Characterisation of intumescent coating performance …fire-research.group.shef.ac.uk/steelinfire/downloads/YZ_10.pdf · Characterisation of intumescent coating performance for performance-based

Characterisation of intumescent coating

performance for performance-based design

Yong Zhang

Prof. Yong.C. Wang

20-04-10

Background

• Standard fire test methods are used to assess intumescent coating.– treated as non-reactive materials.

– thermal properties are considered the same – thermal properties are considered the same during tests.

• Different fire exposures lead to different results.– Complex chemical reactions and phase

transition undergo when heated.

Background

• There is a strong demand for new, performance based assessing method.

– A model to predict the performance of

intumescent coatings under different fire intumescent coatings under different fire

scenarios.

• Steel temperature-time relationship

• Effective thermal conductivity.

– New, practical test methods.

Modelling

• A modelling has been built by my predecessor.

– final thickness was taken as

an input. an input.

– limit to specific fire

conditions.

Key Parameters

• Chemical factors– Composition (proportions of reactants)

– Chemical kinetic (Activation Energy, Pre-exponent factor)

• Thermal factors

TGA

• Thermal factors– Char structure (bubble size)

– Char size (final thickness)

• Other factors– Steel thickness

– Heat flux etc

– Pressure

Modelling

0.1

1

1.5

2

2.5

3

3.5

4

4.5

5

Ap

pare

nt

therm

al co

nd

ucti

vit

y (

W/m

.K)

Td

Q

∆=

*λ

d= final expansion thickness

Apparent thermal conductivity

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

0 100 200 300 400 500 600

Steel temperature (C)

Eff

ec

tive

th

erm

al c

on

du

cti

vit

y (

W/m

.K)

0

0.5

0 100 200 300 400 500 600

Steel temperature (C)

d= initial thickness

Effective thermal conductivity

Theory and Assumptions

• How to predict the final expansion thickness?

• The key issues:

– Q1: How does bubble grow?

– Q2:When does expansion stop?

– Q3: How much gas contributing to bubble growth?– Q3: How much gas contributing to bubble growth?

• Gas participation: mass loss + trapped gas = total decomposition

Initialise

Compute gas

generatedin each layer

Calculate gas

trapped among bubbles

Calculate

expansion growth

Determine

bubble burst andexpansion stop

Calculate volume

fractions of gas and polymer

Determine

material propertiesCalculate

temperature fieldendt>limit?Y

N

Q1Q2Q3

Theory and Assumptions

• Main assumptions

– Q1: How does bubble grow?

• Gases trapped in bubble follow ideal gas law

nrTPV =wS

mrTPx =i.e.

– Q2:When does expansion stop?

• Expansion stops when bubble bursts

nrTPV =wS

Px =i.e.

Bubble Theory

• Consider an individual bubble

• The forces across the bubble have

to be balanced according to the

Young-Laplace equation:•

Where ∆p is pressure difference , η -viscosity, σ-surface tension, R-bubble size and dR/dt -expand rate.

• Bubble bursts (so no further

expansion) when

R

R

RP

•

+=∆ησ 42

R

R

RP

•

+>∆ησ 42

From bitsandpieces.us

Bubble theory

• Combining the ideal gas law equation, after some

mathematical work

•

+= RRRdt

dmm

wPS

rkπηπσ 168*

2expansion force

Surface tension resistance

• The expansion is mainly due to gas release from

chemical reactions and temperature increase

• The resistances (viscosity &surface tension) are

mainly a function of temperature.

+= RRRdt

mwPS

πηπσ 168*

Viscosity resistance

Bubble theory

Barrier force

The extract

barrier force

curve is

unknown, we

only know the

trend

Bubble growth Bubble burst

It should be safe to say that bubble bursts when the

driving force reaches maximum

trend

Bubble Theory

• Answer to Q1&Q2:

Bubble grows following

the ideal gas law and

stops growing when

reaching maximum 0.00000

0.00005

0.00010

0.00015

0.00020

m(d

m/d

t) m(dm/dt)

dt

dmmreaching maximum

• Q3: How much gases

are trapped?

400 600 800 1000 1200

Tempdtm

Results from the past

• Furnace tests

– Three fire profiles

– Structural samples1200

0 1000 2000 3000 4000 5000 6000

0

200

400

600

800

1000

1200

Tem

p

t (s)

Iso Fire

Slow Fire

Fasf Fire

Gas trapped ratio (beta)

• So far, no suitable experimental solution is

found.

1.0

0.6

0.8

1.0

β=(423/T)n

200 400 600 800 1000 1200

0.0

0.5

be

ta

temp /k

β

200 400 600 800 1000 1200

0.0

0.2

0.4

0.6

be

ta

temp /k

β=(423/T)n

Beta relationship:gas trapped ratio

T temperature5.01

0

0x

T

T+

=β

T0 temperature start to release gas

x0 initial coating thickness

Different beta curves used for the modelling

Standard Fire

0

100

200

300

400

500

600

700

800

Th

ickn

ess

Te

mp

Experiment

Modelling

Thickness

Beam Web ST 2 initial thickness 1.116 mm

-5

0

5

10

15

20

25

30

35

40

45-17%

0 10 20 30 40 50 60

0

100

200

300

400

500

600

700

800

Te

mp

Experiment

Modelling

Thickness

Beam Web ST 4 initial thickness 1.167 mm

0

10

20

30

40

2%

0 10 20 30 40 50 60

Time

-5 0 10 20 30 40 50 60

Time

0 10 20 30 40 50 60

0

100

200

300

400

500

600

700

800

Thic

kne

ss

Mo

de

lling

Time

Experiment

Temp

Thickness

Beam Web ST 1 initial thickness 1.075 mm-10%

-5

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60

0

100

200

300

400

500

600

700

800

Tem

p

Time

Experiment

Modelling

Thickness

Column Web ST 5 initial thickness 0.964 mm

0

10

20

30

402.7%

Slow Fire

0

200

400

600

Te

mp

Web

Modelling

Thickness

Slow Fire 254 web initial thickness 0.615 mm-10%

-2

0

2

4

6

8

10

12

14

16

18

20

22

24

26

0

200

400

600

Te

mp

Web

Modelling

Thickness

Slow Fire 203 Web initial thickness 0.834 mm

0

5

10

15

20

25

30

35

40

45

-2%

0 10 20 30 40 50 60

Time

-2

0 10 20 30 40 50 60

0

200

400

Te

mp

Time

Flange

Modelling

Thickness

Slow Fire 203 Flange initial thickness 0.834 mm 11%

0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60

0

50

100

150

200

250

300

350

400

450

500

Th

ickn

ess

Tem

p

Time

flange

Experiment

Thickness

Slow fire 254 Flange initial thickness 0.615 mm

0

5

10

15

20

256%

0 10 20 30 40 50 60

Time

Fast Fire

0 10 20 30 40 50 60

0

200

400

600

800

Th

ickn

ess

Te

mp

Time

Experimental

Modelling

Thickness

Fast Fire 254 Web initial thicknes 0.569 mm

1%

0

5

10

15

20

25

0 10 20 30 40 50 60

0

200

400

600

800

Tem

p

Time

Flange

Modelling

Thickness

Fast Fire 254 Flange initial 0.569 mm 25%

0

5

10

15

20

TimeTime

0 10 20 30 40 50 60

0

200

400

600

800

Thic

kn

ess

Te

mp

Time

Flange

Modelling

Thickness

Fast fire 203 flange initial thickness 0.839 mm14%

0

5

10

15

20

25

30

35

0 10 20 30 40 50 60

0

200

400

600

800

Tem

p

Time

Web

Modelling

Thickness

Fast fire 203 web initial thickness 0.839 mm

0

5

10

15

20

25

30

35

12%

Experiments and results

•A cone calorimeter is employed for modelling development and improvement.

•versatile, small scales, quick experiments and

cost-effective.

•provide feedbacks and inputs for the modelling.

•can also be used for modelling validation.

•A furnace is also built to validate the model

Camera

PC

cone

steelpaint

scale

wool

•A furnace is also built to validate the model using structural samples.

Experiment and results

Experiments and results

40

50

60

70

80

90

100

110W

eig

ht

k=10

k=20

k=30

k=40

40

60

80

100

k=10

Modelling

Weig

ht

0 20 40 60 80 100

30

40

Time

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Temp

Component 1* Component 2 Component 3

Fraction (%) 28 17 55

E (kJ/mol) 54 100 60

A (s-1) 300 1.8e6 5

Component 1 Acid source

Component 2 Blow agent

Component 3 Charring agent

SamplesSample ID Steel thickness

mm

Initial thickness

mm

Heat Flux

kW/m2

SA150LP 5 0.35 50

SA250LP 5 0.37

SB150LP 10 0.79 50

SB250LP 10 0.73SB250LP 10 0.73

SC150LP 20 0.54 50

SC250LP 20 0.65

SA365LP 5 0.68 65

SA465LP 5 0.7

SC165LP 20 0.55 65

SC265LP 20 0.62

SC365LP 20 1.24 65

SC465LP 20 1.23

Results

0

100

200

300

400

500

600

Te

mp

/oC

modelling

measured

measured

modelling

sa150lp initial thickness 0.35 mm

0

5

10

15

20

25

30

Th

ickne

ss /m

m

50 kW/m2

-500 0 500 1000 1500 2000 2500 3000 3500 4000

0

time /s

0

-500 0 500 1000 1500 2000 2500 3000 3500 4000

0

100

200

300

400

500

thic

kne

ss /m

m

tem

p /

oC

time /t

measured

modelling

modelling

measured

0

5

10

15

20

25

30

35

40sb150lp initial thickness 0.8 mm

-500 0 500 1000 1500 2000 2500 3000 3500 4000

0

50

100

150

200

250

300

350

400

450

Th

ickn

ess /m

m

Te

mp

/oC

time /t

modelling

measured

modelling

measured

sc150lp initial thickness 0.55 mm

0

5

10

15

20

25

Results

0

100

200

300

400

500

Te

mp

modelling

measured

measured

modelling

sc165lp initial thickness 0.55

-2

0

2

4

6

8

10

12

14

16

18

20

22

24

26

65 kW/m2

-500 0 500 1000 1500 2000 2500 3000 3500 4000

0

50

100

150

200

250

300

350

400

Tem

p

time

modelling

measured

measured

modelling0

10

20

30

40

sc365lp intial thickness 1.25 mm

-500 0 500 1000 1500 2000 2500 3000 3500 4000

time

-2

-500 0 500 1000 1500 2000 2500 3000 3500 4000

0

100

200

300

400

500

600

700

Th

ickn

ess /m

m

Te

mp /

oC

time /s

modelling

measured

measured

modelling

sa365lp initial thickness 0.7mm

0

5

10

15

20

25

30

35

40

ResultsSample ID Steel thickness Initial thickness Final thickness E R. Modelling

E.R.

Diff.

SA150LP 5 0.35 23 65 59 6%

SA250LP 5 0.37 20 54

SB150LP 10 0.79 40 50 48 4%

SB250LP 10 0.73 38 52

SC150LP 20 0.54 20 37 40 3%SC150LP 20 0.54 20 37 40 3%

SC250LP 20 0.65 14 21

SA365LP 5 0.68 25.6 38 40 5%

SA465LP 5 0.7 26 37

SC165LP 20 0.55 20.4 37 38 4%

SC265LP 20 0.62 26.6 43

SC365LP 20 1.24 35.3 28 27 5%

SC465LP 20 1.23 33 27

Summary

• New assessing method based on this model

– Inputs for the model

• TGA tests (Reactant fractions, chemical kinetic, T0)

• Mean bubble size• Mean bubble size

• Cone Tests for a beta relationship

– the model can be used to predict

• Steel-temperature relationship

• Effective thermal conductivity

Future works

• More cone and furnace tests to validate and improve the model.

– Different affecting parameters

• More products from the market• More products from the market

– Leigths Paint

– Nullifire Fire