COMBUSTION OF LOW VOLATILECOALS

Application to Wall FiredPower Plant

PROJECT SUMMARY 344

OBJECTIVES

• To demonstrate the combustion performance of alow volatile coal burner for wall fired furnaces bytesting of a full-scale burner design in a singleburner test facility.

• To optimise the furnace design and process forutility plant and burner retrofits with regard tocombustion efficiency, NOX emissions, andstability/turndown performance by means of pilotscale testing and advanced modelling techniques.

SUMMARY

The main aim of the project was to demonstrate theproposed low volatile burner at full scale in a singleburner test facility and to establish its performancewith regard to NOX and combustion efficiency.Furthermore, pilot scale testing was undertaken toquantify the effect of the staged addition ofcombustion air. Air staging is generally regarded asan effective, mature technology for NOX reductionfrom bituminous coals, but its impact on low volatilecoals is less well understood.

Flame Stability is a major concern when firing lowvolatile coals. In light of this, laboratory scale teststo assess the flame stability properties of coalswere undertaken by Imperial College London. Themain analyses carried out were High TemperatureWire Mesh (HTWM) and Explosibility. The HTWMtests confirmed that the Q factor, (the ratio of hightemperature volatile yield to the proximate volatilecontent of the coal) remains constant throughoutdevolatilization. This is useful for predicting thevolatile release during devolatilization processes athigh temperatures.

CLEANER FOSSIL FUELS PROGRAMME– CLEANER COAL R&D PROGRAMME

Additional testing was takenon Mitsui Babcock EnergyLtd's NOX Reduction TestFacility (NRTF) to establishthe effect of the mainoperating variables on NOXand burnout under baselineand air staging conditionsfor a low volatile coal. Theresults indicate that furthersignificant NOX reduction isachievable when firing lowvolatile coals using MitsuiBabcock wall burners withan air staging system.

BACKGROUND

Traditionally low volatilecoals and anthracites havebeen utilised in arch firedfurnaces (often referred toas ‘downshot’ firing) so asto overcome the inherentdifficulties of achievingstable and efficientcombustion which arisefrom the lack of volatilematerial to aid in theignition, and the lowreactivity of the remainingchar. The downshot firingsystem is, however, ofhigher initial cost than acomparable wall firedsystem, and if it werepossible to utilise lowvolatile coals in wall firedfurnaces there are cleareconomic benefits. Wherethe wall firing of low volatilecoals has been attempted inthe past (predominantly inthe former Soviet Union) theexperience has been poor -excessively high carbon inash (typically 20 to 30%),high NOX emissions (of the

Figure 1. Low Volatile Coal Flame (courtesy of Mitsui Babcock EnergyLimited)

The Explosibility tests were found to provide a new wayto assess coal ignition coal properties, and supported theconclusion that early release volatiles are crucial for flamestability.

The testing and successful demonstration of theproposed low volatile burner was achieved on MitsuiBabcock Energy Limited's multi-fuel burner test facility.Two low volatile coals, volatile content of ~15% and~10%, were fired during the testing programme. Bothcoal tests achieved stable and unsupported low NOXflames.

Table 1. Coal properties

Coal Chinese UK

Moisture, % as

received

0.8 0.6

Volatile, % as

received

12.6 9.5

Fixed Carbon, %

as received

71.1 82.5

Ash, % as

received

15.5 7.4

GCV, MJ/kg 26.66 32.99

order of 1000 to 1300 mg/Nm3 @ 6% O2),and poor stability and turndownperformance (oil support of around 10% ofthe thermal input being required to sustainthe combustion, even at maximum boilerload). Such a level of performance isunacceptable from both an environmentaland an economic perspective and there is aclear requirement for a burner that can beretrofitted to existing plant in addition toneed for new installations (to extend therange of coals that can be utilised in thelower cost wall fired combustion system).

The overall objective is therefore todemonstrate that low volatile coals can besuccessfully exploited in wall fired plant. Inthe long-term this can only be achieved bymeans of full-scale plant demonstration, butan essential step towards that goal is toprove the performance of a single full-scaleburner and to show how it can be integratedinto an overall firing system.

In phase 1 of the project the keymechanisms for ignition and stabilisation oflow volatile coal flames were identified, andthese were applied in the development of aburner design that could be installed in wallfired plant. To develop the burner processdesign Computational Fluid Dynamics (CFD)combustion modelling, supported bydetailed coal characterisation data (e.g. hightemperature volatile yield and char reactivity)provided by Imperial College London(Imperial), was used. Detailed analysisallowed the optimisation of the variousdesign parameters within the constraints ofpractical plant implementation to beidentified. It was shown that the proposedburner design is likely to have a considerablybetter combustion performance thancompared to the current Soviet design.

RESULTS

The laboratory scale testing investigated lowvolatile coals by conducting characterisationtests in the Imperial College London

Figure 2. High Temperature Wire Mesh Reactor(courtesy of Imperial College London)

laboratory under conditions that are relevantto PF combustion (i.e. small particles, hightemperatures and high rates of heating).

TGA

TGA (Thermo Gravimetric Analysis) non-isothermal analysis is a relatively simple andinexpensive method to characterize coalcombustion properties. Three different kindsof profiles can be obtained using the TGAequipment: Thermal-Gravimetric (TG)showing mass loss as a function oftime/temperature; Derivative TG (DTG), thederivative of TG curves showing the rate ofweight loss as a function of temperature;and Differential Thermal Analysis (DTA)giving the temperature difference betweenthe sample and an inert reference materialas a function of sample temperature.

HIGH TEMPERATURE VOLATILEYIELD

The Q factor is a way to relate slow heatingdevolatilization (from the TGA) and rapidheating devolatilization (from the HTWM).The Q factor is defined as the ratio betweenthe rapid heating volatile yield and thechange in proximate volatile matter.

Imperial College

The key feature of the Q factor is that it'sbeen found to be nearly constantthroughout devolatilization for a wide rangeof coals. Thus while more volatiles arereleased under rapid heating conditions thanin the proximate test, due to capture in andwithin the particles themselves, proportion‘lost’ in the proximate test is constant. Theconstant feature of Q factor providesanother means to calculate the volatilerelease and has a potential to be moreaccurate than the ash trace technique,particularly at lower extents ofdevolatilization and for low ash coalsamples.

Figure 3. DTG test plot (courtesy of Imperial CollegeLondon)

EXPLOSIBILITY TEST

The explosibility test is undoubtedly usefulto understand the role of volatiles in theignition process, with the environmentproduced inside the explosibility chamberregarded as being close to the PFcombustion environment. To investigate theextent of devolatilization when coal dustwas ignited, TGA analyses were carried outon char samples produced. While sampleswere not ideal due to ignitor contamination,the trends observed and new residuecharacterization methods developed were ofvalue in understanding the ignition process.The coal characterisation undertaken wasvaluable in the selection of the coals to betested in the full-scale burner, and forms thebasis of a laboratory scale approach to

assessing the impact of coal properties onignition and flame stability.

Figure 4. Explosion test bomb chamber (courtesy ofNIOSH)

FULL SCALE BURNER TESTS

The proposed low volatile coal burner designwas developed from existing wall firedburner technology. The burner is based onthe ‘internal’ air staging concept where thecombustion air is introduced concentricallyaround the primary air / pulverised coal in aseries of separate annuli. The relativeproportion of air supplied to the threecombustion air annuli (Secondary, Tertiaryand Quaternary) is controlled by sleevedampers, which form the integral part of theburner design. The provision of swirledcombustion air, in combination with the bluffbody effect of the core air tube gives rise toa strong recirculation zone immediately infront of the burner - the hot recirculatedcombustion products provide the heatsource required to initiate the devolatilizationand stabilise the combustion process.

The objective of the testing was to definethe best achievable performance in terms offlame stability and turndown, combustionefficiency and emissions of NOX for theburner when firing low volatile coals.Parameters investigated during these testsincluded excess air level, burner damper andswirl settings, and firing rate.

Testing was performed on the MBTF firing aChinese Low Volatile coal (15% daf volatilematter). NOX emissions of 597 mg/Nm3 at6% O2 and Unburnt losses of 4.3% GCVwere achieved when the burner was set atits optimum geometry.

Figure 5. Burner Schematic (courtesy of MitsuiBabcock Energy Ltd)

The burner was further tested firing a UKLow Volatile coal of 10% daf volatile matter.These tests were also successful with NOXemissions of 436 mg/Nm3 at 6% O2 andUnburnt losses of 11.7% GCV when theburner was set at it's optimum geometry.These unburnt loss figures of 4.3% GCVand 11.7% GCV appears at first sight to behigh, but it does meet commercial plantlimits when differences between the MBTFand plant burner operating Stoichiometrylevels are taken into account.

Figure 6. Plot of NOX vs. O2 (courtesy of MitsuiBabcock Energy Ltd)

When firing the Chinese coal a stableunsupported flame was achieved to aturndown value of 50% MCR, where as thelower volatile UK coal firing achieved a

stable unsupported flame to a turndownvalue of 70% MCR. The turndownperformance was demonstrated to beacceptable, and the results indicate a goodlevel of inherent flame stability for thisburner.

AIR STAGING TESTS

Testing was also performed on the NRTFfiring a Chinese Low Volatile coal (7.1% dafvolatile matter) to establish the effect of themain operating variables on NOX andburnout under baseline and air stagingconditions. The tests undertaken consideredthe effects of excess air, primary zonestoichiometry and residence time, and wereselected to be relevant to a typical moderndesign wall fired furnace for this coal. Thebaseline NOX was 810 mg/Nm3 @ 6% O2 forthis coal. Reducing primary zonestoichiometry to 1.0 and 0.9 reduces NOX to644 and 477mg/Nm3 @ 6% O2 respectively.

The NOX reduction achieved by loweringprimary zone stoichiometry demonstratesthat air staging is an effective NOX reductiontechnology for this coal.

CONCLUSIONS

• As a result of the work carried out to dateit has been proven possible to ignite andfire coals having volatile content as low as10% daf in a wall-fired burner without theuse of a support fuel. Furthermore, it ispossible to achieve ignition and sufficientdevolatilisation in that part of the nearburner region where the suppression ofNOX formation occurs through staging toachieve reductions in NOX emissions.However, the NOX levels resulting fromthe firing of low volatile coals aregenerally higher than those resulting frombituminous coals under similar firingconditions.

Secondary AirQuaternary AirTertiary Air

Primary Air & Fuel

Core Air

Swirl Generators

MBEL Advanced Low NO x Burner

Further information on the Cleaner Fossil Fuels Programme, and

copies of publications, can be obtained from:

Cleaner Fossil Fuels Programme Helpline, Building 329, Harwell International Business Centre, Didcot, Oxfordshire OX11 0QJTel: +44 (0)870 190 6343 Fax: +44 (0)870 190 6713E-mail: helpline@cleanercoal.org.ukWeb: www.dti.gov.uk/cct/

© Crown copyright. First printed August 2005.Printed on paper containing a minimum of 75% post-consumer waste. DTI/Pub FES 05/955

COST

The total cost of this projectis £419,543, with theDepartment of Trade andIndustry (DTI) contributing£159,427. The balance offunding was provided by theparticpants.

DURATION

24 months - July 2002 toJuly 2004

CONTRACTOR

Mitsui Babcock EnergyLimitedTechnologyPorterfield RoadRenfrewPA4 8DJTel: +44 (0) 141 886 4141Web:www.mitsuibabcock.com

COLLABORATORS

Imperial College London

• Appreciable NOX reductions are achievable with airstaging when firing low volatile coal using new designwall burners, even with a modest degree of airstaging.

• Analytical techniques can be used to aid theassessment of low volatile coals for wall burner firing.Not only can Imperial College London apparatusprovide the volatile content and peak devolatilizationrates of a particular coal, they can also provideinformation to help understand the ignition process ofthat coal. In particular the Explosibilty tests were foundto provide a new way to assess coal ignition coalproperties and supported the conclusion that earlyrelease volatiles are crucial for flame stability and theHTWM tests confirmed that the Q factor, that links theTGA and HTWM tests, remains constant throughoutdevolatilization. This is useful for predicting the volatilerelease during devolatilization processes at hightemperatures.

POTENTIAL FOR FUTURE DEVELOPMENT

• Firing a low volatile coal with volatile matter ~5% dafon the modified 70MW Mitsui Babcock Energy Ltdwall burner.

• Full Scale plant demonstration of the modified 70MWMitsui Babcock Energy Ltd wall fired burners firing lowvolatile coal.

• Further develop coal characterisation utilizing anexplosion chamber that has minimal ignitorcontamination.