polymers
Article
Factors for Consideration in an Open-Flame Test
for Assessing Fire Blocking Performance of
Barrier Fabrics
Shonali Nazaré *, William M. Pitts, John Shields and Rick Davis
Fire Research Division, Engineering Laboratory, National Institute of Standards and Technology,
Gaithersburg, MD 20899-8665, USA; [email protected] (W.M.P.); [email protected] (J.S.);
[email protected] (R.D.)
* Correspondence: [email protected]; Tel.: +1-301-975-2499
Academic Editors: Baljinder Kandola, Abderrahim Boudenne and Paul Kiekens
Received: 14 July 2016; Accepted: 6 September 2016; Published: 19 September 2016
Abstract: The main objective of the work reported here is to assess factors that could affect the
outcome of a proposed open flame test for barrier fabrics (BF-open flame test). The BF-open
flame test characterizes barrier effectiveness by monitoring the ignition of a flexible polyurethane
foam (FPUF) layer placed in contact with the upper side of the barrier fabric, exposed to a burner
flame from below. Particular attention is given to the factors that influence the ignitibility of the
FPUF, including thermal resistance, permeability, and structural integrity of the barrier fabrics (BFs).
A number of barrier fabrics, displaying a wide range of the properties, are tested with the BF-open
flame test. Visual observations of the FPUF burning behavior and BF char patterns, in addition to
heat flux measurements on the unexposed side of the barrier fabrics, are used to assess the protective
performance of the BF specimen under the open flame test conditions. The temperature and heat
transfer measurements on the unexposed side of the BF and subsequent ranking of BFs for their
thermal protective performance suggest that the BF-open flame test does not differentiate barrier
fabrics based on their heat transfer properties. A similar conclusion is reached with regard to BF
permeability characterized at room temperature. However, the outcome of this BF-open flame test is
found to be heavily influenced by the structural integrity of thermally degraded BF. The BF-open
flame test, in its current form, only ignited FPUF when structural failure of the barrier was observed.
Keywords: residential upholstered furniture; barrier fabrics; flammability; thermal protective
performance; thermal degradation; flaming ignition; flexible polyurethane foam
1. Introduction
The flammability of residential upholstered furniture (RUF) has long been recognized as a major
contributor to residential fire losses in the United States and elsewhere [1] due to the rapid fire
growth and high heat release rates frequently observed. As an example, fire statistics suggest that
in the US fires involving RUF are responsible for 25% of all fire deaths in residences [2]. In 1975,
California implemented Technical Bulletin (TB) 117-1975, which required that materials, such as
polyurethane foam, used to fill furniture, be able to withstand a small open flame for at least 12 s [3].
Flame retardant (FR) chemicals were used widely in upholstered furniture to meet the FR standards of
the California Bureau of Electronic and Appliance Repair, Home Furnishings, and Thermal Insulation’s
(CBEARHFTI) Technical Bulletin (TB) 117 [4]. Due to the relatively large size of the California market
and its influence on the overall furniture market in the United States, the use of flame retardants in
RUF became common throughout the United States.
Concerns have been raised about the potential for these widely used FRs to have harmful human
health and environmental effects [5,6]. Additionally, the effectiveness of FRs at the levels used in RUF
Polymers 2016, 8, 342; doi:10.3390/polym8090342
www.mdpi.com/journal/polymers
Polymers 2016, 8, 342
2 of 18
for reducing RUF flammability was questioned [7]. The environmental concerns and the effectiveness
of FRs used in RUF were publicized in a series of investigative articles in the Chicago Tribune [8].
In 2012, the governor of California requested the California Bureau of Electronics and Appliance Repair,
Home Furnishing and Thermal Insulation (CBEARHFTI) to review the flaming ignition requirements
of TB 117 [9]. As a result of this review, the flaming ignition test requirements were replaced by
a test based on limiting cigarette (a smoldering source) ignition of upholstery-fabric covered flexible
polyurethane foam (FPUF) [10].
Recent fire loss data has suggested that a very large percentage of the fire losses associated
with RUF take place during flaming fires ignited by either smoldering or flaming sources [2,11].
Flaming fires initially ignited by a smoldering source occur following transition from a smoldering
fire. Clearly, cost-effective alternative approaches for reducing fire growth rates and heat release rate
levels in flaming fires involving RUF are desirable. The inclusion of a material (referred to as a barrier
fabric) between the exterior upholstery fabric and the cushioning designed to slow ignition and/or
limit the burning intensity has been suggested as a possible approach. Barrier fabrics are widely
used to help meet mandatory flammability standards for mattresses [12] and California standards for
reduced-flammability contract furniture [13].
Cal TB 117-2013 includes a provision that allows upholstery fabrics, which fail the smoldering
ignition test, to be utilized when an appropriate barrier fabric, i.e., a material interposed between the
outer upholstery fabric and the interior cushioning materials, is used to protect the interior cushioning
materials from smoldering ignition. Recently, CBEARHFTI proposed an open-flame test [14] for
barrier fabrics (BF-open flame test). The test method can provide insight into the response of barrier
materials to an open-flame ignition source and the ability of the barrier fabric to prevent or slow down
an external flame from reaching the FPUF and igniting it.
Recognizing the potential importance of barrier fabrics in RUF applications, researchers at the
National Institute of Standards and Technology (NIST), have reviewed existing barrier fabrics and
have developed potential methodologies for assessing the effectiveness of barrier fabrics in RUF
fires ignited by flaming and smoldering sources [15–17]. However, several of the test methods that
provide quantitative data use expensive equipment and require skilled personnel, both of which are
not always easily available to the test facilities in the industry. Protective performance testing of barrier
fabrics must be as simple and inexpensive as possible, while still providing science-based information.
The open flame test drafted by CBEARHFTI for barrier fabrics, proposes a possible standard testing
method with a goal of producing upholstered furniture which is safer from the hazards associated
with small open-flame ignition [14].
The BF-open flame test used in this study, which mirrors the CBEARHFTI proposed standard
testing methods, has some similarities to ASTM D7140, Standard Test Method to Measure Heat Transfer
through Textile Thermal Barrier Materials [18]. ASTM D7140 is designed to measure heat transfer
through textile materials and to differentiate barrier materials based on their heat transfer properties.
In this test, the barrier material is exposed to a well-defined and controlled convective heat source
(propane flame from a Meker burner) from underneath for 60 s. The temperature on the unexposed
side of the test specimen is monitored. This test method essentially produces a time-temperature curve;
however, it does not define a heat transfer threshold for textile materials that can be used as a criterion
for fire barrier materials.
The BF-open flame test method consists of a similar application of a pre-mixed butane flame to the
underside of a horizontally mounted specimen of the barrier fabric (see Figure 1). However, unlike in
ASTM D7410, a layer (12.7 mm thick) of non-FR FPUF, having specified properties, is placed on the
unexposed side of the barrier fabrics away from the flame. The test specimen is sandwiched over
an opening formed by two rigid fire-rated insulating boards supported by a metal rack. The barrier
fabric (BF) specimen fails the test if the foam on top of the BF specimen ignites. The differences and
similarities between ASTM D7140 and the BF-open flame test are given in Table 1. The amount of
material required and the exposure area of the test specimen are larger in the case of the BF-open
Polymers 2016, 8, 342
3 of 18
flame test. Both test methods use the same types of Meker burner, and the flame application times are
Polymers 2016, 8, 342
3 of 17
identical. The distance between the top of the burner and the test specimen is almost double for the
BF-open
flame
test than
in the
D7140
test method.
The gas
rate is
specified
in theinBF-open
the BF-open
flame
test than
inASTM
the ASTM
D7140
test method.
Theflow
gas flow
rate
is specified
the BFflame
test,
whereas
in
the
ASTM
D7140
test
method
the
gas
flow
rate
is
adjusted
to
obtain
a heat
flux
open flame test,
whereas
in
the
ASTM
D7140
test
method
the
gas
flow
rate
is
adjusted
to
obtain
a heat
2 from the flame exposure. ASTM D7140 requires the use of a propane flame, while the
of
46
kW/m
flux of 46 kW/m² from the flame exposure. ASTM D7140 requires the use of a propane flame, while
BF-open
flame
test test
usesuses
a butane
flame.
the BF-open
flame
a butane
flame.
Figure
Figure 1.
1. BF-open
BF-open flame
flame test
test apparatus.
apparatus.
Table 1.
BF-openflame
flametests.
tests.
Table
1. Comparison
Comparison of
of ASTM
ASTM 7140
7140 and
and BF-open
Description
Description
TestTest
specimen
specimen
Exposed
specimen
area
Exposed specimen
area
Burner
Burner
Gas
Gas
Gas flow rate
Gas flow
rate and burner
Distance between
specimen
Flame application
time burner
Distance between
specimen and
Heat
flux
of
flame
exposure
Flame application time
Heat flux of flame exposure
ASTM7140
7140
BF-open
flame
ASTM
BF-open
flame
test test
133
mm
×
133
mm
250
mm
×
250
133 mm × 133 mm
250 mm × 250 mm mm
76
mm
×
76
mm
× 127
76 mm × 76 mm
127127
mmmm
× 127
mm mm
Meker burner
burner 38
burner
38 mm
diameter,
Meker
38mm
mmdiameter,
diameter, Meker
Meker
burner
38 mm
diameter,
1.2
size
1.2 1.2
mmmm
orifice
size size
1.2mm
mmorifice
orifice
size
orifice
Natural gas/propane
Butane
Natural
gas/propane
Butane
Not specified
500 ± 10 mL/min
Not
500
10 mL/min
50
± specified
1.6 mm
100±mm
s mm
60100
s mm
50 ±601.6
2
Not specified
46 kW/m
60 s
60 s
2
Not specified
46 kW/m
ASTM
D7140 measures
measuresthe
theheat
heatpenetration
penetrationthrough
through
a BF
when
exposed
toopen
an open
flame.
ASTM D7140
a BF
when
exposed
to an
flame.
The
The
BF-open
flame
test,
on
the
other
hand,
assesses
barrier
effectiveness
by
monitoring
the
ignition
BF-open flame test, on the other hand, assesses barrier effectiveness by monitoring the ignition of a
of
a FPUF
placed
in contact
the unexposed
side
the barrier
Thus, ASTM
while ASTM
FPUF
layerlayer
placed
in contact
with with
the unexposed
side of
theof
barrier
fabric.fabric.
Thus, while
D7140
D7140
is
a
quantitative
test,
the
BF-open
flame
ignition
test
is
a
pass/fail
test.
Moreover,
at
first
glance
is a quantitative test, the BF-open flame ignition test is a pass/fail test. Moreover, at first glance the
the
BF properties
responsible
limiting
ignition
of FPUF
in BF-open
flame
tests
uncertain.
BF properties
responsible
for for
limiting
the the
ignition
of FPUF
in BF-open
flame
tests
areare
uncertain.
In
In
this
study,
we
examine
the
factors
that
influence
the
ignitability
of
FPUF
in
the
BF-open
flame
this study, we examine the factors that influence the ignitability of FPUF in the BF-open flame test
test
method.
BF-open
flame
method
was
evaluatedbybycomparing
comparingthe
the results
results of
of several
method.
The The
BF-open
flame
testtest
method
was
evaluated
several
commercially
available
barrier
fabrics
from
the
BF-open
flame
test.
Additionally,
measurements
commercially available barrier fabrics from the BF-open flame test. Additionally, measurements of
of
heat
transfer
through
the
BFs
were
made
using
a
technique
similar
to
that
described
in
ASTM
D7140.
heat transfer through the BFs were made using a technique similar to that described in ASTM D7140.
Observations
Observations of
of the
the FPUF
FPUF burning
burning behavior
behavior and
and char
char patterns
patterns in
in the
the BF-open
BF-open flame
flame test
test are
are included
included
in
the
assessment
of
the
performance
of
the
BF
specimen
under
the
BF-open-flame
test
conditions.
in the assessment of the performance of the BF specimen under the BF-open-flame test conditions.
2. Materials and Methods
Certain commercial equipment, instruments or materials are identified in this paper in order to
specify the experimental procedure adequately. Such identification is not intended to imply
recommendation or endorsement by the National Institute of Standards and Technology, nor is it
Polymers 2016, 8, 342
4 of 18
2. Materials and Methods
Certain commercial equipment, instruments or materials are identified in this paper in order
to specify the experimental procedure adequately. Such identification is not intended to imply
recommendation or endorsement by the National Institute of Standards and Technology, nor is
it intended to imply that the materials or equipment identified are necessarily the best available for
this purpose.
2.1. Materials
The sample descriptions and physical properties of 17 commercially available BFs tested during
the current study are given in Table 2. For comparison purposes, the sample identification numbers
have been kept identical to those used in our previous studies [16,17]. New BFs included in this
study were assigned higher numbers. The exclusion of BFs studied earlier from the current study
was either due to unavailability of sufficient barrier fabric (BF-6, BF-7, and BF-14) or to the similarity
of fabric structures (BF-11, BF-12 both knitted and similar to BF-13 and BF-17, BF-18 both woven
glass fabrics and similar to BF-19). The list includes a variety of textile structures including high-loft,
nonwoven battings, knitted, woven structures, and back-coated fabrics. The BFs varied in average
thicknesses from 0.1 mm to 7.8 mm. The experimental matrix covers the most extensively used fibers
and fiber blends in the BF industry. These include flame retardant (FR) rayon, low-melt polyester,
FR polyester, glass fiber, aramid fibers, and blends thereof. BFs made from the latest core-yarn
technology and high-performing polyaramid/melamine fiber blends are also included. The exact fiber
blend compositions are proprietary and thus were not available.
Depending on the mode of fire blocking technology employed, the BFs in Table 2 are identified
as passive or active. By providing a physical barrier between the heat source and the cushioning
material, passive barriers can limit pyrolysis and heat release rates. Their effectiveness derives from
serving as a physical and/or thermal barrier between some or all of the fuel and the potential ignition
source. Active BFs have a chemical effect on the fire. Active flame retardants may act in the condensed
(or solid) phase or in the gas phase or as a combination of both. In the condensed phase, the FR
action typically results in enhanced char formation whereas in the gas phase the FR undergoes thermal
breakdown to form compounds that can lead to flame quenching.
BFs can also be distinguished as thermally thick or thermally thin materials. These terms come
from heat transfer analysis and refer to whether or not temperatures inside a material vary spatially,
when the surface is exposed to a heat source or sink. In thermally thin materials, heat transfer occurs
easily and quickly through the material. In these systems, the temperature on the outside of the object
is roughly the same as the temperature throughout the material; i.e., there are minimal temperature
gradients inside the material [19]. In contrast, temperatures inside thermally thick materials can vary
widely, when exposed to a heat source because of their relatively high level of heat transfer resistance.
In fabrics, thermal thickness is a function of heat transfer resistance relative to the physical thickness of
the fabric [20].
Considering the air permeability properties in Table 2, BFs can be characterised as permeable
and impermeable barriers. For the purpose of this study, impermeable barriers are defined as
barriers with air permeabilities (measured with a target pressure drop of 125 Pa [17]) less than 1 m/s.
These impermeable barriers limit the transport of volatile gases through the barrier.
Non-flame retarded FPUF pieces of known dimensions meeting Cal TB 117-2013 [10] were
procured from Innocor Foam Technologies, Coldwater, MS, USA.
Polymers 2016, 8, 342
5 of 18
Table 2. Description and physical properties of barrier fabrics.
Structure
FR system
Thickness
(mm)
Area
density §
(g/m2 )
Bulk
density
(g/cm3 )
Air
permeability
(m/s)
Highloft
Passive
4.1 ± 0.1
155
0.038
2.8 ± 0.2
Highloft
Passive
6.7 ± 0.2
230
0.034
2.0 ± 0.1
Needle punched
Passive
7.8 ± 0.6
240
0.031
2.3 ± 0.1
BF-4
Boric acid treated
cotton/FR
rayon/polyester
Needle
punched/Stratified
Passive
5.7 ± 0.1
230
0.040
2.2 ± 0.2
BF-5
Boric acid treated cotton
Needle punched
Passive
6.9 ± 0.8
230
0.033
1.3 ± 0.1
BF-8
FR rayon/polyester
Needle punched
nonwoven
Passive
4.3 ± 0.1
237
0.055
2.2 ± 0.1
BF-9
FR rayon/polyester
Needle punched
nonwoven
Passive
2.2 ± 0.1
240
0.109
1.5 ± 0.1
BF-10
FR polyester/FR rayon
Stitchbond
Active
0.7 ± 0.1
165
0.236
1.1 ± 0.1
BF-13
Glass fiber core/FR
acrylic fiber (core
spun yarn)
Knitted
Active
1.4 ± 0.1
165
0.118
1.9 ± 0.1
BF-15
Glass fiber core/FR
acrylic fiber
Woven
Active
0.5 ± 0.1
170
0.340
2.1 ± 0.1
BF-16
FR rayon/glass
fiber/Poly Lactic Acid
(PLA) fiber
Nonwoven
Active
2.9 ± 0.1
290
0.097
1.9 ± 0.2
BF-17
Glass filaments
Woven
Passive
0.2 ± 0.1
150
0.750
0
BF-20
Para-aramid/melamine
Woven
Passive
0.77 ± 0.02
264
0.343
0.2 ± 0.1
BF-21
Para-aramid
Nonwoven
Passive
0.67 ± 0.02
69
0.103
2.1 ± 0.1
BF-22
Meta-aramid/
Para-aramid
Woven/non-woven
composite
Passive
1.61 ± 0.11
267
0.166
0.9 ± 0.1
BF-23
Cotton/glass fiber
Knit/backcoated
Active/Passive
1.5 ± 0.1
284
0.189
0
BF-24
Polyester
Nonwoven batting
Passive
8.13 ± 1.1
165
0.020
7.6 ± 0.2
ID
Fiber blend
BF-1
Flame retarded (FR)
rayon/polyester
BF-2
BF-3
§ For textile materials, area density is generally expressed as mass per unit area. The standard uncertainty
(Type B) [21,22] in measuring area density is about ±5 g/m2 . A Type B evaluation of standard uncertainty is
based on scientific judgment using experience with, and general knowledge of, the behavior and property of
relevant materials and instruments.
2.2. BF-Open Flame Test
A schematic of the BF-open flame test apparatus used in this study is shown in Figure 1. A BF
specimen (approximately 250 mm × 250 mm) was sandwiched between two rigid fire-rated insulating
boards, each equipped with a square (127 mm × 127 mm) opening, supported by four supporting
threaded rods. A piece (127 mm × 127 mm × 12.7 mm) of FPUF was placed in the opening directly
above and in contact with the BF test specimen. A Meker burner (Humbolt Manufacturing Co.
Part number H-5600, Elgin, IL, USA) with a top diameter of 32 mm and orifice size of 1.2 mm was
used. A butane flame was used with a fuel flow rate of 500 ± 10 mL/min at ambient conditions,
while keeping the venturi throat in a fully open position [23]. The flame was applied to the test
specimen from underneath. The Meker burner was manually placed at the center of the bottom plate
of the test rig such that the top of the burner was positioned 102 mm below the center of the bottom
surface of the test specimen (Figure 1). The flame was applied for a period of 60 s. A barrier fabric failed
the test if flaming ignition of the FPUF was observed in any of three repeated runs for a given barrier.
2.3. Temperature and Heat Transfer Measurements
The experimental setup for measurement of heat transfer through the barrier fabrics was the
same as shown in Figure 1 except that the piece of FPUF was replaced by a slug calorimeter (Model:
ST-8-21915 S/N 169222, Medtherm Corporation, Huntsville, AL, USA) embedded in an insulating
board and placed face down on the BF specimen. This is similar, but not exactly the same as the set-up
Polymers 2016, 8, 342
Polymers 2016, 8, 342
6 of 18
6 of 17
defined
by ASTM
D 169222,
7140. The
slug calorimeter
consisted
of a blackened
discin40anmm
in diameter
ST-8-21915
S/N
Medtherm
Corporation,
Huntsville,
AL, USA) copper
embedded
insulating
and placed
down
on the
BF specimen.
is similar, but notthermocouples
exactly the samewere
as themounted
setwith board
a thickness
of 1.6face
mm.
Three
32-gauge
Type This
K chromel/alumel
◦
up
defined
by
ASTM
D
7140.
The
slug
calorimeter
consisted
of
a
blackened
copper
disc
40
mm
in to
along the perimeter of the disk at 120 locations. The calibrated slug calorimeter was connected
diameter
with
a
thickness
of
1.6
mm.
Three
32-gauge
Type
K
chromel/alumel
thermocouples
were
a data acquisition system which recorded the rise in temperature of the sensors as a function of time.
mounted along the perimeter of the disk at 120° locations. The calibrated slug calorimeter was
The rates of temperature rise were used in conjunction with the calorimeter constants provided by
connected to a data acquisition system which recorded the rise in temperature of the sensors as a
the manufacturer to compute the heat flux received. At the start of the test, the heat sensors were
function of time. The rates of temperature rise were used in conjunction with the calorimeter
approximately
at room by
temperature.
A particularly
useful
feature
of this test
is atest,
continuous
constants provided
the manufacturer
to compute
the heat
flux received.
At procedure
the start of the
the
calorimetric
tracewere
useful
for analyzingatthe
fabric
heat transfer
characteristics.
heat sensors
approximately
room
temperature.
A particularly
useful feature of this test
The
heat flux
the flame calorimetric
exposure was
calibrated
the slug
procedure
is from
a continuous
trace
useful by
forplacing
analyzing
the calorimeter
fabric heat facing
transferdown
so that
it was exposed directly to the flame. The distance between the bottom of the calorimeter and the
characteristics.
The
heat
fluxatfrom
the flame exposure
was
the rate
slug of
calorimeter
facing the
burner top
was
kept
an approximate
distance
ofcalibrated
102 mm. by
Theplacing
gas flow
butane through
down
so
that
it
was
exposed
directly
to
the
flame.
The
distance
between
the
bottom
of
the
calorimeter
flow meter was maintained at 500 ± 10 mL/min. The heat flux from the butane flame was measured
burner
top was
kept at an approximate
102 mm.
The gas flow
ratepropane
of butane
2 . This
2 ) obtained
as 50and
± 3the
kW/m
is comparable
to the heatdistance
flux (46ofkW/m
using
gas as
through the flow meter was maintained at 500 ± 10 mL/min. The heat flux from the butane flame was
specified in ASTM D 7140 test method.
measured as 50 ± 3 kW/m². This is comparable to the heat flux (46 kW/m²) obtained using propane
gas as specified
in ASTM D 7140 test method.
3. Results
and Discussion
3. Results and Discussion
3.1. General
Experimental Observations during BF-Open Flame Test
3.1.mentioned
General Experimental
Observations section,
during BF-Open
Test
As
in the experimental
duringFlame
the BF-open
flame test the open flame impinges
directly onAsthe
BF at the in
center.
Depending on
the type,
the barrier
fabricflame
ignitestest
or the
undergoes
charring.
mentioned
the experimental
section,
during
the BF-open
open flame
Heat impinges
transferred
through
the BF
BF at
heats
foam
and, if high
enough,
cause the
foam
to undergo
directly
on the
the the
center.
Depending
on the
type, can
the barrier
fabric
ignites
or
thermal
decomposition.
undergoes charring. Heat transferred through the BF heats the foam and, if high enough, can cause
thethe
foam
to undergo thermal
decomposition.
As
temperature
increases,
the foam undergoes both physical and chemical changes. Physically,
Asobserved
the temperature
increases,
theaway
foamfrom
undergoes
physical
and in
chemical
changes.
the foam is
to deform
and move
the heatboth
source,
resulting
an increased
surface
Physically,
the
foam
is
observed
to
deform
and
move
away
from
the
heat
source,
resulting
in
an
area. The FPUF forms a dome-like structure such as the one visible in Figure 2. The center of the FPUF
increased surface area. The FPUF forms a dome-like structure such as the one visible in Figure 2. The
dome is generally positioned over the tip of the applied flame. Such physical changes in the FPUF may
center of the FPUF dome is generally positioned over the tip of the applied flame. Such physical
change the heat transfer between the BF and the FPUF and may have a significant effect on subsequent
changes in the FPUF may change the heat transfer between the BF and the FPUF and may have a
decomposition
processes
of the foam.
significant effect
on subsequent
decomposition processes of the foam.
Figure 2. Digital images showing a thermally thick barrier fabric (BF-3) exposure to the open-flame
Figure 2. Digital images showing a thermally thick barrier fabric (BF-3) exposure to the open-flame
ignition source, dome formation, and no ignition of foam.
ignition source, dome formation, and no ignition of foam.
When heated, the FPUF can initially undergo two main reaction processes. One pathway is a
pyrolysis
involving
reverse
the polymerization
process,
to processes.
the formation
of toluene
When heated,
the the
FPUF
can of
initially
undergo two
main leading
reaction
One
pathway is
diisocyanate
(TDI)-derived
gaseous
products
and
polyol-derived
liquids
[24].
The
second
is toluene
an
a pyrolysis involving the reverse of the polymerization process, leading to the formation of
oxidation process that produces a char. The composition of degradation products depends very much
diisocyanate (TDI)-derived gaseous products and polyol-derived liquids [24]. The second is
on the conditions of pyrolysis [25]. During the first stage of pyrolysis, the heated foam releases
an oxidation process that produces a char. The composition of degradation products depends very
gaseous products (usually seen as a yellow smoke) from the isocyanate component of the foam
muchformulation
on the conditions
of pyrolysis [25]. During the first stage of pyrolysis, the heated foam
[26] and is probably the most volatile fraction of the released species [24]. However, this
releases
gaseous
products
(usually ignite
seen as
a yellow
frommost
theprobably
isocyanate
gaseous fuel does not necessarily
in the
BF-opensmoke)
flame test;
due component
to lack of of
the foam formulation [26] and is probably the most volatile fraction of the released species [24].
However, this gaseous fuel does not necessarily ignite in the BF-open flame test; most probably
due to lack of sufficient heating and/or an ignition source. Careful inspection of residual materials
after the test, especially where ignition of FPUF did not occur, revealed that a thin layer of liquid,
Polymers 2016, 8, 342
7 of 18
Polymers 2016, 8, 342
7 of 17
essentially regenerated polyol, was formed on the surface of FPUF that was in contact with the
especially
where
ignitionobservations
of FPUF did of
notFPUF
occur,pyrolysis
revealed and/or
that a thin
layer ofbehavior,
liquid, essentially
BF. Based
on our
previous
burning
formation of
regenerated
polyol,
was
formed
on
the
surface
of
FPUF
that
was
in
contact
with
the
BF.
Based
on our
a thin layer of liquid, believed to be regenerated polyol, is consistent with current observations.
previous
observations
of
FPUF
pyrolysis
and/or
burning
behavior,
formation
of
a
thin
layer
of
liquid,
Further heating led to pyrolysis of this materials.
believed to be regenerated polyol, is consistent with current observations. Further heating led to
In the presence of thermally thick barriers, where the rate of FPUF heating is much slower
pyrolysis
of this
materials.
Polymers 2016,
8, 342
7 of 17
compared to that for thermally thin barriers, the FPUF decomposition is more likely to be dominated
In the presence of thermally thick barriers, where the rate of FPUF heating is much slower
by the
secondtopathway,
whereby
oxidation
reactions
produce
significant
char.
Highly
porous,
especially
where
of FPUF
did not the
occur,
revealed
that a thin
layer of
liquid,
compared
that
forignition
thermally
thin barriers,
FPUF
decomposition
is more
likely
to beessentially
dominated
was formed
onprovide
the
surface
of FPUF that
was in contact
BF.losses
Based and
on
our
thermally
insulating
highloft
barriers
appropriate
conditions,
i.e.,with
lowthe
heat
sufficient
by regenerated
the
secondpolyol,
pathway,
whereby
oxidation
reactions
produce
significant
char.
Highly
porous,
previous
observations
of FPUF
pyrolysis
and/or
burning
behavior,
formationi.e.,
of asmoldering
thin layer
liquid,
oxygen
supplyinsulating
to allow highloft
the
FPUF
undergo
significant
low-temperature
combustion.
thermally
barriers
provide
appropriate
conditions,
low
heat of
losses
and
believed to be regenerated polyol, is consistent with current observations. Further heating led to
An example
of
the
char
formed
during
this
type
of
non-flaming
mode
of
combustion
is
shown
sufficient oxygen supply to allow the FPUF undergo significant low-temperature smoldering
pyrolysis of this materials.
combustion.
An example of the char formed during this type of non-flaming mode of combustion is
in Figure
3.
In the presence of thermally thick barriers, where the rate of FPUF heating is much slower
compared to that for thermally thin barriers, the FPUF decomposition is more likely to be dominated
by the second pathway, whereby oxidation reactions produce significant char. Highly porous,
thermally insulating highloft barriers provide appropriate conditions, i.e., low heat losses and
sufficient oxygen supply to allow the FPUF undergo significant low-temperature smoldering
combustion. An example of the char formed during this type of non-flaming mode of combustion is
shown in Figure 3.
shown in Figure 3.
Figure
3. Digital
imageofofchar
char removed
removed from
tested
with
BF-2.
Figure
3. Digital
image
fromthe
theFPUF
FPUF
tested
with
BF-2.
As heating from the bottom continued, holes were seen to be formed in the dome structure due
As
heating
from the
bottom
continued,
weretoseen
to be formed
in the dome
structure
due to
to foam
pyrolysis.
This
caused
the domeholes
structure
collapse.
This phenomenon
was
generally
foamobserved
pyrolysis.
the dome
structure
collapse.
generally
observed
in This
casescaused
of thermally
thin BFs
(BF-8, to
BF-9,
BF-10, This
BF-13,phenomenon
BF-15, BF-19, was
BF-20,
BF-21, BF-22
Figure
3. for
Digital
image
of substantial
char removed
from
the FPUF
tested
with
BF-2.
in cases
thermally
thin
BFs
(BF-8,
BF-9,
BF-10,
BF-13,
BF-15,
BF-20,
BF-21, to
BF-22
andin
BF-23)
andof
BF-23)
suggesting
that
these
tests
amounts
of BF-19,
heat
were
transferred
the foam
a
very
short
period.
Gaseous
products
were
visually
observed
being
released
immediately
following
suggesting that for these tests substantial amounts of heat were transferred to the foam in a very short
As heating from the bottom continued, holes were seen to be formed in the dome structure due
the hole
formation.
The volatilized
gases came
into contact with
fresh air
and althoughfollowing
the pyrolyzed
period.
products
visually
observed
released
toGaseous
foam pyrolysis.
Thiswere
caused
the dome
structurebeing
to collapse.
Thisimmediately
phenomenon was
generallythe hole
foam
and
BF
surface
may
have
been
at
a
high
temperature,
flaming
ignition
was
not
observed.
The
formation.
The
volatilized
gases
came
into
contact
with
fresh
air
and
although
the
pyrolyzed
observed in cases of thermally thin BFs (BF-8, BF-9, BF-10, BF-13, BF-15, BF-19, BF-20, BF-21, BF-22 foam
thermal
degradation
of
FPUF
and
subsequent
ignition
requires
a
mixture
of
combustible
gases
and
and BF-23)
suggesting
that at
forathese
substantial amounts
heat were
transferred
to the foam
and BF surface
may
have been
hightests
temperature,
flamingofignition
was
not observed.
Theinthermal
air athat
within
the flammability
limits
and
also a observed
sufficientbeing
heat source for
the mixture
to initiate
veryare
short
period.
products
were
visually
immediately
degradation
of
FPUF
andGaseous
subsequent
ignition
requires
a mixture ofreleased
combustible
gasesfollowing
and air that are
exothermic
reactions The
andvolatilized
ignite [27,28].
In this
open
configuration,
and
escape of
the hole formation.
gases came
into
contact
with fresh airheat
and losses
although
thethe
pyrolyzed
within
the flammability
and also a sufficient
heat
source
for the mixture
to initiate exothermic
gaseous
products
to limits
the
surroundings
high
rates
which
be was
the not
explanation
foam and
BF surface
may
have been atoccur
a highattemperature,
flamingcould
ignition
observed.for
Thethe
reactions
and
[27,28].
In
thisand
open
configuration, heat losses and the escape of gaseous products
failure
of ignite
pyrolysis
gases
auto-ignite.
thermal
degradation
of to
FPUF
subsequent ignition requires a mixture of combustible gases and
to the surroundings
occur
at high
which
could
be the
the
failure
of from
pyrolysis
of flammability
the
foam rates
waslimits
only
observed
the explanation
BF char
cracked
open
and
flames
airFlaming
that are ignition
within
the
and
also
a when
sufficient
heat
source
forfor
the
mixture
to initiate
gasesthe
toexothermic
auto-ignite.
burner came
into direct
contact
withInmaterial
top of the barrier.
When
ignition
reactions
and ignite
[27,28].
this openonconfiguration,
heat losses
andflaming
the escape
of
gaseousthe
products
to
the surroundings
occurobserved
at
high rates
which
be of
the
explanation
forand
the in
occurred
barrierof
failed
the test.
Following
ignition
the
complete
foam
was
engulfed
Flaming
ignition
the
foam
was
only
when
thecould
BFpiece
char
cracked
open
flames
of pyrolysis
gases
auto-ignite.
Digital
images
ofdirect
BFtoexposure
towith
the open-flame
ignition
ignitionWhen
and burning
of ignition
BF,
from flames.
thefailure
burner
came
into
contact
material on
top ofsource,
the barrier.
flaming
Flaming ignition
the foam was
only observed
the BFcrack
char cracked
open and
from
FPUFthe
degradation
and of
subsequent
ignition
of ignition
FPUF when
following
formation
the flames
BF
areengulfed
shown in
occurred
barrier failed
the test. Following
the complete
piece ofinfoam
was
the
burner
came
into
direct
contact
with
material
on
top
of
the
barrier.
When
flaming
ignition
in Figure 4.
flames. occurred
Digital images
of failed
BF exposure
to the open-flame
burning
the barrier
the test. Following
ignition theignition
completesource,
piece ofignition
foam wasand
engulfed
in of BF,
FPUF degradation
subsequent
ignition
FPUF following
crack formation
the BF
flames. Digitaland
images
of BF exposure
to theofopen-flame
ignition source,
ignition and in
burning
of are
BF, shown
FPUF
in Figure
4. degradation and subsequent ignition of FPUF following crack formation in the BF are shown
in Figure 4.
Figure 4. Digital images showing stages of a thermally thin, highly permeable barrier (BF-1) exposure
to the open-flame ignition source, flaming of BF-1, FPUF degradation and dome formation, dome
collapse and release of gaseous products, and flaming ignition of FPUF following the opening of a
direct
pathway
through
BF. stages of a thermally thin, highly permeable barrier (BF-1) exposure
Figure
4. Digital
imagesthe
showing
Figure 4. Digital images showing stages of a thermally thin, highly permeable barrier (BF-1) exposure to
to the open-flame ignition source, flaming of BF-1, FPUF degradation and dome formation, dome
the open-flame
ignition source, flaming of BF-1, FPUF degradation and dome formation, dome collapse
collapse and release of gaseous products, and flaming ignition of FPUF following the opening of a
and release
gaseous
products,
directof
pathway
through
the BF. and flaming ignition of FPUF following the opening of a direct
pathway through the BF.
Polymers 2016, 8, 342
Polymers 2016, 8, 342
8 of 17
8 of 18
collapse and release of gaseous products, and flaming ignition of FPUF following the opening of a
direct pathway through the BF.
A variety of FPUF behaviors are observed depending on the barrier properties, e.g., thermal
thickness,
gas permeability,
flammability,
and structural
integrity.
Images
barrier fabrics
FPUF
A variety
of FPUF behaviors
are observed
depending
on the
barrierofproperties,
e.g., and
thermal
specimens
(12.7
thick) after
a 60 s exposure
to the open
flame ignition
arefabrics
shownand
in Figure
thickness,
gas mm
permeability,
flammability,
and structural
integrity.
Imagessource
of barrier
FPUF 5.
Only
three
BFs;
BF-1,
BF-10,
and
BF-24
failed
this
test.
For
the
BFs
that
failed
the
BF-open
flame
test,
specimens (12.7 mm thick) after a 60 s exposure to the open flame ignition source are shown in Figure
holes
or
crack
formation
in
the
BFs
can
be
clearly
seen.
The
images
for
BF-13,
BF-15,
and
BF-17
5. Only three BFs; BF-1, BF-10, and BF-24 failed this test. For the BFs that failed the BF-open flame in
Figure
clearly
show formation
that the FPUF
was
pyrolysed;
noBF-13,
flaming
ignition
FPUF
test, 5holes
or crack
in the
BFssignificantly
can be clearly
seen. Thehowever,
images for
BF-15,
and of
BF-17
was
duringshow
the test
60significantly
s. The test pyrolysed;
duration ofhowever,
60 s wasnonot
sufficient
to form
inobserved
Figure 5 clearly
that duration
the FPUF of
was
flaming
ignition
of
FPUF
was
observed
during
the
test
duration
of
60
s.
The
test
duration
of
60
s
was
not
sufficient
to
openings (holes/cracks) in the BFs and the use of 12.7 mm (1/2”) thick foam provided insufficient
form
openingsbeyond
(holes/cracks)
the BFs
and theThus,
use ofthe
12.7BF-open
mm (1/2″)
thicktest
foam
provided
insufficient
fuel
to pyrolyze
the 60in
s test
duration.
flame
method,
does
not create
fuel to pyrolyze
beyond
the 60high
s testrates
duration.
Thus,
the BF-open
flame
method,
flammable
conditions
despite
of FPUF
pyrolysis.
The
test test
results
also does
seemnot
to create
depend
flammable
conditions
despite
high
rates
of
FPUF
pyrolysis.
The
test
results
also
seem
to
depend
on
on whether or not FPUF is fully decomposed prior to barrier breakthrough. In order to provide
whether
or
not
FPUF
is
fully
decomposed
prior
to
barrier
breakthrough.
In
order
to
provide
additional insight into the test behavior, these BFs were retested using 25.4 mm (1”) thick foam pieces.
into
the test
these BFs
were
retested
usingapplication
25.4 mm (1″)
thick
foam pieces.
Theadditional
modifiedinsight
BF-open
flame
testbehavior,
was terminated
after
5 min
of flame
if no
ignition
of FPUF
The
modified
BF-open
flame
test
was
terminated
after
5
min
of
flame
application
if
no
ignition
of
was observed. Table 3 shows the results of the BF-open flame test with 12.7 mm FPUF and the modified
FPUF was observed. Table 3 shows the results of the BF-open flame test with 12.7 mm FPUF and the
BF-open flame test with 25.4 mm FPUF. An additional four BFs (BF-2, BF-13, BF-15, and BF-20) failed
modified BF-open flame test with 25.4 mm FPUF. An additional four BFs (BF-2, BF-13, BF-15, and BFthe modified open flame test. Again, the FPUF ignition occurred due to direct flame impingement on
20) failed the modified open flame test. Again, the FPUF ignition occurred due to direct flame
the FPUF through openings formed in the BF.
impingement on the FPUF through openings formed in the BF.
Figure
5. Digital
images
fabrics and
andFPUF
FPUFspecimens
specimens(12.7
(12.7
mm
thick)
Figure
5. Digital
imagesofofunexposed
unexposedsides
sides of
of barrier
barrier fabrics
mm
thick)
after
60 60
s ofs of
exposure
toto
open
after
exposure
openflame
flameignition
ignitionsource.
source.
The following section describes the effects of BF properties on thermal decomposition of FPUF in
the BF-open flame test.
Polymers 2016, 8, 342
9 of 18
Table 3. Response of FPUF in BF-open flame test configurations.
12.7 mm (1/2”) thick foam
ID
Structure
BF-1
BF-2
25.4 mm (1”) thick foam
Pass/Fail
Comments
Ignition of
FPUF
Time to
ignition (s)
Comments
Highloft
Fail
Foam ignites due to
hole/crack formation in
the BF
Yes
40
Foam ignites due to
hole/crack formation in
the BF
Highloft
Pass
Dome formation
Yes
120
Foam ignites due to
hole/crack formation in
the BF
BF-3
Highloft
Pass
Dome formation
No
-
Dome formation
BF-4
Needle punched
nonwoven
Pass
Dome formation
No
-
Dome formation
BF-5
Needle punched
nonwoven
Pass
Dome formation
No
-
Dome formation
BF-8
Needle punched
flat
Pass
Dome formation
No
-
Dome structure collapses
due to formation of hole
in the FPUF
BF-9
Needle punched
flat
Pass
Dome formation
No
-
Dome structure collapses
due to formation of hole
in the FPUF
BF-10
Stitchbond
Fail
Foam ignites due to
hole/crack formation in
the BF
Yes
130
Foam ignites due to
hole/crack formation in
the BF
BF-13
Knitted
Pass
Dome structure collapses
due to formation of hole
in the FPUF
Yes
186
Foam ignites due to
hole/crack formation in
the BF
BF-15
Woven
Pass
Dome structure collapses
due to formation of hole
in the FPUF
Yes
165
Foam ignites due to
hole/crack formation in
the BF
BF-16
Nonwoven
Pass
Dome structure collapses
due to formation of hole
in the FPUF
No
-
Dome structure collapses
due to formation of hole
in the FPUF
BF-17
Woven glass
Pass
No dome formation,
FPUF forms liquid
No
-
No dome formation,
FPUF forms liquid
BF-20
Woven
Pass
Dome structure collapses
due to formation of hole
in the FPUF
No
BF-21
Nonwoven
Pass
Dome structure collapses
due to formation of hole
in the FPUF
Yes
240
Dome structure collapses
due to formation of hole
in the FPUF
BF-22
Woven/nonwoven
composite fabric
Pass
Dome structure collapses
due to formation of hole
in the FPUF
No
-
Dome structure collapses
due to formation of hole
in the FPUF
BF-23
Knit/backcoated
Pass
No dome formation,
FPUF forms liquid
No
-
No dome formation,
FPUF forms liquid
BF-24
Polyester batting
Fail
Foam ignites due to
hole/crack formation in
the BF
Yes
4
Foam ignites due to
hole/crack formation in
the BF
Dome structure collapses
due to formation of hole
in the FPUF
3.1.1. Thermally Insulating Barriers
Highloft nonwoven BFs are characterized by high volumes of air that exceed the volume of fiber.
BF-1, BF-2, BF-3, BF-4, BF-5, and BF-24 are highloft barriers. BF-1, BF-2, BF-3, BF-4, and BF-5 are made
up of char-forming fiber blends. These BFs vary in area densities, thicknesses, and gas permeabilities.
All of these BFs passed the BF-open flame test except BF-1. All of the tested highloft barriers ignited
and burned briefly when exposed to the open flame ignition source. BF-1 burned and formed a very
fragile char with holes in it. The flames from the ignition source reached the flammable material above
the barrier, which caused FPUF ignition. BF-1 thus failed the BF-open flame test. The lower area
density and thickness in combination with higher permeability of BF-1 compared to the other highloft
barriers likely explain its poor performance.
Polymers 2016, 8, 342
10 of 18
In the case of BF-2, BF-3, BF-4, and BF-5, the thermally insulating carbonaceous char was formed
following the ignition and/or thermal degradation of component fibers. The carbonaceous thick char
enhances flame and heat resistance. Moreover, the char was found to be structurally intact with no
holes or cracks, thus preventing the burner flames from reaching the FPUF. The undamaged dome
structure of the FPUF surface, particularly for BF-3, BF-4, and BF-5 (see Figure 5), indicated that heat
transfer into the FPUF was limited by the thermally thick barrier. In an attempt to assess the robustness
of this carbonaceous insulative char, we ran additional tests; whereby FPUF thickness was increased
from 12.7 mm to 25.4 mm and flame was applied until ignition of FPUF was observed. The times to
ignition of FPUF are given in Table 3. It was noted that the FPUF ignited only when a hole or a crack
was formed in the barrier.
BF-24 is comprised of 100% polyester fibers and is generally referred to as polyester batting.
Traditionally, this material has been used in upholstered furniture to give superior appearance and
performance characteristics to the finished product. In residential mattresses, a thin layer of such
polyester batting is often used to enhance smoldering ignition resistance. A similar approach has been
suggested for upholstered furniture whereby a polyester batting could act as a barrier to a smoldering
ignition source [29–31]. However, this material fails the BF-open flame test as the thermoplastic
polyester melts as soon as it is exposed to this open-flame ignition source. As the polyester batting
melts and shrinks to form a large hole (Figure 5), the FPUF placed directly above is exposed to the
flaming ignition source. The FPUF ignites and burns completely, thus failing the test.
3.1.2. Impermeable or Barriers with Low Gas Permeability
Needle-punched barriers are nonwoven materials characterized by high area density and low
thickness, resulting in higher bulk densities (see Table 2). These barriers also have very low air
permeabilities. BF-8, BF-9, and BF-16 in Table 2 are characterized as needle-punched flat barriers.
These barriers protect the FPUF sufficiently well to pass the BF-open flame test. In fact, the char of the
thermally degraded barriers was structurally intact even after 5 min of exposure to the open flame
ignition source.
BF-17, BF-20, BF-22, and BF-23 are additional BFs with very low or zero air permeabilities.
Examples of images of residual FPUF char taken after completion of these test are shown in Figure 5.
Even though the FPUF shows signs of thermal degradation, there is no trace of ignition and these BFs
passed the BF-open flame test. These BFs remain structurally intact, i.e., no hole or crack formation,
and the flame does not come into contact with the FPUF or the volatiles.
It is interesting to note that the FPUF does not form a uniform dome-like structure in the presence
of thermally thin, gas impermeable, high thermal conductivity barriers, for example, BF-17. The FPUF
forms a transient dome-like structure, however, further heating and rapid heat transfer via conduction
and radiation, results in faster decomposition of FPUF and a subsequent collapse of the dome-like
structure. The images of residual char in Figure 5 suggest that the FPUF above BF-17 pyrolyzed to
a greater extent when compared to those on top of BF-20, BF-22, and BF-23. This can be associated
with the thickness of the barrier. BF-17 is almost half the thickness of BF-20 and therefore, a higher
heat transfer rate is to be expected. The heat flux measurements on the unexposed side of these BFs,
discussed in the sections below, showed that the total amount of heat transferred through BF-17 is
indeed higher than that through BF-20, BF-22, and BF-23. In the case of BF-23, which is a BF with
flame retardant back-coating, some of the heat from the flaming ignition source is dissipated in the
endothermic reaction of the FR back coating.
3.1.3. Thermally Thin, Permeable Barriers
BF-13, BF-15 and BF-21 are thermally thin, highly permeable barrier fabrics. BF-13 is a knitted
fabric, and BF-15 is a woven fabric, both made from core-spun yarn. BF-21 is a nonwoven fabric
comprised of fire resistant para-aramid fibers. These BFs pass the BF-open flame test, however,
BF-13 and B-15 failed the longer exposure open flame test. In the case of these thermally thin,
Polymers 2016, 8, 342
11 of 18
permeable BFs, the FPUF dome structure formed and collapsed as holes formed in the FPUF. At the
same time, rapid heat transfer through the thermally thin, permeable BFs led to rapid decomposition
of FPUF. Liquefied FPUF dripped away from the FPUF and formed a pool of low viscous flammable
Polymers 2016, 8, 342
11 of 17
liquid on top of the BF. In some of the tests, liquid FPUF was actually observed to seep downward
heat transfer
through
thermally
permeable
BFs
ledbarrier.
to rapidSince
decomposition
FPUF.
throughrapid
the barrier
material
andthe
burn
on thethin,
burner
side of
the
the FPUFofdid
not ignite
Liquefied
FPUF
dripped
away
from
the
FPUF
and
formed
a
pool
of
low
viscous
flammable
liquid
on
or burn above the barrier during the test period, the BF-13 and BF-15 (see Figure 6) met the BF-open
topcriterion
of the BF.even
In some
of the tests,
liquid
FPUF
was actually
observed
seep downward
through
flame test
though
flames
were
clearly
generated
from to
material
associated
withthe
the FPUF.
barrier material and burn on the burner side of the barrier. Since the FPUF did not ignite or burn
For BF-13 and BF-15, dripping of liquefied FPUF was beneficial in terms of ignition resistance above
above the barrier during the test period, the BF-13 and BF-15 (see Figure 6) met the BF-open flame
the barrier
fabric. BF-21,
a thermally
thin,
non-woven
withassociated
gas permeability
comparable
to
test criterion
even though
flames were
clearly
generatedmaterial
from material
with the FPUF.
For
BF-15 (2.1
±
0.1
m/s),
also
passed
the
BF-open
flame
test.
In
the
case
of
BF-21,
the
FPUF
undergoes
BF-13 and BF-15, dripping of liquefied FPUF was beneficial in terms of ignition resistance above the
barrier fabric.The
BF-21,
a thermally
thin,
non-woven
material
withliquid
gas permeability
comparable
to BF- as seen
little degradation.
FPUF
formed
smaller
amounts
of the
decomposition
product,
15 (2.1
0.1 m/s),This
also passed
the BF-open
In the case
of BF-21,
the FPUF undergoes
for BF-13
and± BF-15.
is remarkable
inflame
that test.
the initial
physical
properties
of these little
BFs (BF-13,
degradation. The FPUF formed smaller amounts of the liquid decomposition product, as seen for BFBF-15 and
BF-21) are comparable. A possible explanation is that heating of the FPUF can be different
13 and BF-15. This is remarkable in that the initial physical properties of these BFs (BF-13, BF-15 and
for BF-13
and BF-15 as the BF properties e.g., gas permeability, are likely to change due to thermal
BF-21) are comparable. A possible explanation is that heating of the FPUF can be different for BF-13
degradation
of the
constituent
fibers.
Thepermeability,
fire resistant
fibers
BF-21degradation
char in place and
and BF-15
as the
BF properties
e.g., gas
arepara-aramid
likely to change
due tointhermal
therefore
gas permeability
may
of the
the constituent
fibers. The
fire remain
resistant unaltered.
para-aramid fibers in BF-21 char in place and therefore the
gas permeability may remain unaltered.
Figure 6. Digital images showing (a) flaming of liquefied FPUF below BF-15 and (b) unburned,
Figure 6. Digital images showing (a) flaming of liquefied FPUF below BF-15 and (b) unburned,
thermally degraded FPUF.
thermally degraded FPUF.
It can be noted from Table 3 that ignition times for BF-15 in the modified BF-open flame test
werebe
in noted
excess of
60 s.Table
A knitted
fabric
with a glass-fiber
(BF-13)
and
nonwovenBF-open
para-aramid
It can
from
3 that
ignition
times forblend
BF-15
in the
modified
flame test
barrier (BF-21) both failed to protect the FPUF from heat and flames in the modified BF-open flame
were in excess of 60 s. A knitted fabric with a glass-fiber blend (BF-13) and nonwoven para-aramid
test. When exposed to an open flame for longer duration, holes are formed in the thin BFs, through
barrier (BF-21)
both failed to protect the FPUF from heat and flames in the modified BF-open flame test.
which the flames from the ignition source penetrated and ignited the FPUF.
When exposed
to an open
longer
holes
arefrom
formed
in the
through which
As discussed,
notflame
all BFsfor
succeed
in duration,
protecting the
FPUF
ignition.
Thethin
levelBFs,
of protection
the flames
fromon
the
ignition
source penetrated
FPUF. gas permeabilities of the BF,
depends
complex
relationships
between theand
heatignited
transferthe
properties,
structuralnot
integrity
of succeed
BF char. The
properties ofthe
barrier
fabrics
governedThe
by the
fiber
Asand
discussed,
all BFs
in protecting
FPUF
fromareignition.
level
oftype
protection
In order to understand
degradation
of the FPUF
and heat fluxes, to
dependsand
onconstruction.
complex relationships
between the
thethermal
heat transfer
properties,
gas permeabilities
of the BF,
which the FPUF is exposed, we measured temperatures and heat fluxes on the unexposed sides of
and structural integrity of BF char. The properties of barrier fabrics are governed by the fiber type and
barrier fabrics using the instrumentation described earlier. The following section describes the heat
construction.
order toofunderstand
thermal
degradation
of the FPUF and heat fluxes, to which
transferIn
properties
the BFs testedthe
in the
BF-open
flame test configuration.
the FPUF is exposed, we measured temperatures and heat fluxes on the unexposed sides of barrier
3.2. Heat
Flux
Measurements on the
Unexposedearlier.
Side of BFs
fabrics using
the
instrumentation
described
The following section describes the heat transfer
properties ofInthe
BFs
tested
in
the
BF-open
flame
test
configuration.
order to compare the results of the BF-open flame
test to the heat transfer properties of the
BFs, the heat flux on the unexposed side of the BF was measured using the technique described in
3.2. HeatSection
Flux Measurements
on the
Unexposed
2.3. The heat flux
recorded
duringSide
the of
60BFs
s exposure time and the total amount of heat
transferred (THT) through the BF at the end of the 60 s test duration are given in Table 4. Uncertainties
In order
to compare the results of the BF-open flame test to the heat transfer properties of the
in measurement of various heat flux related parameters are reported as Type A uncertainties [21,22]
BFs, theinheat
on the
unexposed
side ofuncertainty
the BF was
measuredasusing
the technique
described
Tableflux
4. A Type
A evaluation
of standard
is determined
the standard
deviation of
a
in Section
2.3.of The
heat flux
recorded Descriptions
during the of
60the
s exposure
time
the total
amount
series
independent
observations.
BF chars are
alsoand
included
in Table
4. It is of heat
evident
fromthrough
the char the
description
thatend
the of
BFsthe
that
in the
BF-openare
flame
test, in
failTable
as the4.result
of
transferred
(THT)
BF at the
60fail
s test
duration
given
Uncertainties
structural
failure
of
the
barriers.
in measurement of various heat flux related parameters are reported as Type A uncertainties [21,22]
Derived quantities such as the heat transfer factor (HTF) and thermal protective indices (TPI)
in Table 4. A Type A evaluation of standard uncertainty is determined as the standard deviation of
given in Table 4 and plotted in Figure 7, respectively, have been used to characterize the heat transfer
a series properties
of independent
observations.
of the BF THT
chars
arehaving
also included
inHTF
Table
of the barrier
fabrics. The Descriptions
HTF is a mass normalized
value
units of J/g.
is 4. It is
evident from the char description that the BFs that fail in the BF-open flame test, fail as the result of
structural failure of the barriers.
Polymers 2016, 8, 342
12 of 18
Derived quantities such as the heat transfer factor (HTF) and thermal protective indices (TPI)
given in Table 4 and plotted in Figure 7, respectively, have been used to characterize the heat transfer
properties of the barrier fabrics. The HTF is a mass normalized THT value having units of J/g.
2
HTF is the
ratio
of8,the
in g/m2 .
Polymers
2016,
342 total heat transferred in MJ/m (THT) value to the fabric area density
12 of 17
The inverse of the HTF was previously defined as the thermal protective index (TPI) for a BF and was
ratio
of for
the their
total heat
transferred
in MJ/m²
(THT) value to[16].
the fabric
area density
in g/m².
Thethermal
used to the
rank
BFs
thermal
protective
performances
The ranking
of BFs
using
inverse of the HTF was previously defined as the thermal protective index (TPI) for a BF and was
protective indices derived from heat transfer data obtained from the thermal protective performance
used to rank BFs for their thermal protective performances [16]. The ranking of BFs using thermal
(TPP) test
deviceindices
[16] and
the from
BF-open
flame data
test obtained
setup are
plotted
in Figure
7. The
ranking order
protective
derived
heat transfer
from
the thermal
protective
performance
is almost
the test
same
for [16]
bothand
sets
data. The
theare
TPI
value,
the better
the
thermal
protective
(TPP)
device
theof
BF-open
flamehigher
test setup
plotted
in Figure
7. The
ranking
order
is
almostof
the
same for both sets of data. The higher the TPI value, the better the thermal protective
performance
a BF.
performance of a BF.
10
9
Thermal Protective Performance (TPP) data from [16]
Data from BF-open flame test configuration
Thermal protective index (TPI)
8
7.2
7
6.3
5.9
6
5.1
5
5.1
4.7
4.6
4
3.7
3.0
2.7
3
2.2
2
1.5
1.8
1.6
1.7
0.8
1
0
7. Comparison
of BF
rankingsusing
using TPI
TPI from
different
test test
configurations.
FigureFigure
7. Comparison
of BF
rankings
fromtwo
two
different
configurations.
The experimental data suggest that different barriers have different heat transfer properties such
Thethat
experimental
dataofsuggest
that different
have
properties
the total amount
heat transferred
throughbarriers
the barriers
is different
“different”heat
and transfer
the maximum
heat such
that the flux
totalmeasured
amounton
of the
heat
transferred
the also
barriers
“different”
and
the maximum heat flux
unexposed
sidethrough
of the barrier
variesiswith
the type of
barrier.
measured on the unexposed side of the barrier also varies with the type of barrier.
Table 4. Heat transfer data for BFs exposed to BF-open flame test.
ID
Tableheat
4. Heat
data
forofBFs
Maximum
flux at transfer
Total
amount
heatexposed to BF-open flame test.
Heat transfer
unexposed side of barrier
transferred at 60 s
Visual observations and comments
factor (kJ/g)
at 60 s (kW/m²)
Maximum
heat flux at
Total (J/cm²)
amount of
ID
BF-1
unexposed17
side
± 2 of barrier
at 60 s (kW/m2 )
heat transferred
at
95 ± 12
60 s (J/cm2 )
BF-2
BF-1
BF-3
BF-4
BF-2
BF-5
BF-3
BF-8
BF-4
BF-5
BF-9
BF-8
BF-10
BF-9
BF-13
BF-10
BF-15 *
BF-13
BF-16
BF-15 *
BF-17
BF-16
BF-19
BF-17
BF-20
BF-19
BF-21
BF-20
BF-22 *
BF-21
BF-23
BF-22
*
8±1
17
2
6 ±±0.5
7
±
1
8±1
5 ±0.5
1
6±
713±±16
514±±12
13
15±± 61
14
29±± 21
15 18
±1
29
1
10 ±
± 0.3
18
17 ± 1
10 ± 0.3
16 ± 1
17 ± 1
13±± 12
16
15±± 23
13
9 3
15 ±
129± 1
43 ± 1
9535±±12
3
40±± 1
2
43
30±± 3
1
35
73±± 2
4
40
30
1
76 ±
± 10
73
88±± 4
76
± ±104
163
88104
±4
163
58 ±
± 14
104
98 ± 2
58 ± 1
93 ± 7
98 ± 2
73±
±16
93
7
87±
± 16
15
73
5215
87 ±
69
52± 6
21
61
15
18
21
13
15
31
18
13
32
31
53
32
99
53
61
99
20
61
65
20
29
65
28
29
126
28
19
126
24
19
BF-23
12 ± 1
33 ± 6
190 ± 33
115
BF-24
33 ± 6
190 ± 33
115
BF-24
69 ± 6
Heat transfer
factor61(kJ/g)
24
Thin,
fragile char with
Visual
observations
and hole
comments
formations
Thermally thick, insulating char
Thin, fragile char with
Thermallyhole
thick,
insulating char
formations
Thermally
char
Thermallythick,
thick,insulating
insulating
char
Thermally
char
Thermallythick,
thick,insulating
insulating
char
Undamaged
char
Thermally
thick, insulating
char
Thermally
thick, insulating
char
Undamaged
char
Undamaged
char
Thermally
thin, cracked
barrier
charbarrier
ThermallyUndamaged
thin, permeable
Thermally
thin,
crackedbarrier
barrier
Thermally
thin,
permeable
Thermally
thin, permeable
barrier
Undamaged
char
Thermally thin, permeable barrier
Undamaged char
Undamaged char
Undamaged char
Undamaged char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char
Undamaged
char slug
Barrier melts
and exposes
Barrier
melts and
exposes
calorimeter
to the
flame slug
calorimeter to the flame
* Single measurements were taken due to the unavailability of test specimens.
* Single measurements were taken due to the unavailability of test specimens.
Polymers 2016, 8, 342
13 of 18
Polymers 2016, 8, 342
13 of 17
The degree of thermal protection provided by the barrier fabrics can also be assessed by
The degree
of thermal
protection
provided
the barrier
fabrics
can also
be assessed
by
monitoring
the heat
flux versus
time data
collectedbyduring
the tests
run with
the slug
calorimeter
monitoring
the
heat
flux
versus
time
data
collected
during
the
tests
run
with
the
slug
calorimeter
(see
(see Figure 8). Thermal responses of (a) thermally thick, highloft; (b) thermally thin, permeable;
Figure
8). Thermal
responses ofand
(a)(d)
thermally
thick, highloft;
(b) thermally
(c)
thermally
thin, impermeable;
flat non-woven
BFs are shown
in Figurethin,
8. permeable; (c)
thermally
thin,
impermeable;
and
(d)
flat
non-woven
BFs
are
shown
in
Figure
8.
The heat flux versus time curve for direct slug calorimeter exposure (without barrier fabric) is
The heat flux versus time curve for direct slug calorimeter exposure (without barrier fabric) is
included for reference. It can be seen that high heat flux values of 45 ± 5 kW/m2 are reached almost
included
for reference. It can be seen that high heat flux values of 45 ± 5 kW/m² are reached almost
instantaneously in the absence of a barrier. Immediately after the peak is reached, the heat flux curve
instantaneously
in the absence of a barrier. Immediately after the peak is reached, the heat flux curve
starts to decline. The decrease in the heat flux measurement could be due to decreasing the difference
starts to decline. The decrease in the heat flux measurement could be due to decreasing the difference
between the heat source and the surface temperature of the slug and/or due to heat losses from the
between the heat source and the surface temperature of the slug and/or due to heat losses from the
slug to its holder [32]. Correcting the slug calorimeter results for such effects in order to achieve more
slug to its holder [32]. Correcting the slug calorimeter results for such effects in order to achieve more
accuracy is beyond the scope of this work. It is important to note here that the heat flux data is used to
accuracy is beyond the scope of this work. It is important to note here that the heat flux data is used
compare the heat transfer properties of different barriers. It is also worth mentioning that the heat flux
to compare the heat transfer properties of different barriers. It is also worth mentioning that the heat
versus time curves reach steady states after a peak is reached in the presence of a barrier, suggesting
flux versus time curves reach steady states after a peak is reached in the presence of a barrier,
no additional changes in barrier effectiveness with time. Generally, it can be seen that only a third of
suggesting no additional changes in barrier effectiveness with time. Generally, it can be seen that
the applied heat flux is transferred through a char forming barrier.
only a third of the applied heat flux is transferred through a char forming barrier.
There are no great differences in the BF behavior during heat flux measurements as compared
There are no great differences in the BF behavior during heat flux measurements as compared
to the BF-open flame test with FPUF. For example, BF-24 melted and shrank away from the flame,
to the BF-open flame test with FPUF. For example, BF-24 melted and shrank away from the flame,
BF-10 formed cracks, and BF-1 had holes in it as described in Table 4. However, none of these
BF-10 formed cracks, and BF-1 had holes in it as described in Table 4. However, none of these BFs
BFs show an instantaneous increase in heat flux on the formation of cracks or holes except BF-24,
show an instantaneous increase in heat flux on the formation of cracks or holes except BF-24, for
for which the heat flux increases significantly as soon as the polyester batting melts and forms a hole
which the heat flux increases significantly as soon as the polyester batting melts and forms a hole to
to expose the slug calorimeter to the flame. As discussed earlier, the FPUF ignites, and BF-24 fails
expose the slug calorimeter to the flame. As discussed earlier, the FPUF ignites, and BF-24 fails the
the test. Comparing heat flux versus time curves for the BFs that failed the test, it can be noted from
test. Comparing heat flux versus time curves for the BFs that failed the test, it can be noted from
Figure 8 that BFs can fail the test without reaching high heat flux values, i.e., the barriers fail due to
Figure 8 that BFs can fail the test without reaching high heat flux values, i.e., the barriers fail due to
structural failure. The FPUF ignites only when the flames from ignition source come into contact with
structural failure. The FPUF ignites only when the flames from ignition source come into contact with
volatiles. The transient response of the slug calorimeter to an instantaneously imposed high heat flux
volatiles. The transient response of the slug calorimeter to an instantaneously imposed high heat flux
is determined by a thermal energy balance of the slug [33] and solicits further studies to investigate
is determined by a thermal energy balance of the slug [33] and solicits further studies to investigate
instantaneous heat transfer due to structural failure of the BFs.
instantaneous heat transfer due to structural failure of the BFs.
50
(a)
45
No Barrier
40
BF-1
air permeability = 7.6 m/s
35
Heat flux, kW/m²
BF-2
30
BF-3
25
20
BF-4
air permeability =2.8 m/s
15
BF-5
10
air permeability = 2.0 m/s- 2.3 m/s
BF-24
5
air permeability =1.3 m/s
0
0
10
20
30
40
Time, s
Figure 8. Cont.
50
60
Polymers 2016, 8, 342
Polymers 2016, 8, 342
14 of 18
14 of 17
50
(b)
45
No Barrier
40
Heat flux, kW/m²
35
air permeability = 2.1
BF-10
30
25
BF-15
air permeability = 1.9 m/s
20
air permeability = 1.1 m/s
15
BF-13
10
air permeability = 0.2 m/s
5
BF-20
0
0
10
20
30
Time, s
40
50
50
60
(c)
45
No Barrier
40
Heat flux, kW/m²
35
30
BF-17
25
20
BF-22
15
air permeability = 0 m/s
10
BF-23
5
0
0
10
20
30
40
Time, s
Figure 8. Cont.
50
60
Polymers 2016, 8, 342
Polymers 2016, 8, 342
15 of 18
15 of 17
Figure 8. Thermal response of (a) thermally thick, highloft; (b) thermally thin, permeable; (c)
Figure 8. Thermal response of (a) thermally thick, highloft; (b) thermally thin, permeable; (c) thermally
thermally thin, impermeable; and (d) flat non-woven BFs.
thin, impermeable; and (d) flat non-woven BFs.
4. Conclusions
4. Conclusions
The BF-open flame test is an easy to operate, cost-effective test method designed to assess the
The BF-open flame test is an easy to operate, cost-effective test method designed to assess the
protective performance of fire blocking barrier fabrics when exposed to a flaming ignition source.
protective performance of fire blocking barrier fabrics when exposed to a flaming ignition source.
Several commercially available barrier fabrics were tested using the proposed test method. In the BFSeveral commercially available barrier fabrics were tested using the proposed test method. In the
open flame test, ignition of the FPUF is only observed when the BF breaks open and flames from the
BF-open flame test, ignition of the FPUF is only observed when the BF breaks open and flames from
ignition source come into direct contact with decomposing FPUF or its volatiles. Once flaming
the ignition source come into direct contact with decomposing FPUF or its volatiles. Once flaming
ignition occurs, the complete piece of foam is engulfed in flames, and the barrier fabric fails the test.
ignition occurs, the complete piece of foam is engulfed in flames, and the barrier fabric fails the test.
However, if the BFs remain intact, the foam does not ignite, resulting in the barrier fabric passing the
However, if the BFs remain intact, the foam does not ignite, resulting in the barrier fabric passing the
test. Thus, the BF-open flame test that was considered in this study only differentiates barriers based
test. Thus, the BF-open flame test that was considered in this study only differentiates barriers based
on their structural integrity. However, relative barrier fabric effectiveness is known to vary with such
on their structural integrity. However, relative barrier fabric effectiveness is known to vary with such
properties as flammability, thermal resistance, and permeability, in addition to structural integrity
properties as flammability, thermal resistance, and permeability, in addition to structural integrity [16].
[16]. Since the BF-open flame test method seems to be primarily sensitive to barrier structural
Since the BF-open flame test method seems to be primarily sensitive to barrier structural integrity, it
integrity, it does not differentiate in terms of other important properties of BFs described earlier in
does not differentiate in terms of other important properties of BFs described earlier in [16].
[16].
Heat flux measurements suggested that roughly only one third of heat from the open flame
Heat flux measurements suggested that roughly only one third of heat from the open flame
ignition source is transferred through char forming barriers. Moreover, it was observed that the
ignition source is transferred through char forming barriers. Moreover, it was observed that the BFs
BFs can fail the test without reaching high heat flux values, i.e., the barriers fail primarily due to
can fail the test without reaching high heat flux values, i.e., the barriers fail primarily due to structural
structural failure. In many cases the FPUF pyrolysed significantly however, the test set up does not
failure. In many cases the FPUF pyrolysed significantly however, the test set up does not create
create flammable conditions despite high rates of FPUF pyrolysis as long as the barrier remains intact.
flammable conditions despite high rates of FPUF pyrolysis as long as the barrier remains intact. The
The 12.5 mm thick foam used in the BF-open flame test method provides a relatively small quantity
12.5 mm thick foam used in the BF-open flame test method provides a relatively small quantity of
of fuel that is, in some cases, either fully pyrolysed or charred before the end of the test duration
fuel that is, in some cases, either fully pyrolysed or charred before the end of the test duration of 60
of 60 s. A larger number of BFs failed in protecting the FPUF beyond the 60 s test duration when the
s. A larger number of BFs failed in protecting the FPUF beyond the 60 s test duration when the FPUF
FPUF thickness was increased from 12.5 mm to 25 mm. However, ignition times were significantly
thickness was increased from 12.5 mm to 25 mm. However, ignition times were significantly longer
longer and FPUF ignitions still occurred due to hole/crack formation in the BF, suggesting that there
and FPUF ignitions still occurred due to hole/crack formation in the BF, suggesting that there exists
a time factor in the ability of the BF to protect the foam based on the time to structural failure of the
barrier.
Polymers 2016, 8, 342
16 of 18
exists a time factor in the ability of the BF to protect the foam based on the time to structural failure of
the barrier.
The inverted test configuration with the FPUF placed on top of the BF as opposed to real
furnishing configuration, where the FPUF would be located under the fire blocking barrier fabric,
presents an implausible challenge. There may be burning scenarios where cushion could be insulted
by flames from below, but the reverse seems much more probable. In a real furnishing configuration,
the FPUF would pyrolyze and collapse when exposed to heat and flames. In the BF-open flame test
configuration, the FPUF often swells away from the ignition source, forming a dome-shaped structure
over the BF. The dome formation is a result of a specific test configuration and this would not happen
in the case of a real furniture fire. The shape and volume of the dome structure formed due to thermal
decomposition of the foam sample varied strongly with the type of BF. The variation in size and shape
likely caused a significant variation of flame heat transfer to the FPUF, which would strongly affect the
likelihood of FPUF ignition in real-world configurations.
Based on this limited experimental series, while open flame test method offers several advantages
with regard to its simplicity and ease of use, the BF-open flame test used in this study appears to assess
barrier fabrics with regard to their resistance to breaking open and exposing the protected material to
direct flame impingement. Since BF effectiveness is known to be determined by a number of other
parameters [16], it may be necessary to revise the BF-open flame test methodology to more adequately
replicate BF performance in actual applications.
Acknowledgments: Authors would like to acknowledge COST MP1105 for covering the costs to publish in
open access.
Author Contributions: Shonali Nazaré and Rick Davis conceived the project; Shonali Nazaré designed and
performed the experiments; Shonali Nazaré and William M. Pitts analyzed the data; John Shields contributed
designing and building the test device as well as data acquisition tools; Shonali Nazaré wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Troitzsch, J.H. Fires, statistics, ignition sources, and passive fire protection measures. J. Fire Sci. 2016, 34,
171–198. [CrossRef]
Hall, J.R. Estimating fires when a product is the primary fuel but not the first fuel, with an application to
upholstered furniture. Fire Technol. 2015, 51, 381–391. [CrossRef]
State of California Department of Consumer Affairs. Technical Bulletin TB-117: Requirements, Test Procedure,
and Apparatus for Testing and Flame Retardancy of Filling Materials and Fabrics; Bureau of Home Furnishings,
Department of Consumer Affairs, State of California: North Highlands, CA, USA, 1977.
California Senate Bill No. 1019, Chapter 862, Approved by Governor on September 30, 2014. Available
online: https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201320140SB1019 (accessed on
22 February 2016).
Alaee, M.; Wenning, R.J. The Significance of Brominated Flame Retardants in the Environment: Current
Understanding, Issues and Challenges. Chemosphere 2002, 46, 579–796. [CrossRef]
Letcher, R.; Behnisch, P. The State-of-Science and Trends of BFRs in the Environment. Environ. Int. 2003, 29,
663–886. [CrossRef]
Babrauskas, V.; Blum, A.; Daley, R.; Birnbaum, L. Flame Retardants in Furniture Foam: Benefits and Risks.
Fire Saf. Sci. 2011, 10, 265–278. [CrossRef]
Hawthorne, M. Toxic Roulette: Flame Retardants Get a Pass from Regulators with Little Assessment of Potential
Health Risks; Chicago Tribune: Chicago, IL, USA. Available online: http://www.pulitzer.org/files/finalists/
2013/chictrib2013/chictrib05.pdf (accessed on 26 January 2016).
Governor Brown Directs State Agencies to Revise Flammability Standards. Office of the Governor:
Sacramento, CA, USA, 2012. Available online: http://gov.ca.gov/news.php?id=17598 (accessed on
4 February 2016).
Polymers 2016, 8, 342
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
17 of 18
State of California Department of Consumer Affairs. Technical Bulletin Cal-117–2013: Requirements,
Test Procedure and Apparatus for Testing the Smolder Resistance of Materials Used in Upholstered Furniture;
Bureau of Electronic and Appliance Repair Home Furnishings and Thermal Insulation: Sacramento, CA,
USA, 2013.
Gann, R.G. The Challenge of Realizing Low Flammability Home Furnishing. In Interflam 2013, Proceedings
of the Thirteenth International Conference, Windsor, UK, 24–26 June 2013; Interscience Communications:
Greenwich, UK, 2013.
Damant, G.H.; Nurbakhsh, S. Heat release tests of mattresses and bedding systems. J. Fire Sci. 1992, 10,
386–410. [CrossRef]
Damant, G.H. Use of barriers and fire blocking layers to comply with full-scale fire tests for furnishings.
J. Fire Sci. 1996, 14, 3–25. [CrossRef]
State of California Department of Consumer Affairs. Technical Bulletin Cal-117–2013: Proposed Open Flame Test
for Barrier Materials; Bureau of Electronic and Appliance Repair Home Furnishings and Thermal Insulation:
Sacramento, CA, USA, 2013.
Nazaré, S.; Davis, R.D. A Review of Fire Blocking Technologies for Soft Furnishings. Fire Sci. Rev. 2012, 1,
1–23. [CrossRef]
Nazaré, S.; Pitts, W.M.; Flynn, S.; Shields, J.; Davis, R.D. Evaluating fire blocking performance of barrier
fabrics. Fire Mater. 2014, 38, 695–716. [CrossRef]
Nazaré, S.; Pitts, W.M.; Matko, S.; Davis, R.D. Evaluating Smoldering Behavior of Barrier Fabrics. J. Fire Sci.
2014, 32, 539–562. [CrossRef]
ASTM. ASTM D7140-07: Standard Test Method to Measure Heat Transfer through Textile Thermal Barrier Materials;
ASTM International: West Conshohocken, PA, USA, 2007.
Ohlemiller, T.; Shields, J. One- and two-sided burning of thermally thin materials. Fire Mater. 1993, 17,
103–110. [CrossRef]
Shields, T.J.; Silock, G.W.; Murray, J.J. Evaluating ignition data using the flux time product. Fire Mater. 1994,
18, 243–254. [CrossRef]
Taylor, B.N.; Kuyatt, C.E. NIST Technical Note 1297: Guidelines for Evaluating and Expressing the Uncertainty of
NIST Measurement Results; NIST: Gaithersburg, MD, USA, 1994.
JCGM 100:2008: Evaluation of Measurement Data—Guide to the Expression of Uncertainty in
Measurement. Available online: http://www.bipm.org/utils/common/documents/jcgm/JCGM_100_
2008_E.pdf (accessed on 19 May 2016).
Eiseman, J.H. A study of laboratory Bunsen burner for natural gas. J. Res. Stand. 1949, 42, 541–556. [CrossRef]
Ravey, M.; Pearce, E.M. Flexible polyurethane foam. I. Thermal decomposition of a polyether-based,
water-blown commercial type of flexible polyurethane foam. J. Appl. Polym. Sci. 1997, 63, 47–74. [CrossRef]
Rogers, F.E.; Ohlemiller, T. Smolder characteristic of flexible polyurethane foams. J. Fire Flammabl. 1980, 11,
32–44.
Bustamante, V.L.; Rogaume, T.; Guillaume, E.; Rein, G.; Torero, J.L. Analysis of principal gas products during
combustion of polyether polyurethane foam at different irradiance levels. Fire Saf. J. 2009, 44, 933–940.
[CrossRef]
Drysdale, D. An Introduction to Fire Dynamics, 2nd ed.; Wiley: Chichester, UK, 1998.
Tse, S.D.; Carlo, A.; Nde-Pello, F.; Miyasaka, K. Controlling mechanisms in the transition from smoldering to
flaming of flexible polyurethane foam. Symp. Int. Combust. 1996, 26, 1505–1513. [CrossRef]
Eggestad, J.; Johnsen, A.C. Effects of interliners on the ignitability of upholstered furniture. J. Fire Sci. 1987,
5, 152–161. [CrossRef]
Damant, G.H. Cigarette ignition of upholstered furniture. J. Fire Sci. 1995, 3, 337–349. [CrossRef]
Gandhi, S.; Spivak, S.M. A survey of upholstered furniture fabrics and implications for furniture flammability.
J. Fire Sci. 1994, 12, 284–312. [CrossRef]
Polymers 2016, 8, 342
32.
33.
18 of 18
Hightower, T.M.; Olivares, R.A.; Philippidis, D. Thermal Capacitance (Slug) Calorimeter Theory including
Heat Losses and other Decaying Processes. In Proceedings of the Thermal and Fluids Analysis Workshop
(TFAWS) 2008, San Jose, CA, USA, 18–22 August 2008; National Aeronautics and Space Administration:
Washington, DC, USA, 2008. Available online: http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/
20090008662.pdf (accessed on 1 August 2016).
Filipczak, R.; Lyon, R.E. DOT/FAA/AR-TN00/38: Heat Flux Measurements in the Ohio State University Rate of Heat
Release Apparatus; Federal Aviation Administration, Airport and Aircraft Safety Research and Development
Division: Atlantic, NJ, USA, 2000.
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).