Fire Resistive Materials: MICROSTRUCTURE Performance Assessment and Optimization of Fire Resistive Materials NIST July 14, 2005.

Fire Resistive Materials: MICROSTRUCTURE

Performance Assessment and Optimization of Fire Resistive Materials

NIST

July 14, 2005

MicrostructureExperimental

3-D Tomography2-D optical, SEM

Confocal microscopy

Modeling3-D Reconstruction

ParametersPorosity

Pore SizesContact Areas

Properties(all as a function of T)

ThermalHeat CapacityConductivity

DensityHeats of Reaction

AdhesionPull-off strength

Peel strengthAdhesion energy

Fracture toughness

EquipmentTGA/DSC/STASlug calorimeter

DilatometerBlister apparatus

Materials Science-Based Studies of Fire Resistive Materials

EnvironmentalInterior

Temperature, RH, load

ExteriorTemperature, RH, UV, load

Performance PredictionLab scale testingASTM E119 Test

Real structures (WTC)

Importance of Microstructure• As with most materials, microstructure of FRMs will significantly

influence many performance properties– Adhesion and mechanical properties (fracture toughness)– Heat transfer

• Conduction• Convection (of steam and hot gases)• Radiation

• What are some critical microstructural parameters for porous FRMs?– Porosity (density, effective thermal conductivity)– Pore size (radiation transfer)– Pore connectivity (convection, radiation)

• What are some applicable techniques for characterizing the microstructure of FRMs?– Optical microscopy– Scanning electron microscopy (SEM)– X-ray microtomography

Optical Microscopy

• Can apply to cast or fracture surfaces

• Minimal specimen preparation required

• Can estimate total coarse porosity and “maximum” pore size

• Two-dimensional so limited information on porosity connectivity

Scanning Electron Microscopy

• Higher resolution view than optical microscopy

• Tradeoffs between magnification and having a representative field of view

• More specimen preparation may be required

• Two-dimensional

X-ray Microtomography• Inherently three-

dimensional• Intensity of signal

based on x-ray transmission (local density) of material

• Voxels dimensions of 10 μm readily available– 1 μm at specialized

facilities (e.g., ESRF in France)

BFRL Experience with Microtomography

• With CSTB and ESRF (France), in 2001, created the visible cement dataset, a first-of-its-kind view of the 3-D microstructure of hydrating cement paste and plaster of Paris– http://visiblecement.nist.gov

• With FHWA, Penn State, and others, have imaged thousands of three-dimensional coarse and fine aggregates as part of the ongoing VCCTL consortium

• In 2004, with Penn State, imaged a variety of fire resistive materials including fiber-based, gypsum-based, and intumescent materials

X-ray Microtomography of Unexposed Gypsum-Based FRM

From Center for Quantitative Imaging, Penn State Univ.

X-ray Microtomography of Flame-Exposed Intumescent Coating FRM

From Center for Quantitative Imaging, Penn State Univ.

Microstructure Thermal Conductivity

• Segment 3-D microstructures into pores and solids (binary image)

• Extract a 200x200x200 voxel subvolume from each microstructure data set

• Separate and quantify volume of each “pore” (erosion/dilation, watershed segmentation-Russ, 1988, Acta Stereologica)

• Input segmented subvolume into finite difference program to compute thermal conductivity (compare to measured values)

Three-Dimensional X-ray Microtomography

• Three-dimensional images of isolated pores

Gypsum-based Fiber/cement-based

Thermal Conductivity Computation

-Use finite difference technique with conjugate gradient solver (Garboczi, 1998, NISTIR)-Put a temperature gradient across the sample and solve for heat flow at each node-Compute equivalent k value for composite material

Q = -kA (dT/dx)

Porosity: kpore

“Solid”: ksolid

Q-Need to know values for kpore and ksolid (itself microporous)

Thermal Conductivity of Porous Solids

ppppv

pvpkk solidPS

3/23/2

3/23/2

1)(

1

where

• v = kpore/ksolid

• ksolid = thermal conductivity of solid material,

• p = porosity = (max –matl)/max

max = density of solid material in the porous system,

matl = density of the porous material, and

• kpore = thermal conductivity of pore = kgas-cond + krad

Theory of Russell (1935, J Amer Ceram Soc)

Radiation Term

3

3

16ETrkrad

where

• σ = Stefan-Boltzmann constant

(5.669x10-8 W/m2/K4),

• E = emissivity of solid (1.0 for a black body),

• T = absolute temperature (K), and

• r = radius of pore (m)

kpore= krad + kgas-cond

For spherical pores (Loeb, 1954, J Amer Ceram Soc):

Applicability of Russell/Loeb Theory

FRM A and B are both fiber/portland cement-basedPore sizes estimated as “maximum radius” from optical microscopy

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 200 400 600 800 1000 1200

Temperature (oC)

k [

W(/

m•K

)]

FRM-A 0.5 mm pores

FRM-A (measured)

FRM-B (measured)

FRM-B 0.75 mm pores

Microstructure Modeling Results: Gypsum-Based Material FRM C

Complication- gypsum to anhydrite conversionkgypsum ≈ 1.2 W/m•K kanhydrite≈ 4.8 W/m•K (Horai, 1971)

0.05

0.10

0.15

0.20

0.25

0.30

0 200 400 600 800 1000 1200

Temperature ( oC)

k [W

/(m

•K)]

Measured data

Model (gypsum/anhydrite)

Model (constant k-gypsum)

Microstructure Modeling Results: Fiber/Cement-Based Material FRM B

Complications– Anisotropy of microstructure– Radiation transfer through connected pores (Flynn and Gorthala, 1997)

0.000.050.100.150.200.250.300.350.400.450.50

0 200 400 600 800 1000 1200

Temperature ( oC)

k (

W/m

•K)

Measured data

No erosions (x,y)

No erosions (z)

k vs. Porosity and Pore Size

FRM ρ(kg/m3)

Porosity Pore

radius (mm)

k

(23 oC)[W/(m•K)]

k (1000 oC)

[W/(m•K)]

A-fiber 313.7 87.5 % 0.5 0.0534 0.3708

B-fiber 236.8 91.2 % 0.75 0.0460 0.5010

C-gypsum 292.4 87.2 % 0.2 0.0954 0.2618

Summary• Microstructure is paramount to thermal

performance of FRMs• Powerful microstructure characterization

techniques exist and are becoming more commonplace

• Computational techniques are readily available for predicting thermal conductivity from microstructure-based inputs

• Opens possibilities for microstructure-based design and optimization of new and existing FRMs