 1 1

Group 10 1 1 1 1

Group 11 1

Group 12 1

Night Shift Day Shift Night Shift

The Unit has been operating with the new boiler control settings and sootblowing
schedule since September 2002 and no slagging incidents have been reported since
then.

REFERENCES

1. Couch, G., “Understanding Slagging and Fouling During PF Combustion”, IEA

Coal Research, IEACR/72, IEA Coal Research, London, 1972.
2. Russell, N. V., Wigley, F., and Williamson, J., “The Roles of Lime and Iron Oxide
on the Formation of Ash and Deposits in PF Combustion”, Fuel, JFUE1744,
2001.
3. Gupta, S. K., Wall, T. F., Creelman, R. A., and Gupta, R. P., “Ash Fusion
Temperatures and the Transformations of Coal Ash Particles to Slag”, Fuel
Processing Technology, vol. 56, pp. 33-43, 1998.
4. Conn, R. Personal Communications.

18
5. Boni, A. A. et al., “Transformation of Inorganic Coal Constituents in Combustion
Systems”, DOE of U.S., Contract No. AC22-86PC90751, March 1990.
6. Seggiani, M., “Empirical Correlations of the Ash Fusion Temperatures and
Temperature of Critical Viscosity for Coal and Biomass Ashes”, Fuel, vol. 78, pp.
1121-1125, 1999.

19