A Two-Step Reaction Mechanism for Combustion of Polymeric Solids

Containing Flame Retardants

Haiqing Guoa, Richard E Lyonb , Natallia Safronavac, Richard N Waltersb, Sean Crowleyb

a C-FAR Services, 303 Quail Drive, Marmora, NJ

b FAA WJHTC Atlantic City International Airport, NJ c Technology and Management International, LLC ,1433 Hooper Ave, Toms River, NJ

Background A ban of highly effective bromine containing flame-retardant

compounds that are widely used in aircraft cabins.

Advance in phosphorous containing flame-retardant compounds as bromine replacement.

Use of MCC (micro scale combustion calorimeter) as a tool for gas-phase combustion kinetics study (one-step global reaction).

MCC is modified for measurements of CO and CO2.

2

Objectives

Obtain incomplete combustion of polymeric solids containing flame retardants in MCC by varying the combustor temperature.

Implement a 2-step reaction mechanism, including a CO generation and a CO oxidation step, to describe the kinetics in the combustor.

Correlate the kinetics from MCC test with cone calorimeter test results.

3

MCC Modification

4

Combustor Temperature Combustor temperature is not perfectly uniform at each set point. Surface fitting was performed with the measured temperatures at varied

locations and temperature set points. Pyrolyzer temperature is maintained at 430 °C.

0900

200

850

400

25800

Com

bust

or T

empe

ratu

re (

oC

)

600

20

Set Temperature (o

C)

750

800

15

Axial Distance (cm)

1000

700 10650 5

600 0

5

2-Step Reaction Kinetics The kinetics in the MCC combustor involves a 2-step reaction mechanism.

Both reactions are second-order: first-order dependent on the fuel and the oxidizer concentration.

Wall effect is negligible.

Heat release through chemical reaction does not affect combustor temperature.

The molar number change in the combustor is neglected.

6

22

22

2

1

21 COOCO

productOHCOOFuel

k

k

→+

++→+

2-Step Reaction Kinetics MCC measures species (O2, CO and CO2) molar fraction in a dry-basis.

Water is corrected based on molar balance. The combustor is divided into 200 elements and the residence time in

each element is evaluated with the bulk velocity at ambient temperature and corrected with local temperature.

The initial fuel concentration into the combustor is estimated with the CO2 measurement at the complete combustion condition.

Species concentration in the next time step is updated with d[i]/dt and also accounts for temperature-related volume change.

Evolution of the species within the combustor are simulated following the 2-step mechanism using MATLAB.

The appropriate value for the pre-exponential factor (A) and the activation energy (E) are obtained by minimizing the error in fitting the measured and the simulated species concentrations (i.e., O2, CO, and CO2) 7

Materials

8

Sample Formula Structure

Polystyrene PS C8H8

Decabromodiphenyl oxide

DBDPO C12Br10O

Triphenylphosphine oxide

TPPO C18H15PO

9,10-dihydro- 9-oxa-10-phosphaphenanthrene-10-oxide

DOPO C12H9PO2

9

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-4

-3

-2

-1

0

1

2

3

4

Yie

ld (

g / g

)

PS

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 2.00e12

A2 = 5.01e27

E1 = 230,000

E2 = 532,000

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-4

-3

-2

-1

0

1

2

3

4

Yie

ld (

g / g

)

PS + 10% Br

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 2.51e09

A2 = 1.26e11

E1 = 178,000

E2 = 234,000

Results _ Bromine Containing

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-4

-3

-2

-1

0

1

2

3

4

Yie

ld (

g / g

)

PS + 15% Br

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 7.94e06

A2 = 2.24e09

E1 = 136,000

E2 = 204,000

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-4

-3

-2

-1

0

1

2

3

4

Yie

ld (

g / g

)

PS + 20% Br

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 2.24e07

A2 = 2.82e08

E1 = 146,000

E2 = 188,000

Results _ Phosphorous Containing

10

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-4

-3

-2

-1

0

1

2

3

4

Yie

ld (

g / g

)

PS

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 2.00e12

A2 = 5.01e27

E1 = 230,000

E2 = 532,000

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-3

-2

-1

0

1

2

3

Yie

ld (

g / g

)

DOPO

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 7.08e14

A2 = 3.16e35

E1 = 296,000

E2 = 716,000

550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o

C )

-3

-2

-1

0

1

2

3

Yie

ld (

g / g

)

TPPO

CO Predict

CO Measure

CO2

Predict

CO2

Measure

O2

Predict

O2

Measure

A1 = 5.62e29

A2 = 2.82e39

E1 = 582,000

E2 = 784,000

Results _ CO Oxidation The second reaction is examined with pure CO burning in the

combustor. N2 line was replaced 4.06% CO mixed in N2 balance.

11

500 550 600 650 700 750 800 850 900 950 1000

Combustor Set Temperature ( o C )

18.4

18.6

18.8

19

19.2

19.4

19.6

19.8

20

20.2

20.4

Exh

aust

O2

( %

)

CO

Predict

Measure

A2 = 1.05e40

E2 = 830,000

Fuel Dependence

The measured CO molar fraction should be proportional to the initial sample mass if the proposed 1st order fuel dependence is satisfied.

Combustor temperature is fixed at 675 °C

12

0

1

2

3

4

5

6

7

350 400 450 500 550

CO M

olar

Fra

ctio

n (%

)

Pyrolyzer T (°C)

m1 = 2.02 mg

m2 = 2.62 mg

m3 = 3.33 mg

0.0

0.5

1.0

1.5

2.0

350 400 450 500 550

Nor

mal

ized

CO

(%

/mg)

Pyrolyzer T (°C)

m1 = 2.02 mg

m2 = 2.62 mg

m3 = 3.33 mgPS PS

Oxygen Dependence

PS was tested at T=690 °C with varied initial O2 molar fraction. Reasonable agreement was observed.

13

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 5 10 15 20 25 30 35 40 45

Yiel

d (g

/g)

Oxygen Molar Fraction (%)

T_combust = 690 °C

CO_measure

CO2_measure

CO_predict

CO2_predict

Reaction Rates _ Bromine

Presence of bromine suppresses both reactions. The effect on the 2nd reaction is larger. The bromine becomes less efficient with the increase of

additive concentration.

14

-2

-1

0

1

2

3

4

5

6

7

8

0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10

ln (k

1)

1000/T (K-1)

k1 PredictionPS

PS+10%Br

PS+15%Br

PS+20%Br

-10

-5

0

5

10

15

20

0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10

ln (k

2)

1000/T (K-1)

k2 PredictionPS

PS+10%Br

PS+15%Br

PS+20%Br

Reaction Rates _ Phosphorous

Phosphorous containing materials’ reaction rates are comparable to that of the pure PS.

TPPO has a much faster reaction rate on the 1st reaction at higher temperatures.

15

-10

-5

0

5

10

15

20

0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10

ln (k

1)

1000/T (K-1)

k1 PredictionPS

TPPO

DOPO

-10

-5

0

5

10

15

20

0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10

ln (k

2)

1000/T (K-1)

k2 PredictionPS

TPPO

DOPO

Diffusion and Reaction Rates

Damkohler number (diffusion time scale / chemical reaction time scale) is used to evaluate the condition in fire.

Diffusion time scale is evaluated with:

For bimolecular reactions where [O2]0>>[Fuel]0, chemical reaction time scale is evaluated with:

16

𝜏𝜏𝑐𝑐𝑐𝑐𝑐𝑐𝑐 =1

𝑜𝑜2 𝑘𝑘𝑐𝑐𝑐𝑐𝑐𝑐𝑐

𝜏𝜏𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 =𝐿𝐿2

𝐷𝐷

Damkohler Number for Materials

17

1E+00

1E+02

1E+04

1E+06

1E+08

1E+10

1E+12

1500 1600 1700 1800 1900 2000

Da N

umbe

r (-)

T (K)

PS

PS+10%Br

PS+15%Br

PS+20%Br

TPPO

DOPO

At typical fire temperatures, the 2nd reaction is much faster than the 1st reaction. The total chemical reaction is limited by the 1st reaction (CO generation).

Da number estimated for the 1st reaction for all materials. o Da number is always higher than

unity. o For brominated materials, Da number

is close to 1. o For Phosphorous containing materials,

Da number is significantly higher than 1.

o It is noted the turbulent mixing in real fires should be faster than the diffusion.

Comparison with Cone Tests Cone test with bromine containing

material has a similar CO yield but much more smoke yield.

Cone test with phosphorous containing material has a significant increase in incomplete combustion products.

18

DOPO

DOPO

Smoke Yield (m2/kg) CO Yield (kg/kg)PS 1244 0.12

PS+10%Br 1751 0.14PS+15%Br 1940 0.14PS+20%Br 1999 0.15

TPPO > 2369 0.27DOPO > 2416 0.37

Summary A MCC was used to study gas phase combustion of phosphorus and

bromine containing polymeric materials. 2-step reaction kinetics were developed. And appropriate values for

A and E were obtained by optimization. Presence of bromine suppresses both reactions with larger effect

on the 2nd reaction. Presence of phosphorous has a higher temperature dependence. The Damkohler number evaluated at typical flame temperature is

higher than unity, indicating a oxygen starving condition in fires. The cone calorimeter tests show a correlation between Damkohler

number and carbon monoxide yield. For materials in which chemical reaction rate is significantly higher than mixing rate, CO yield increases.

19

Future Work

Mix phosphorous additives (TPPO / DOPO) with other polymeric materials and study their kinetics. (degrading at the similar temperature)

Modify current method of MCC testing. Use high-temperature furnace and different oxygen concentrations to achieve flame alike condition.

The authors are grateful to the companies for providing samples used in the current experiments.

20

Acknowledgement