Fuel flexible power stations: utilisation of ash co-products as

additives for NOx emissions controlRichard Birley

Supervisors

University of Leeds: Prof JM Jones, Prof A Williams, Dr LI Darvell, BF2RA: Dr J Ashman

School of Chemical and Process EngineeringFACULTY OF ENGINEERING

To establish if the use of biomass pulverised fly ash (PFA) and/or furnace bottom ash (FBA)

can be used as an additive during coal combustion to reduce NOx in large scale

power production.

Ash additives applied at 15% w/w to study theeffects on combustion and devolatilisation

Establish the effects of the additives on a range of fuels with varying reactivity

Aim and objectives of the research

• The use of fuel flexible furnaces are being used for efficiency and emissions control

• Industrial Emissions Directive has set targets of 150 mg/Nm3 for 2020

• NOx can lead to:• Acid rain

• Increased tropospheric ozone

• Increased particulate matter

Justification for this research

The ashes have high levels of potential reactive elements

• Alkali metals K, Na

• Alkaline earth metal Ca, Mg

• Unburnt carbon

The reactive elements may act as catalysts as a mechanism for the reduction of NOx during coal combustion

Other considerations for catalysis are iron contents (although low contents in biomass)

Concept

Tar N

Char N

Vol/Gas NNH3/HCN

Fuel N

HCN/NH3

CN O2/O2-

Fuel Rich CxHy/C

N2/CO2

NO

CxHy

NOO2/O2-

λ<1

Catalysis on surface of ash compounds or char

Tar cracking to volatile N on surface of reactive compounds

O2/O2-

CO2/N2/HCN

Evolution of fuel nitrogen

Catalysis on surface of ash compounds or char

Carbon a

Hydrogen a

Nitrogen a

Sulphur a

Oxygen a,c

Volatiles a

Fixed carbon ac

Ash a

Moisture b

Fuel ratio

Ffos-y-fan 78.39 3.07 0.92 0.80 12.52 8.09 87.61 4.30 3.86 10.83

Shotton 78 4.1 1.7 0.98 5.40 32.31 57.87 9.82 7.4 1.79

La Loma 64.30 4.70 1.30 0.50 17.06 38.78 49.08 12.14 5.40 1.27

Galatia 69.3 4.5 1.7 1.1 13.08 40 49.68 10.32 12.4 1.24

FBA 39.48 1.99 0.12 0.00 0.44 25.74 16.29 57.97 2.4 3.28

PFA 4.50 0.07 0.03 0.10 7.62 2.88 9.44 87.68 0.4 0.63

Olive cake 46.81 6.15 2.70 0.35 34.12 72.29 17.84 9.87 13.41 0.25

Coal PFA 3.54 0.05 0.00 0.00 0.47 2.26 1.80 95.94 0.45 0.80

a=dbb=arc=calculated by difference

Fuel ratio=𝐹𝑖𝑥𝑒𝑑 𝑐𝑎𝑟𝑏𝑜𝑛

𝑉𝑜𝑙𝑎𝑡𝑖𝑙𝑒𝑠

Proximate and elemental analysis

Experimental procedure

Fuel characterisation

Combustion fuel/ fuel +15% w/w additives

Devolatilisationfuel/ fuel +15% w/w additives

NO

Volatile NGas phase N

Char NCarbon conversion

Drop tube furnace

Heated to 1373K

Heating rates 104 – 105 K sec-1

Residence time 0.75 sec

Combustion

Air atmosphere

DTF at Northeastern University (NEU)

NOx emissions through combustion

X. Ren, R. Sun, X. Meng, N. Vorobiev, M. Schiemann, and Y. A. Levendis, "Carbon, sulfur and nitrogen oxide emissions from combustion of pulverized raw and torrefied biomass," Fuel, vol. 188, pp. 310-323, // 2017.

0

100

200

300

400

500

600

700

Coal Coal pls FBA Coal plus PFA

NO coals (ppm)

Shotton

La Loma

Galatia

0

1

2

3

4

5

6

7

8

9

Coal Coal pls FBA Coal plus PFA

NO coals (mg/g dry fuel)

Results: Combustion NO emissions

characteristics (DTF)

During combustion in a DTF:

• Two methods of analysing the raw data has been

shown, ppm and mg/g of dry fuel

• Galatia and La Loma show reductions of ~10%

with the addition of PFA

Results: Combustion NO kg/GJ as

a function of fuel ratio

y = -170.2x + 531.55R² = 0.896

y = 97.463x2 - 426.9x + 674.19R² = 0.9747

y = 124.21x2 - 418.37x + 583.73R² = 0.8334

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

NO

Kg

/GJ

dry

co

al

Fuel ratio

NOx emissions kg/GJ (dry coal) as a function of Fuel Ratio

Coals

Coals plus PFA

Coals plus FBA

Olive cake

Olive cake plus coal PFA

All fuels no additives

All fuels plus PFA

Linear (Coals plus FBA)

Poly. (All fuels no additives)

Poly. (All fuels plus PFA)

Fuels no additives

Fuels with PFA

Fuels with FBA

Results: Combustion NO emissions

compared to (Ca+Mg+K+Na)/N

y = 31.854x2 - 520.07x + 2362.6R² = 1

y = 1.1261x2 + 2.1625x + 208.09R² = 0.9583

y = 5.2901x2 - 113.46x + 821.96R² = 0.9916

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00

NO

Kg

/GJ

dry

co

al

∑ Basic oxides/N

NOx emissions kg/GJ (dry coal) as a function of (Ca+Mg+K+Na)/N

Coals

Coals plus PFA

Coals plus

Olive cake

Olive cake plus coal PFA

All fuels no additives

Fuels plus PFA

Poly. (Coals plus)

Poly. (All fuels no additives)

Poly. (Fuels plus PFA)

Fuels no additives

Fuels with FBA

Fuels with PFA

Devolatilisation and N partitioning at

the University of Leeds (UoL)

a. DTF Heated to 1373K

Heating rates 104 – 105 K sec-1

Residence time 0.50 sec (UoL)

N2 atmosphere, 2% O2

b. TGA heated to 1273K

Heating rate 33K sec-1

N2 atmosphere

McNamee P., Darvell L.I., Jones J.M., Williams A. The combustion characteristics of high-heating-rate chars from untreated and torrefied biomass fuels. Biomass and Energy, vol 82 pp63-72.

Low N furnaces with NOx control

For this research we are

mainly concerned with the

primary controls

Primary

• Low NOx burners

• Air/burner staging

• Overfire air

• Stoichiometry

• Furnace temperature

Secondary

• SNCR

• SCR

Hot gases out (NOx)To: SCR/SNCR

Garner L. Experience with low NOx burners. IEACR/99. London, UK. 1997

0

10

20

30

40

50

60

70

80

90

0 10 20 30 40 50 60 70 80 90Hig

h t

emp

erat

ure

vo

lati

le y

ield

(d

af%

)

Proximate volatile yield (daf%)

Comparison of high temperature volatile yields to proximate volatile yields

ICSTM CRE IFRF ICSTM DTI#208 Hemweg

ICSTM DTI #208 Australian Sub-bituminous coals UoL Linear 1:1 Coals plus FBA

Coal PFA Olive cake Olive cake plus PFA Linear (ICSTM)

Linear (IFRF) Linear (IFRF) Linear (Linear 1:1)

Results: High temperature volatile yield

compared to proximate volatile yield

Adapted from- Dr L. King . Doosan Babcock How UK thermal power plant cleaned up their act......for what future? 65th Energy Science Lecture, University House, University of Leeds 20th September 2016

Sub-bituminous coals volatile increase ~5-10%Anthracite increase 18-30%FBAPFA

y = 1.2588x - 19.185

y = -0.2762x + 77.595

y = -0.0241x2 + 3.6309x - 48.408

0

10

20

30

40

50

60

70

80

90

20 30 40 50 60 70 80 90

HH

R v

ola

tile

N %

Total volatiles (daf %)

HHR volatile N release compared to total volatiles

Coal 1273K Coal 1873K Sub-bituminous coals with PFA UoL

Sub-bituminous coals FBA UoL Fuels UoL Fuels plus PFA 1273K UoL

1:1 Linear (Coal 1873K) Linear (Sub-bituminous coals FBA UoL)

Poly. (Fuels plus PFA 1273K UoL) Linear (1:1)

Fuels no additives

Fuels with PFA

Fuels with FBA

Results: TGA HHR volatiles compared

to total volatiles heating rate 33K sec-1

Adapted from Gibbins JR, et al (1995a). Implications of nitrogen release from coals at elevated temperatures for NOx formation during pf combustion. In coal science proceedings of the 8th international conference in coal science. Oviedo, Spain, 10-15 Sept 1995. Amsterdam, Netherlands, Elsevier Science BV, vol 1, pp755-758 (1995)

Results: Devolatilisation Comparison

of N partitioning to carbon conversion

y = 1.0559x

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

Vo

lati

le N

rel

ease

%

Carbon conversion %

Volatile N compared to carbon conversion

Low Volatile coals Pittsburgh Sub-bituminous 1 Sub-bituminous UoL

1:1 Linear Sub-bituminous UoL FBA Sub-bituminous UoL PFA Olive cake

Olive cake plus PFA Linear (Sub-bituminous 1) Linear (1:1 Linear)

Kambara S, et al (1995). Relationship between functional forms of coal nitrogen and NOx emissions from pulverised coal combustion. Fuel 74(9) 1247-1253. And others

FBA

PFA

Discussion points

Tar N

Char N

Vol/Gas NNH3/HCN

Fuel N

HCN/NH3

CN O2/O2-

Fuel Rich CH/C

N2/CO2

NO

CxHy

NO

CO2/N2/HCN

O2/O2-

λ<1

Catalysis on surface of ash compounds or char

Tar cracking to volatile N on surface of reactive

compoundsO2/O2-

()

()

Catalysis on surface of ash compounds

or char

Conclusion

There is experimental evidence for the additive catalysing:

N-release during devolatilisation, where a marked

effect was observed

Increased conversion of char N to volatile N

Carbon burn-out

Enhanced release of volatile-N is beneficial for NOx

reduction strategies through air staging.

The trade off between carbon burnout and NO reduction on

char can impact the measured NOx emission from the char

combustion stage (not shown)

Acknowledgement

Supervisors UoL: Prof J Jones, Prof A Williams, Dr L Darvell

Industrial: Dr J Ashman

Prof Y Levendis, Aidin Panhari and Emad Rokni

Northeastern University Boston

Consultant: Dr D Waldron

BF2RA grant 23

EPSRC funding grant-EP/L014912/1