ArticlePDF Available

Fire safety of wood construction

Authors:
  • USDA Forest Service, Forest Products Laboratory
CHAPTER 18
Fire Safety of Wood Construction
Robert H. White, Research Forest Products Technologist
Mark A. Dietenberger, Research General Engineer
18–1
Contents
Fire Safety Design and Evaluation 18–1
Types of Construction 18–2
Ignition 18–2
Exterior Fire Exposure in the Wildland–Urban
Interface 18–3
Fire Growth Within Compartment 18–3
Containment to Compartment of Origin 18–6
Fire-Performance Characteristics of Wood 18–8
Thermal Degradation of Wood 18–8
Ignition 18–9
Heat Release and Smoke 18–11
Flame Spread 18–12
Charring and Fire Resistance 18–13
Fire-Retardant-Treated Wood 18–15
Pressure Treatments 18–15
Performance Requirements 18–16
Fire-Retardant Coatings 18–17
Literature Cited 18–18
Additional References 18–18
General 18–18
Fire Test Standards 18–19
Ignition 18–19
Flame Spread 18–20
Flashover and Room/Corner Tests 18–20
Heat Release and Heat of Combustion 18–20
Combustion Products 18–21
Fire Resistance 18–21
Fire-Retardant-Treated Wood 18–22
Fire safety is an important concern in all types of construc-
tion. The high level of national concern for re safety is
reected in limitations and design requirements in building
codes. These code requirements and related re performance
data are discussed in the context of re safety design and
evaluation in the initial section of this chapter. Because
basic data on re behavior of wood products are needed to
evaluate re safety for wood construction, the second major
section of this chapter provides additional information on
re behavior and re performance characteristics of wood
products. The chapter concludes with a discussion of re-
retardant treatments that can be used to reduce the combus-
tibility of wood.
Fire Safety Design and Evaluation
Fire safety involves prevention, detection, evacuation, con-
tainment, and extinguishment. Fire prevention basically
means preventing the sustained ignition of combustible
materials by controlling either the source of heat or the
combustible materials. This involves proper design, instal-
lation or construction, and maintenance of the building and
its contents. Proper re safety measures depend upon the
occupancy or processes taking place in the building. Smoke
and heat detectors can be installed to provide early detection
of a re. Early detection is essential for ensuring adequate
time for egress. Egress, or the ability to escape from a re,
often is a critical factor in life safety. Statutory requirements
pertaining to re safety are specied in building codes or
re codes. Design deciencies are often responsible for
spread of heat and smoke in a re. Spread of a re can be
prevented with designs that limit re growth and spread
within a compartment and contain re to the compartment
of origin. Sprinklers provide improved capabilities to extin-
guish a re in its initial stages. These requirements fall into
two broad categories: material requirements and building
requirements. Material requirements include such things as
combustibility, ame spread, and re resistance. Building
requirements include area and height limitations, restops
and draftstops, doors and other exits, automatic sprinklers,
and re detectors.
Adherence to codes will result in improved re safety. Code
ofcials should be consulted early in the design of a build-
ing because the codes offer alternatives. For example, oor
areas can be increased if automatic sprinkler systems are
added. Code ofcials have the option to approve alternative
materials and methods of construction and to modify
18–2
General Technical Report FPLGTR190
provisions of the codes when equivalent re protection and
structural integrity are documented.
Most current building codes in the United States are based
on the model building code produced by the International
Code Council (ICC) (International Building Code® (IBC))
and related International Code® (I-Codes®) documents).
In addition to the documents of the ICC, the National Fire
Protection Association’s (NFPA’s) Life Safety Code (NFPA
101) provides guidelines for life safety from re in buildings
and structures. NFPA also has a model building code known
as NFPA 5000. The provisions of the ICC and NFPA docu-
ments become statutory requirements when adopted by local
or state authorities having jurisdiction.
Information on re ratings for different products and as-
semblies can be obtained from industry literature, evaluation
reports issued by ICC Evaluation Service, Inc. (ICC-ES)
and other organizations, and listings published by testing
laboratories or quality assurance agencies. Products listed
by Underwriters Laboratories, Inc. (UL), Intertek, and other
such organizations are stamped with the rating information.
The eld of re safety engineering is undergoing rapid
changes because of the development of more engineering
and scientic approaches to re safety. This development
is evidenced by the publication of the fourth edition of The
Society of Fire Protection Engineers (SFPE) Handbook of
Fire Protection Engineering. Steady advances are being
made in the elds of re dynamics, re hazard calculations,
re design calculations, and re risk analysis. Such efforts
support the worldwide trend to develop alternative building
codes based on performance criteria rather than prescriptive
requirements. Additional information on re protection can
be found in various publications of the NFPA and SFPE.
In the following sections, various aspects of building code
provisions pertaining to re safety of building materials are
discussed under the broad categories of (a) types of con-
struction, (b) ignition, (c) re growth within compartment,
(d) containment to compartment of origin, and (e) exterior
res. These are largely requirements for materials. Informa-
tion on prevention and building requirements not related to
materials (for example, detection) can be found in NFPA
publications.
Types of Construction
A central aspect of the re safety provisions of building
codes is the classication of buildings by types of construc-
tion and use or occupancy. Based on classications of build-
ing type and occupancy, the codes set limits on areas and
heights of buildings. Building codes generally recognize
ve classications of construction based on types of materi-
als and required re resistance ratings. The two classica-
tions known as Type I (re-resistant construction) and Type
II (noncombustible construction) basically restrict
the building elements to noncombustible materials. Wood
is permitted to be used more liberally in the other three
classications, which are Type III (ordinary), Type IV
(heavy timber), and Type V (light-frame). Type III construc-
tion allows smaller wood members to be used for interior
walls, oors, and roofs including wood studs, joists, trusses,
and I-joists. For Type IV (heavy timber) construction, in-
terior wood columns, beams, oors, and roofs are required
to satisfy certain minimum dimensions and no concealed
spaces are permitted. In both Types III and IV construc-
tion, exterior walls must be of noncombustible materials,
except that re-retardant-treated (FRT) wood is permitted
within exterior wall assemblies of Type III construction
when the requirements for re resistance ratings are 2-h or
less. In Type V construction, walls, oors, and roofs may be
of any dimension lumber and the exterior walls may be of
combustible materials. Types I, II, III, and V constructions
are further subdivided into two parts—A (protected) and B
(unprotected), depending on the required re resistance rat-
ings. In Type V-A (protected light-frame) construction, most
of the structural elements have a 1-h re resistance rating.
No general re resistance requirements are specied for
buildings of Type V-B (unprotected light-frame) construc-
tion. The required re resistance ratings for exterior walls
also depend on the re separation distance from the lot line,
centerline of the street, or another building. Such property
line setback requirements are intended to mitigate the risk of
exterior re exposure.
Based on their performance in the ASTM E 136 test (see list
of re test standards at end of chapter), both untreated and
FRT wood are combustible materials. However, building
codes permit substitution of FRT wood for noncombustible
materials in some specic applications otherwise limited to
noncombustible materials. Specic performance and treat-
ment requirements are dened for FRT wood used in such
applications.
In addition to type of construction, height and area limita-
tions also depend on the use or occupancy of a structure.
Fire safety is improved by automatic sprinklers, property
line setbacks, or more re-resistant construction. Building
codes recognize the improved re safety resulting from
application of these factors by increasing allowable areas
and heights beyond that designated for a particular type of
construction and occupancy. Thus, proper site planning and
building design may result in a desired building area classi-
cation being achieved with wood construction.
Ignition
The most effective ways to improve re safety are pre-
ventive actions that will reduce or eliminate the risks of
ignition. Some code provisions, such as those in electrical
codes, are designed to address this issue. Other such provi-
sions are those pertaining to separations between heated
pipes, stoves, and similar items and any combustible ma-
terial. In situations of prolonged exposures and conned
spaces, wood has been known to ignite at temperatures
much lower than the temperatures normally associated with
18–3
Chapter 18 Fire Safety of Wood Construction
wood ignition. To address this concern, a safe margin of re
safety from ignition even in cases of prolonged exposures
can be obtained if surface temperatures of heated wood are
maintained below about 80 °C, which avoids the incipient
wood degradation associated with reduction in the ignition
temperature.
Other examples of regulations addressing ignition are re-
quirements for the proper installation and treatment of cel-
lulosic installation. Proper chemical treatments of cellulosic
insulation are required to reduce its tendency for smoldering
combustion and to reduce ame spread. Cellulosic insula-
tion is regulated by a product safety standard of the U.S.
Consumer Product Safety Commission. One of the required
tests is a smoldering combustion test. Proper installation
around recessed light xtures and other electrical devices is
necessary.
Exterior Fire Exposure in the Wildland–Urban
Interface
In areas subjected to wildres, actions to remove ignition
sources around the home or other structures and prevent
easy re penetration into such buildings can signicantly
improve the chances that a structure will survive a wildre.
This includes appropriate landscaping to create a defensible
space around the structure. Particular attention should be
paid to the removal of vegetation and other combustible
exterior items (such as rewood, fence, landscape mulch)
that are close to openings (vents, windows, and doors),
combustible surfaces of the building, and softs. Openings
in building exteriors can allow the re to penetrate into the
building and cause interior ignitions. Building design and
maintenance should be done to limit the accumulation of
combustible debris that could be ignited by rebrands that
originate from burning trees and buildings, with particular
attention paid to nooks and crannies that allow accumulation
of debris. The rebrands’ distribution is such that they can
cause destruction of unprotected structures that are some
distance from the actual ames of the wildre. Regardless
of the type of material used for the exterior membrane, the
type and placement of the joints of the membrane can affect
the likelihood that a re will penetrate the exterior mem-
brane. For example, birdstops should be installed at the ends
of clay tile barrel roof coverings to prevent rebrands from
igniting the underlining substrate.
Rated roof covering materials are designated Class A, B,
or C according to their performance in the tests described
in ASTM E 108, Fire Tests of Roof Coverings. This test
standard includes intermittent ame exposure, spread of
ame, burning brand, ying brand, and rain tests. Each of
the three classes has a different version of the pass–fail test.
The Class A test is the most severe, Class C the least. In the
case of the burning brand tests, the brand for the Class B test
is larger than that for the Class C test. FRT wood shingles
and shakes are available that carry a Class B or C re rating.
A Class A rated wood roof system can be achieved by using
Class B wood shingles with specied roof deck and
underlayment.
For other exterior applications, FRT wood is tested in accor-
dance with ASTM E 84. An exterior treatment is required to
have no increase in the listed ame spread index after being
subjected to the rain test of ASTM D 2898. At the present
time, a commercial treated-wood product for exterior appli-
cations is either treated to improve re retardancy or treated
to improve resistance to decay and insects, not both.
Various websites (such as www.rewise.org) provide addi-
tional information addressing the protection of homes in the
wildland–urban interface. The national Firewise Communi-
ties program is a multi-agency effort designed to reach be-
yond the re service by involving homeowners, community
leaders, planners, developers, and others in the effort to pro-
tect people, property, and natural resources from the risk of
wildland re, before a re starts. The Firewise Communities
approach emphasizes community responsibility for planning
in the design of a safe community and effective emergency
response, along with individual responsibility for safer home
construction and design, landscaping, and maintenance.
The ICC’s International Wildland–Urban Interface Code
provides model code regulations that specically address
structures and related land use in areas subjected to wild-
res. NFPA 1144 is a standard that focuses on individual
structure hazards from wildland res. In response to losses
due to wildres, the California State Fire Marshal’s Of-
ce (www.re.ca.gov) has implemented ignition-resistant
construction standards for structures in the wildland–urban
interface. These test requirements intended to address ignit-
ability of the structure are based on tests developed at the
University of California for exterior wall siding and sheath-
ing, exterior windows, under eave, and exterior decking.
Fire Growth within Compartment
Flame Spread
Important provisions in the building codes are those that
regulate the exposed interior surface of walls, oors, and
ceilings (that is, the interior nish). Codes typically exclude
trim and incidental nish, as well as decorations and fur-
nishings that are not afxed to the structure, from the more
rigid requirements for walls and ceilings. For regulatory
purposes, interior nish materials are classied according
to their ame spread index. Thus, ame spread is one of
the most tested re performance properties of a material.
Numerous ame spread tests are used, but the one cited by
building codes is ASTM E 84 (also known as NFPA 255
and UL 723), the “25-ft tunnel” test. In this test method, the
508-mm-wide, 7.32-m-long specimen completes the top of
the tunnel furnace. Flames from a burner at one end of the
tunnel provide the re exposure, which includes forced draft
conditions. The furnace operator records the ame front
position as a function of time and the time of maximum
ame front travel during a 10-min period. The standard
18–4
General Technical Report FPLGTR190
prescribes a formula to convert these data to a ame spread
index (FSI), which is a measure of the overall rate of ame
spreading in the direction of air ow. In the building codes,
the classes for ame spread index are A (FSI of 0 to 25), B
(FSI of 26 to 75), and C (FSI of 76 to 200). Generally, codes
specify FSI for interior nish based on building occupancy,
location within the building, and availability of automatic
sprinkler protection. The more restrictive classes, Classes A
and B, are generally prescribed for stairways and corridors
that provide access to exits. In general, the more ammable
classication (Class C) is permitted for the interior nish
of other areas of the building that are not considered exit
ways or where the area in question is protected by automatic
sprinklers. In other areas, no ammability restrictions are
specied on the interior nish, and unclassied materials
(that is, more than 200 FSI) can be used. The classication
labels of I, II, and III have been used instead of A, B, and C.
The FSI for most domestic wood species is between 90
and 160 (Table 18–1). Thus, unnished lumber, 10 mm or
thicker, is generally acceptable for interior nish applica-
tions requiring a Class C rating. Fire-retardant treatments
are necessary when a Class A ame spread index is required
for a wood product. Some domestic softwood species meet
the Class B ame spread index without treatment. Other
domestic softwood species have FSIs near the upper limit of
200 for Class C. All available data for domestic hardwoods
Table 18–1. ASTM E 84 flame spread indexes for 19-mm-thick solid lumber of
various wood species as reported in the literaturea
SpeciesbFlame spread
indexc
Smoke
developed
indexcSourced
Softwoods
Yellow-cedar (Pacific Coast yellow cedar) 78 90 CWC
Baldcypress (cypress) 145–150 UL
Douglas-fir 70–100 UL
Fir, Pacific silver 69 58 CWC
Hemlock, western (West Coast) 60–75 UL
Pine, eastern white (eastern white, northern white) 85, 120–215
f
122, — CWC, UL
Pine, lodgepole 93 210 CWC
Pine, ponderosa 105–230e UL
Pine, red 142 229 CWC
Pine, Southern (southern) 130–195
f
UL
Pine, western white 75
f
UL
Redcedar, western 70 213 HPVA
Redwood 70 UL
Spruce, eastern (northern, white) 65 UL, CWC
Spruce, Sitka (western, Sitka) 100, 74 —, 74 UL, CWC
Hardwoods
Birch, yellow 105–110 UL
Cottonwood 115 UL
Maple (maple flooring) 104 CWC
Oak (red, white) 100 100 UL
Sweetgum (gum, red) 140–155 UL
Walnut 130–140 UL
Yellow-poplar (poplar) 170–185 UL
aAdditional data for domestic solid-sawn and panel products are provided in the AF&PA–AWC DCA
No. 1, “Flame Spread Performance of Wood Products.”
bIn cases where the name given in the source did not conform to the official nomenclature of the Forest
Service, the probable official nomenclature name is given and the name given by the source is given in
parentheses.
cData are as reported in the literature (dash where data do not exist). Changes in the ASTM E 84 test
method have occurred over the years. However, data indicate that the changes have not significantly
changed earlier data reported in this table. The change in the calculation procedure has usually resulted in
slightly lower flame spread results for untreated wood. Smoke developed index is not known to exceed
450, the limiting value often cited in the building codes.
dCWC, Canadian Wood Council (CWC 1996); HPVA, Hardwood Plywood Manufacturers Association
(Tests) (now Hardwood Plywood & Veneer Assoc.); UL, Underwriters Laboratories, Inc. (Wood-fire
hazard classification. Card Data Service, Serial No. UL 527, 1971).
eFootnote of UL: In 18 tests of ponderosa pine, three had values over 200 and the average of all tests is
154.
fFootnote of UL: Due to wide variations in the different species of the pine family and local connotations
of their popular names, exact identification of the types of pine tested was not possible. The effects of
differing climatic and soil conditions on the burning characteristics of given species have not been
determined.
18–5
Chapter 18 Fire Safety of Wood Construction
are for Class C. Some high-density imported hardwood spe-
cies have FSIs in Class B. Additional FSI data for domestic
solid-sawn and panel products are provided in the American
Forest and Paper Association (AF&PA)–American Wood
Council (AWC) design for code acceptance (DCA)
No. 1 (see list of references at end of chapter). Report 128
of APA–The Engineered Wood Association (APA) discusses
the ame spread indexes of construction plywood panels.
Code provisions pertaining to oors and oor coverings
include those based on the critical radiant ux test (ASTM
E 648). In the critical radiant ux test, the placement of the
radiant panel is such that the radiant heat being imposed on
the surface has a gradient in intensity down the length of
the horizontal specimen. Flames spread from the ignition
source at the end of high heat ux (or intensity) to the other
end until they reach a location where the heat ux is not suf-
cient for further propagation. This is reported as the critical
radiant ux (CRF). Thus, low CRF reects materials with
high ammability.
Depending on location and occupancy, building code re-
quirements are for a minimum critical radiant ux level
of 2.2 kW m–2 (0.22 W cm–2) for Class II or 4.5 kW m–2
(0.45 W cm–2) for Class I. These provisions are mainly
intended to address the re safety of some carpets. One
section in the International Building Code (IBC) (Sec. 804)
where this method is cited exempts wood oors and other
oor nishes of a traditional type from the requirements.
This method is also cited in standards of the National Fire
Protection Association (NFPA) such as the Life Safety
Code. Very little generic data is published on wood prod-
ucts tested in accordance with ASTM E 648. In one report
published during the development of the test, a CRF of ap-
proximately 3.5 to 4.0 kW m–2 was cited for oak ooring
(Benjamin and Davis 1979). Company literature for propri-
etary wood oor products indicates that such products can
achieve CRF in excess of the 4.5 kW m–2 for Class I. For
wood products tested in accordance with the similar Euro-
pean radiant panel test standard (EN ISO 9239-1 (2002))
(Östman and Mikkola 2006, Tsantaridis and Östman 2004),
critical heat ux (CHF) ranged from 2.6 to 5.4 kW m–2 for
25 wood oorings tested without a surface coating. Most
densities ranged from 400 to 600 kg m–3. One additional
wood ooring product had a CHF of 6.7 kW m–2. Additional
results for the wood ooring products tested with a wide
range of coating systems indicated that the non-re-
retardant coatings may signicantly improve the CHF
to levels above 4.5 kW m–2.
The critical radiant ux apparatus is also used to test the
ammability of cellulosic insulation (ASTM E 970). There
are many other test methods for ame spread or ammabil-
ity. Most are used only for research and development or
quality control, but some are used in product specications
and regulations of materials in a variety of applications.
Other tests for ammability include those that measure heat
release.
Flashover
With sufcient heat generation, the initial growth of a re
in a compartment leads to the condition known as ashover.
The visual criteria for ashover are full involvement of
the compartment and ames out the door or window
(Figure 18–1). The intensity over time of a re starting in
one room or compartment of a building depends on the
amount and distribution of combustible contents in the
room and the amount of ventilation.
The standard full-scale test for pre-ashover re growth
is the room-corner test (ASTM E 2257). In this test, a gas
burner is placed in the corner of the room, which has a sin-
gle door for ventilation. Three of the walls are lined with the
test material, and the ceiling may also be lined with the test
material. Other room-corner tests use a wood crib or similar
item as the ignition source. Such a room-corner test is used
to regulate foam plastic insulation, a material that is not
properly evaluated in the ASTM E 84 test. Observations are
made of the growth of the re and the duration of the test
until ashover occurs. Instruments record the heat genera-
tion, temperature development within the room, and the
heat ux to the oor. Results of full-scale room-corner
tests are used to validate re growth models and bench-
scale test results. In a series of room-corner tests using a
100/300-kW burner and no test material on the ceiling, the
ranking of the different wood products was consistent with
their ame spread index in the ASTM E 84 test (White and
others 1999). Another room-corner test standard (NFPA
286) is cited in codes as an alternative to ASTM E 84 for
evaluating interior wall or ceiling nishes for Class A
applications.
Figure 18–1.
Flashover in
standard room
test.
Smoke and Toxic Gases
One of the most important problems associated with evacu-
ation during a re is the smoke produced. The term smoke
is frequently used in an all-inclusive sense to mean the
mixture of pyrolysis products and air that is present near
the re site. In this context, smoke contains gases, solid
particles, and droplets of liquid. Smoke presents potential
hazards because it interacts with light to obscure vision and
because it contains noxious and toxic substances. Generally,
two approaches are used to deal with the smoke problem:
limit smoke production and control the smoke that has been
produced. The control of smoke ow is most often a factor
in the design and construction of large or tall buildings. In
these buildings, combustion products may have serious ef-
fects in areas remote from the actual re site.
The smoke yield restrictions in building codes are also
based on data from the ASTM E 84 standard. Smoke mea-
surement is based on a percentage attenuation of white light
passing through the tunnel exhaust stream and detected
by a photocell. This is converted to the smoke developed
index (SDI), with red oak ooring set at 100. Flame spread
requirements for interior nish generally are linked to an
added requirement that the SDI be less than 450. Available
SDI data for wood products are less than 450 (Table 181).
In the 1970s, the apparatus known as the NBS smoke cham-
ber was developed and approved as an ASTM standard for
research and development (ASTM E 662). This test is a
static smoke test because the specimen is tested in a closed
chamber of xed volume and the light attenuation is re-
corded over a known optical path length. The corresponding
light transmission is reported as specic optical density as a
function of time. Samples are normally tested in both am-
ing (pilot ame) and nonaming conditions using a radiant
ux of 25 kW m–2. Some restrictions in product specica-
tions are based on the smoke box test (ASTM E 662). As
discussed in a later section, dynamic measurements of
smoke can be obtained with the cone calorimeter
(ASTM E 1354) and the room-corner test (ASTM E 2257).
Toxicity of combustion products is a concern. Fire victims
are often not touched by ames but die as a result of ex-
posure to smoke, toxic gases, or oxygen depletion. These
life-threatening conditions can result from burning contents,
such as furnishings, as well as from the structural materials
involved. The toxicity resulting from the thermal decompo-
sition of wood and cellulosic substances is complex because
of the wide variety of types of wood smoke. Composition
and the concentration of individual constituents depend
on such factors as the re exposure, oxygen and moisture
present, species of wood, any treatments or nishes that
may have been applied, and other considerations. The vast
majority of res that attain ashover do generate dangerous
levels of carbon monoxide, independent of what is burning.
Carbon monoxide is a particularly insidious toxic gas and is
often generated in signicant amounts in wood res. Small
amounts of carbon monoxide are particularly toxic because
the hemoglobin in the blood is much more likely to combine
with carbon monoxide than with oxygen, even with plenty
of breathable oxygen (carboxyhemoglobin) present.
Containment to Compartment of Origin
The growth, intensity, and duration of the re is the “load”
that determines whether a re is conned to the room of ori-
gin. Whether a given re will be contained to the compart-
ment depends on the re resistance of the walls, doors, ceil-
ings, and oors of the compartment. Requirements for re
resistance or re resistance ratings of structural members
and assemblies are another major component of the building
code provisions. In this context, re resistance is the ability
of materials or their assemblies to prevent or retard the pas-
sage of excessive heat, hot gases, or ames while continu-
ing to support their structural loads. Fire resistance ratings
are usually obtained by conducting standard re tests. The
standard re resistance test (ASTM E 119) has three failure
criteria: element collapse, passage of ames, or excessive
temperature rise on the non-re-exposed surface (average
increase of several locations exceeding 139 or 181 °C at
a single location).
Doors can be critical in preventing the spread of res. Doors
left open or doors with little re resistance can easily defeat
the purpose of a re-rated wall or partition. Listings of re-
rated doors, frames, and accessories are provided by vari-
ous re testing agencies. When a re-rated door is selected,
details about which type of door, mounting, hardware, and
closing mechanism need to be considered.
Fires in buildings can spread by the movement of hot re
gases through open channels in concealed spaces. Codes
specify where reblocking and draftstops are required in
concealed spaces, and they must be designed to interfere
with the passage of the re up or across a building. In addi-
tion to going along halls, stairways, and other large spaces,
heated gases also follow the concealed spaces between oor
joists and between studs in partitions and walls of frame
construction. Obstruction of these hidden channels provides
an effective means of restricting re from spreading to
other parts of the structure. Fireblockings are materials used
to resist the spread of ames via concealed spaces within
building components such as oors and walls. They are gen-
erally used in vertical spaces such as stud cavities to block
upward spread of a re. Draftstops are barriers intended
to restrict the movement of air within concealed areas of a
building. They are typically used to restrict horizontal dis-
persion of hot gases and smoke in larger concealed spaces
such as those found within wood joist oor assemblies with
suspended dropped ceilings or within an attic space with
pitched chord trusses.
Exposed Wood Members
The self-insulating quality of wood, particularly in the large
wood sections of heavy timber construction, is an important
18–6
General Technical Report FPLGTR190
factor in providing a degree of re resistance. In Type IV or
heavy timber construction, the need for re resistance re-
quirements is achieved in the codes by specifying minimum
sizes for the various members or portions of a building and
other prescriptive requirements. In this type of construction,
the wood members are not required to have specic re
resistance ratings. The acceptance of heavy timber construc-
tion is based on historical experience with its performance
in actual res. Proper heavy timber construction includes
using approved fastenings, avoiding concealed spaces under
oors or roofs, and providing required re resistance in the
interior and exterior walls.
The availability and code acceptance of a procedure to
calculate the re resistance ratings for large timber beams
and columns have allowed their use in re-rated buildings
not classied as Type IV (heavy timber) construction. In
the other types of construction, the structural members and
assemblies are required to have specied re resistance
ratings. There are two accepted procedures for calculating
the re ratings of exposed wood members. In the rst such
procedure, the equations are simple algebraic equations that
only need the dimensions of the beam or column and a load
factor. Determination of the load factor requires the mini-
mum dimension of column, the applied load as a percentage
of the full allowable design load, and the effective column
length. The acceptance of this procedure is normally limited
to beams and column with nominal dimensions of 152 mm
(6 in.) or greater and for re ratings of 1 h or less. This pro-
cedure is applicable to glued-laminated timbers that utilize
standard laminating combinations. Because the outer tension
laminate of a glued-laminated beam is charred in a 1-h re
exposure, a core lamination of a beam needs to be removed
and the equivalent of an extra nominal 51-mm- (2-in.-) thick
outer tension lamination added to the bottom of the beam.
Details on this procedure can be found in various industry
publications (American Institute of Timber Construction
(AITC) Technical Note 7, AF&PA-AWC DCA #2, APA
Publication EWS Y245A) and the IBC.
A second more exible mechanistic procedure was incor-
porated within the National Design Specication for Wood
Construction (NDS®) in 2001 and is referred to as the NDS
Method. As an explicit engineering method, it is applicable
to all wood structural members covered under the NDS,
including structural composite lumber wood members. Nor-
mal engineering calculations of the ultimate load capacity
of the structural wood element are adjusted for reductions in
dimensions with time as the result of charring. As discussed
more in a later section, a char depth of 38 mm (1.5 in.) at
1 h is generally used for solid-sawn and structural glued-
laminated softwood members. The char depth is adjusted
upward by 20% to account for the effect of elevated tem-
peratures on the mechanical properties of the wood near the
wood–char interface. This procedure also requires that core
lamination(s) of glued-laminated beams be replaced by extra
outer tension laminate(s). A provision of the NDS procedure
addresses the structural integrity performance criteria for
timber decks, but the thermal separation criteria are not ad-
dressed. This second procedure was developed by the Amer-
ican Wood Council and is fully discussed in their Technical
Report No. 10. Fire resistance tests on glued-laminated
specimens and structural composite lumber products loaded
in tension are discussed in FPL publications.
The re resistance of glued-laminated structural members,
such as arches, beams, and columns, is approximately
equivalent to the re resistance of solid members of similar
size. Laminated members glued with traditional phenol,
resorcinol, or melamine adhesives are generally considered
to be at least equal in their re resistance to a one-piece
member of the same size. In recent years, the re resistance
performance of structural wood members manufactured
with adhesives has been of intense interest. As a result of
concerns about some adhesives that were being used in
ngerjointed lumber, industry test protocols and accep-
tance criteria were developed to address this issue. When a
wood-frame assembly is required to have a re resistance
rating, any nger-jointed lumber within the assembly must
include the HRA designation for heat-resistant adhesives in
the grademark. The designation is part of the Glued Lumber
Policy of the American Lumber Standard Committee, Inc.
The activities to address questions concerning the adhesives
have included the development of ASTM standard test
methods and revisions to the ASTM standard specications
for the applicable wood products.
Light-Frame Assemblies
Light-frame wood construction can provide a high degree
of re containment through use of gypsum board as the in-
terior nish. This effective protective membrane provides
the initial re resistance rating. Many recognized assemblies
involving wood-frame walls, oors, and roofs provide a
1- or 2-h re resistance rating. Fire-rated gypsum board
(Type X or C) is used in rated assemblies. Type X and the
higher grade Type C gypsum boards have textile glass la-
ments and other ingredients that help to keep the gypsum
core intact during a re. Fire resistance ratings of various
assemblies are listed in the IBC and other publications such
as the Gypsum Association Fire Resistance Design Manual,
AF&PA-AWC DCA #3, and product directories of listing
organizations, such as UL and Intertek. Traditional construc-
tions of regular gypsum wallboard (that is, not re rated)
or lath and plaster over wood joists and studs have re
resistance ratings of 15 to 30 min. In addition to re-rated
assemblies constructed of sawn lumber, there are rated as-
semblies for I-joists and wood trusses.
Fire-rated assemblies are generally tested in accordance
with ASTM E 119 while loaded to 100% of the allowable
design load calculated using the NDS. The calculation of the
allowable design load of a wood stud wall is described in
ASTM D 6513. Some wood stud wall assemblies were
tested with a load equivalent to 78% of the current design
18–7
Chapter 18 Fire Safety of Wood Construction
load (NDS dated 2005) calculated using a le/d of 33. Less
than full design load in the re test imposes a load restric-
tion on the rated assembly.
While re resistance ratings are for the entire wall, oor, or
roof assembly, the re resistance of a wall or oor can be
viewed as the sum of the resistance of the interior nish and
the resistance of the framing members. In a code-accepted
procedure, the re rating of a light-frame assembly is cal-
culated by adding the tabulated times for the re-exposed
membrane to the tabulated times for the framing. For ex-
ample, the re resistance rating of a wood stud wall with
16-mm-thick Type X gypsum board and rock wool insula-
tion is computed by adding the 20 min listed for the stud
wall, the 40 min listed for the gypsum board, and the 15 min
listed for the rock wool insulation to obtain a rating for the
assembly of 75 min. Additional information on this compo-
nent additive method (CAM) can be found in the IBC and
AF&PA DCA No. 4. More sophisticated mechanistic models
have been developed.
The relatively good structural behavior of a traditional wood
member in a re test results from the fact that its strength is
generally uniform through the mass of the piece. Thus, the
unburned fraction of the member retains high strength, and
its load-carrying capacity is diminished only in proportion
to its loss of cross section. Innovative designs for structural
wood members may reduce the mass of the member and
locate the principal load-carrying components at the outer
edges where they are most vulnerable to re, as in structural
sandwich panels. With high strength facings attached to a
low-strength core, unprotected load-bearing sandwich pan-
els have failed to support their load in less than 6 min when
tested in the standard test. If a sandwich panel is to be used
as a load-bearing assembly, it should be protected with gyp-
sum wallboard or some other thermal barrier. In any protect-
ed assembly, the performance of the protective membrane is
the critical factor in the performance of the assembly.
Unprotected light-frame wood buildings do not have the
natural re resistance achieved with heavier wood members.
In these, as in all buildings, attention to good construction
details is important to minimize re hazards. Quality of
workmanship is important in achieving adequate re resis-
tance. Inadequate nailing and less than required thickness of
the interior nish can reduce the re resistance of an assem-
bly. The method of fastening the interior nish to the fram-
ing members and the treatment of the joints are signicant
factors in the re resistance of an assembly. The type and
quantity of any insulation installed within the assembly may
also affect the re resistance of an assembly.
Any penetration in the membrane must be addressed with
the appropriate re protection measures. This includes the
junction of re-rated assemblies with unrated assemblies.
Fire stop systems are used to properly seal the penetration
of re-rated assemblies by pipes and other utilities.
Through-penetration re stops are tested in accordance with
ASTM E 814. Electrical receptacle outlets, pipe chases, and
other through openings that are not adequately restopped
can affect the re resistance. In addition to the design of
walls, ceilings, oors, and roofs for re resistance, stair-
ways, doors, and restops are of particular importance.
Fire-Performance Characteristics
of Wood
Several characteristics are used to quantify the burning
behavior of wood when exposed to heat and air, including
thermal degradation of wood, ignition from heat sources,
heat and smoke release, ame spread in heated environ-
ments, and charring rates in a contained room.
Thermal Degradation of Wood
As wood reaches elevated temperatures, the different chemi-
cal components undergo thermal degradation that affect
wood performance. The extent of the changes depends on
the temperature level and length of time under exposure
conditions. At temperatures below 100 °C, permanent re-
ductions in strength can occur, and its magnitude depends
on moisture content, heating medium, exposure period, and
species. The strength degradation is probably due to depo-
lymerization reactions (involving no carbohydrate weight
loss). The little research done on the chemical mechanism
has found a kinetic basis (involving activation energy, pre-
exponential factor, and order of reaction) of relating strength
reduction to temperature. Chemical bonds begin to break at
temperatures above 100 °C and are manifested as carbohy-
drate weight losses of various types that increases with the
temperature. Literature reviews by Bryan (1998), Shaza-
deh (1984), Atreya (1983), and Browne (1958) reveal the
following four temperature regimes of wood pyrolysis and
corresponding pyrolysis kinetics.
Between 100 and 200 °C, wood becomes dehydrated and
generates water vapor and other noncombustible gases
including CO2, formic acid, acetic acid, and H2O. With pro-
longed exposures at higher temperatures, wood can become
charred. Exothermic oxidation reactions can occur because
ambient air can diffuse into and react with the developing
porous char residue.
From 200 to 300 °C, some wood components begin to un-
dergo signicant pyrolysis and, in addition to gases listed
above, signicant amounts of CO and high-boiling-point tar
are given off. The hemicelluloses and lignin components are
pyrolyzed in the range of 200 to 300 °C and 225 to 450 °C,
respectively. Much of the acetic acid liberated from wood
pyrolysis is attributed to deactylation of hemicellulose. De-
hydration reactions beginning around 200 °C are primarily
responsible for pyrolysis of lignin and result in a high char
yield for wood. Although the cellulose remains mostly un-
pyrolyzed, its thermal degradation can be accelerated in the
presence of water, acids, and oxygen. As the temperature
18–8
General Technical Report FPLGTR190
increases, the degree of polymerization of cellulose de-
creases further, free radicals appear and carbonyl, carboxyl,
and hydroperoxide groups are formed. Overall pyrolysis
reactions are endothermic due to decreasing dehydration
and increasing CO formation from porous char reactions
with H2O and CO2 with increasing temperature. During this
“low-temperature pathway” of pyrolysis, the exothermic
reactions of exposed char and volatiles with atmospheric
oxygen are manifested as glowing combustion.
The third temperature regime is from 300 to 450 °C because
of the vigorous production of ammable volatiles. This be-
gins with the signicant depolymerization of cellulose in the
range of 300 to 350 °C. Also around 300 °C, aliphatic side
chains start splitting off from the aromatic ring in the lignin.
Finally, the carbon–carbon linkage between lignin structural
units is cleaved at 370 to 400 °C. The degradation reaction
of lignin is an exothermic reaction, with peaks occurring be-
tween 225 and 450 °C; temperatures and amplitudes of these
peaks depend on whether the samples were pyrolyzed un-
der nitrogen or air. All wood components end their volatile
emissions at around 450 °C. The presence of minerals and
moisture within the wood tend to smear the separate pyroly-
sis processes of the major wood components. In this “high-
temperature pathway,” pyrolysis of wood results in overall
low char residues of around 25% or less of the original dry
weight. Many re retardants work by shifting wood degra-
dation to the “low-temperature pathway,” which reduces the
volatiles available for aming combustion.
Above 450 °C, the remaining wood residue is an activated
char that undergoes further degradation by being oxidized to
CO2, CO, and H2O until only ashes remain. This is referred
to as afterglow.
The complex nature of wood pyrolysis often leads to select-
ing empirical kinetic parameters of wood pyrolysis appli-
cable to specic cases. Considering the degrading wood to
be at low elevated temperature over a long time period and
ignoring volatile emissions, a simple rst-order reaction fol-
lowing the Arrhenius equation, dm/dt = –mA exp(–E/RT),
was found practical. In this equation, m is mass of specimen,
t is time, A is the preexponential factor, E is activation en-
ergy, R is the universal gas constant, and T is temperature in
kelvins. The simplest heating environment for determination
of these kinetic parameters is isothermal, constant pressure,
and uniform ow gas exposures on a nominally thick speci-
men. As an example, Stamm (1955) reported on mass loss
of three coniferous wood sticks (1 by 1 by 6 in.)—Southern
and white pine, Sitka spruce, and Douglas-r—that were
heated in a drying oven in a temperature range of 93.5 to
250 °C. The t of the Arrhenius equation to the data re-
sulted in the values of A = 6.23 × 107 s–1 and E = 124 kJ
mol–1. If these same woods were exposed to steam instead
of being oven dried, degradation was much faster. With the
corresponding kinetic parameters, A = 82.9 s–1 and E = 66
kJ mol–1, Stamm concluded that steam seemed to act as a
catalyst because of signicant reduction in the value of acti-
vation energy. Shazadeh (1984) showed that pyrolysis pro-
ceeds faster in air than in an inert atmosphere and that this
difference gradually diminishes around 310 °C. The value of
activation energy reported at large for pyrolysis in air varied
from 96 to 147 kJ mol–1.
In another special case, a simple dual reaction model could
distinguish between the low- and high- temperature path-
ways for quantifying the effect of re retardant on wood
pyrolysis. The reaction equation, dm/dt = (mendm)[A1
exp(–E1/RT) + A2 exp(–E2/RT)] , was found suitable by
Tang (1967). In this equation, mend is the ending char mass,
and subscripts 1 and 2 represent low- and high-temperature
pathways, respectively. A dynamic thermogravimetry was
used to span the temperature to 500 °C at a rate of 3 °C per
minute using tiny wood particles. The runs were made in
triplicate for ponderosa pine sapwood, lignin, and alpha-
cellulose samples with ve different inorganic salt treat-
ments. Tang’s derived values for the untreated wood are
mend = 0.21 of initial weight, A1 = 3.2 × 105 s–1, E1 = 96
kJ mol–1, A2 = 6.5e+16 s–1, and E2 = 226 kJ mol–1. A well-
known re-retardant-treatment chemical, monobasic am-
monium phosphate, was the most effective chemical tested
in that char yield was increased to 40% and E1 decreased to
80 kJ mol–1, thereby promoting most volatile loss through
the low-temperature pathway. The alpha-cellulose reacted
to the chemicals similarly as the wood, while the lignin did
not seem to be affected much by the chemicals. From this
we conclude that ammable volatiles generated by the cellu-
lose component of wood are signicantly reduced with re
retardant treatment. For applications to biomass energy and
re growth phenomology, the kinetic parameters become
essential to describe ammable volatiles and their heat of
combustion but are very complicated (Dietenberger 2002).
Modern pyrolysis models now include competing reactions
to produce char, tar, and noncondensing gases from wood as
well as the secondary reaction of tar decomposition.
Ignition
Ignition of wood is the start of a visual and sustained com-
bustion (smoldering, glow, or ame) fueled by wood pyroly-
sis. Therefore the ow of energy or heat ux from a re or
other heated objects to the wood material to induce pyroly-
sis is a necessary condition of ignition. A sufcient condi-
tion of aming ignition is the mixing together of volatiles
and air with the right composition in a temperature range
of about 400 to 500 °C. An ignition source (pilot or spark
plug) is therefore usually placed where optimum mixing of
volatiles and air can occur for a given ignition test. In many
such tests the surface temperature of wood materials has
been measured in the range of 300 to 400 °C prior to piloted
ignition. This also coincides with the third regime of wood
pyrolysis in which there is a signicant production of am-
mable volatiles. However, it is possible for smoldering or
18–9
Chapter 18 Fire Safety of Wood Construction
glow to exist prior to aming ignition if the imposed radia-
tive or convective heating causes the wood surface to reach
200 °C or higher for the second regime of wood pyrolysis.
Indeed, unpiloted ignition is ignition that occurs where no
pilot source is available. Ignition associated with smoldering
is another important mechanism by which res are initiated.
Therefore, to study aming or piloted ignition, a high heat
ux (from radiant heater) causes surface temperature to rap-
idly reach at least 300 °C to minimize inuence of unwanted
smoldering or glow at lower surface temperatures. Surface
temperature at ignition has been an elusive quantity that
was experimentally difcult to obtain, but relatively recent
studies show some consistency. For various horizontally
orientated woods with specic gravities ranging from 0.33
to 0.69, the average surface temperature at ignition increases
from 347 °C at imposed heat ux of 36 kW m–2 to 377 °C
at imposed heat ux of 18 kW m–2. This increase in the ig-
nition temperature is due to the slow decomposition of the
material at the surface and the resulting buildup of the char
layer at low heat uxes (Atreya 1983). In the case of natu-
rally high charring material such as redwood that has high
lignin and low extractives, the measured averaged ignition
temperatures were 353, 364, and 367 °C for material thick-
nesses of 19, 1.8, and 0.9 mm, respectively, for various
heat ux values as measured in the cone calorimeter
(ASTM E 1354) (Dietenberger 2004). This equipment
along with the lateral ignition and ame spread test (LIFT)
apparatus (ASTM E 1321) are used to obtain data on time to
piloted ignition as a function of heater irradiance. From such
tests, values of ignition temperature, critical ignition ux
(heat ux below which ignition would not occur), and ther-
mophysical properties have been derived using a transient
heat conduction theory (Table 18–2). In the case of red-
wood, the overall piloted ignition temperature was derived
to be 365 °C (638 K) in agreement with measured values,
regardless of heat ux, thickness, moisture content, surface
orientation, and thin reective paint coating. The critical
heat ux was derived to be higher on the LIFT apparatus
than on the cone calorimeter primarily due to the different
convective coefcients (Dietenberger 1996). However, the
heat properties of heat capacity and thermal conductivity
were found to be strongly dependent on density, mois-
ture content, and internal elevated temperatures. Thermal
conductivity has an adjustment factor for composite, engi-
neered, or treated wood products. Critical heat uxes
for ignition have been calculated to be between 10 and
13 kW m–2 for a range of wood products. For exposure to
a constant heat ux, ignition times for solid wood typically
ranged from 3 s for heat ux of 55 kW m–2 to 930 s for heat
ux of 18 kW m–2. Estimates of piloted ignition in various
scenarios can be obtained using the derived thermal proper-
ties listed in Table 18–2 and an applicable heat conduction
theory (Dietenberger 2004).
18–10
General Technical Report FPLGTR190
Table 18
2. Derived wood-based thermophysical parameters of ignitability
Material
Thickness
(mm)
Density
(kg m–3)
ρ
Moisture
content (%)
M
Material
emissivity ra
Tig
(K)
k/
ca
(m2/s)
x107
k
ca
(kJ2 m–4 K–2 s–1)
Gypsum board, Type X 16.5 662 0.9 N/A 608.5 3.74 0.451
FRT Douglas-fir plywood 11.8 563 9.48 0.9 0.86 646.8 1.37 0.261
Oak veneer plywood 13 479 6.85 0.9 1.11 563 1.77 0.413
FRT plywood (Forintek) 11.5 599 11.17 0.9 0.86 650 1.31 0.346
Douglas-fir plywood (ASTM) 11.5 537 9.88 0.85 0.863 604.6 1.37 0.221
FRT Southern Pine plywood 11 606 8.38 0.9 1.43 672 2.26 0.547
Douglas-fir plywood (MB) 12 549 6.74 0.89 0.86 619 1.38 0.233
Southern Pine plywood 11 605 7.45 0.88 0.86 620 1.38 0.29
Particleboard 13 794 6.69 0.88 1.72 563 2.72 0.763
Oriented strandboard 11 643 5.88 0.88 0.985 599 1.54 0.342
Hardboard 6 1,026 5.21 0.88 0.604 593 0.904 0.504
Redwood lumber 19 421 7.05 0.86 1.0 638 1.67 0.173
White spruce lumber 17 479 7.68 0.82 1.0 621 1.67 0.201
Southern Pine boards 18 537 7.82 0.88 1.0 644 1.63 0.26
Waferboard 13 631 5.14 0.88 1.62 563 2.69 0.442
aFormulas for wood thermal conductivity k, heat capacity c, and density ρ, at elevated temperatures used to calculate thermal inertia kρc and
thermal diffusivity k/ρc are as follows:
 
113
m
3
od KkWm1029701864.010004064.01941.0 TMrk
 
11
mKkgkJ297025.0125.1
TMc
 
3
od mkg01.01
M
where Tig is ignition temperature, ambient temperature Ta = 297 K, mean temperature Tm = (Ta + Tig)/2, and the parameter r is an adjustment factor
used in the calculation of the thermal conductivity for composite, engineered, or treated wood products (Dietenberger 2004).
Some, typically old, apparatuses for testing piloted ignition
measured the temperature of the air ow rather than the
imposed heat ux with the time to ignition measurement.
These results were often reported as the ignition temperature
and as varying with time to ignition, which is misleading.
When the imposed heat ux is due to a radiant source, such
reported air ow ignition temperature can be as much as
100 °C lower than the ignition surface temperature. For a
proper heat conduction analysis in deriving thermal proper-
ties, measurements of the radiant source ux and air ow
rate are also required. Because imposed heat ux to the sur-
face and the surface ignition temperature are the factors that
directly determine ignition, some data of piloted ignition are
inadequate or misleading.
Unpiloted ignition depends on special circumstances that re-
sult in different ranges of ignition temperatures. At this time,
it is not possible to give specic ignition data that apply to
a broad range of cases. For radiant heating of cellulosic
solids, unpiloted transient ignition has been reported at
600 °C. With convective heating of wood, unpiloted
ignition has been reported as low as 270 °C and as high as
470 °C. Unpiloted spontaneous ignition can occur when a
heat source within the wood product is located such that the
heat is not readily dissipated. This kind of ignition involves
smoldering and generally occurs over a longer period of
time. Continuous smoking is visual evidence of smoldering,
which is sustained combustion within the pyrolyzing mate-
rial. Although smoldering can be initiated by an external
ignition source, a particularly dangerous smoldering is that
initiated by internal heat generation. Examples of such res
are (a) panels or paper removed from the press or dryer and
stacked in large piles without adequate cooling and (b) very
large piles of chips or sawdust with internal exothermic re-
actions such as biological activities. Potential mechanisms
of internal heat generation include respiration, metabolism
of microorganisms, heat of pyrolysis, abiotic oxidation, and
adsorptive heat. These mechanisms, often in combination,
may proceed to smoldering or aming ignition through a
thermal runaway effect within the pile if sufcient heat is
generated and is not dissipated. The minimum environmen-
tal temperature to achieve smoldering ignition decreases
with the increases in specimen mass and air ventilation, and
can be as low as air temperatures for large ventilating piles.
Therefore, safe shipping or storage with wood chips, dust,
or pellets often depends on anecdotal knowledge that ad-
vises maximum pile size or ventilation constraints, or both
(Babrauskas 2003).
Unpiloted ignitions that involve wood exposed to low-level
external heat sources over very long periods are an area of
dispute. This kind of ignition, which involves considerable
charring, does appear to occur, based on re investigations.
However, these circumstances do not lend themselves easily
to experimentation and observation. There is some evidence
that the char produced under low heating temperatures can
have a different chemical composition, which results in a
somewhat lower ignition temperature than normally re-
corded. Thus, a major issue is the question of safe working
temperature for wood exposed for long periods. Tempera-
tures between 80 and 100 °C have been recommended as
safe surface temperatures for wood. As noted earlier, to ad-
dress this concern, a safe margin of re safety from ignition
can be obtained if surface temperatures of heated wood are
maintained below about 80 °C, which avoids the incipient
wood degradation associated with reduction in ignition
temperature.
Heat Release and Smoke
Heat release rates are important because they indicate the
potential re hazard of a material and also the combustibil-
ity of a material. Materials that release their potential chemi-
cal energy (and also the smoke and toxic gases) relatively
quickly are more hazardous than those that release it more
slowly. There are materials that will not pass the current
denition of noncombustible in the model codes but will
release only limited amounts of heat during the initial and
critical periods of re exposure. There is also some criticism
of using limited ammability to partially dene noncom-
bustibility. One early attempt was to dene combustibility
in terms of heat release in a potential heat method (NFPA
259), with the low levels used to dene low combustibil-
ity or noncombustibility. This test method is being used to
regulate materials under some codes. The ground-up wood
sample in this method is completely consumed during the
exposure to 750 °C for 2 h, which makes the potential heat
for wood identical to the gross heat of combustion from the
oxygen bomb calorimeter. The typical gross heat of combus-
tion averaged around 20 MJ kg–1 for ovendried wood, de-
pending on the lignin and extractive content of the wood.
A better or a supplementary measure of degrees of combus-
tibility is a determination of the rate of heat release (RHR)
or heat release rate (HRR). This measurement efciently
assesses the relative heat contribution of materials—thick,
thin, untreated, or treated—under re exposure. The cone
calorimeter (ASTM E 1354) is currently the most common-
ly used bench-scale HRR apparatus and is based on
the oxygen consumption method. An average value of
13.1 kJ g–1 of oxygen consumed was the constant found for
organic solids and is accurate with very few exceptions to
within 5%. In the specic case of wood volatiles aming
and wood char glowing, this oxygen consumption constant
was reconrmed at the value of 13.23 kJ g–1 (Dietenberger
2002). Thus, it is sufcient to measure the mass ow rate
of oxygen consumed in a combustion system to determine
the net HRR. The intermediate-scale apparatus (ASTM E
1623) for testing 1- by 1-m assemblies or composites and
the room full-scale test (ASTM E 2257) also use the oxygen
consumption technique to measure the HRR of res at larger
scales.
18–11
Chapter 18 Fire Safety of Wood Construction
The cone calorimeter is ideal for product development with
its small specimen size of 100 by 100 mm. The specimen is
continuously weighed by use of a load cell. In conjunction
with HRR measurements, the effective heat of combustion
as a function of time is calculated by the ASTM E 1354
method. Basically, the effective heat of combustion is the
HRR divided by the mass loss rate as determined from the
cone calorimeter test as a function of time. Typical HRR
proles, as shown in Figure 18–2, begin with a sharp peak
upon ignition, and as the surface chars, the HRR drops to
some minimum value. After the thermal wave travels com-
pletely through the wood thickness, the back side of a wood
sample reaches pyrolysis temperature, thus giving rise to a
second, broader, and even higher HRR peak. For FRT wood
products, the rst HRR peak may be reduced or eliminated.
Heat release rate depends upon the intensity of the imposed
heat ux. Generally, the averaged effective heat of combus-
tion is about 65% of the oxygen bomb heat of combustion
(higher heating value), with a small linear increase with ir-
radiance. The HRR itself has a large linear increase with the
heat ux. This information along with a representation of
the heat release prole shown in Figure 18–2 has been used
to model or correlate with large scale re growth such as
the Steiner tunnel test and the room-corner re test (Dieten-
berger and White 2001)
The cone calorimeter is also used to obtain dynamic mea-
surements of smoke consisting principally of soot and CO in
the overventilated res and of white smoke during unignited
pyrolysis and smoldering. The measurements are dynamic
in that smoke continuously ows out the exhaust pipe where
optical density and CO are measured continuously. This
contrasts with a static smoke test in which the specimen is
tested in a closed chamber of xed volume and the light at-
tenuation is recorded over a known optical path length. In
the dynamic measurements of smoke, the appropriate smoke
parameter is the smoke release rate (SRR), which is the opti-
cal density multiplied by the volume ow rate of air into the
exhaust pipe and divided by the product of exposed surface
area of the specimen and the light path length. Often the
smoke extinction area, which is the product of SRR and the
specimen area, is preferred because it can be correlated lin-
early with HRR in many cases. This also permits compari-
son with the smoke measured in the room-corner re test
because HRR is a readily available test result (Dietenberger
and Grexa 2000). Although SRR can be integrated with time
to get the same units as the specic optical density, they
are not equivalent because static tests involve the direct ac-
cumulation of smoke in a volume, whereas SRR involves
accumulation of freshly entrained air volume ow for each
unit of smoke. Methods investigated to correlate smoke be-
tween different tests included alternative parameters such as
particulate mass emitted per area of exposed sample. As per-
taining to CO production, some amount of correlation has
been obtained between the cone calorimeter’s CO mass ow
rate as normalized by HRR to the corresponding parameter
measured from the post ashover gases during the room-
corner re test. Thermal degradation of white smoke from
wood into simpler gases within the underventilated re test
room during post ashover is not presently well understood
and can have dramatic effects on thermal radiation within
the room, which in turn affects wood pyrolysis rates.
Flame Spread
The spread of ames over solids is a very important phe-
nomenon in the growth of compartment res. Indeed, in
res where large fuel surfaces are involved, increase in HRR
with time is primarily due to increase in burning area. Much
data have been acquired with the ame spread tests used in
building codes. Table 18–1 lists the FSI and smoke index
of ASTM E 84 for solid wood. Some consistencies in the
FSI behavior of the hardwood species can be related to their
density (White 2000). Considerable variations are found for
wood-based composites; for example, the FSI of four struc-
tural akeboards ranged from 71 to 189.
As a prescriptive regulation, the ASTM E 84 tunnel test is
a success in the reduction of re hazards but is impractical
in providing scientic data for re modeling or in useful
bench-scale tests for product development. Other full-scale
tests (such as the room-corner test) can produce quite differ-
ent results because of the size of the ignition burner or test
geometry. This is the case with foam plastic panels that melt
and drip during a re test. In the tunnel test, with the test
material on top, a material that melts can have low amma-
bility because the specimen does not stay in place. With an
adequate burner in the room-corner test, the same material
will exhibit very high ammability.
A ame spreads over a solid material when part of the
fuel, ahead of the pyrolysis front, is heated to the critical
18–12
General Technical Report FPLGTR190
0
50
100
150
200
250
300
350
100 200 300 400 500 600 700 800 900 1,000
Time (s)
Heat release rate (kW m-2)
65 kW m-2
50 kW m-2
35 kW m-2
20 kW m-2
Figure 18–2. Heat release rate curves for 12-mm-thick
oriented strandboard (OSB) exposed to constant heat
flux of 20, 35, 50 and 65 kW m–2.
condition of ignition. The rate of ame spread is controlled
by how rapidly the fuel reaches the ignition temperature in
response to heating by the ame front and external sources.
The material’s thermal conductivity, heat capacitance,
thickness, and blackbody surface reectivity inuence the
material’s thermal response, and an increase in the values of
these properties corresponds to a decrease in ame spread
rate. On the other hand, an increase in values of the ame
features, such as the imposed surface uxes and spatial
lengths, corresponds to an increase in the ame spread rate.
Flame spread occurs in different congurations, which are
organized by orientation of the fuel and direction of the
main ow of gases relative to that of ame spread. Down-
ward and lateral creeping ame spread involves a fuel ori-
entation with buoyantly heated air owing opposite of the
ame spread direction. Related bench-scale test methods are
ASTM E 162 for downward ame spread, ASTM E 648 for
horizontal ame spread to the critical ux level, and ASTM
E 1321 (LIFT apparatus) for lateral ame spread on verti-
cal specimens to the critical ux level. Heat transfer from
the ame to the virgin fuel is primarily conductive within a
spatial extent of a few millimeters and is affected by ambi-
ent conditions such as oxygen, pressure, buoyancy, and ex-
ternal irradiance. For most wood materials, this heat transfer
from the ame is less than or equal to surface radiant heat
loss in normal ambient conditions, so that excess heat is not
available to further raise the virgin fuel temperature; ame
spread is prevented as a result. Therefore, to achieve creep-
ing ame spread, an external heat source is required in the
vicinity of the pyrolysis front (Dietenberger 1994).
Upward or ceiling ame spread involves a fuel orientation
with the main air owing in the same direction as the ame
spread (assisting ow). Testing of ame spread in assisting
ow exists in both the tunnel tests and the room-corner burn
tests. The heat transfer from the ame is both conductive
and radiative, has a large spatial feature, and is relatively un-
affected by ambient conditions. Rapid acceleration in ame
spread can develop because of a large, increasing magnitude
of ame heat transfer as a result of increasing total HRR in
assisting ows (Dietenberger and White 2001). These com-
plexities and the importance of the ame spread processes
explain the many and often incompatible ame spread tests
and models in existence worldwide.
Charring and Fire Resistance
As noted earlier in this chapter, wood exposed to high tem-
peratures will decompose to provide an insulating layer of
char that retards further degradation of the wood (Figure
18–3). The load-carrying capacity of a structural wood
member depends upon its cross-sectional dimensions. Thus,
the amount of charring of the cross section is the major
factor in the re resistance of structural wood members.
When wood is rst exposed to re, the wood chars and
eventually ames. Ignition occurs in about 2 min under the
standard ASTM E 119 re-test exposures. Charring into the
depth of the wood then proceeds at a rate of approximately
0.8 mm min–1 for the next 8 min (or 1.25 min mm–1). There-
after, the char layer has an insulating effect, and the rate
decreases to 0.6 mm min–1 (1.6 min mm–1). Considering the
initial ignition delay, the fast initial charring, and then the
slowing down to a constant rate, the average constant char-
ring rate is about 0.6 mm min–1 (or 1.5 in. h–1) (Douglas-r,
7% moisture content). In the standard re resistance test,
this linear charring rate is generally assumed for solid wood
directly exposed to re. There are differences among species
associated with their density, anatomy, chemical composi-
tion, and permeability. In a study of the re resistance of
structural composite lumber products, the charring rates
of the products tested were similar to that of solid-sawn
lumber. Moisture content is a major factor affecting char-
ring rate. Density relates to the mass needed to be degraded
and the thermal properties, which are affected by anatomi-
cal features. Charring in the longitudinal grain direction
is reportedly double that in the transverse direction, and
chemical composition affects the relative thickness of the
char layer. Permeability affects movement of moisture be-
ing driven from the wood or that being driven into the wood
beneath the char layer. Normally, a simple linear model for
charring where t is time (min), C is char rate (min mm–1),
and xc is char depth (mm) is
(18–1)
The temperature at the base of the char layer is generally
taken to be 300 °C or 550 °F (288 °C). With this tempera-
ture criterion, empirical equations for charring rate have
18–13
Chapter 18 Fire Safety of Wood Construction
Figure 18–3. Illustration of charring of wood slab.
been developed. Equations relating charring rate under
ASTM E 119 re exposure to density and moisture content
are available for Douglas-r, Southern Pine, and white oak.
These equations for rates transverse to the grain are
C = (0.002269 + 0.00457m)r + 0.331 for Douglas-r
(18–2a)
C = (0.000461 + 0.00095m)r + 1.016 for Southern Pine
(18–2b)
C = (0.001583 + 0.00318m)r + 0.594 for white oak
(18–2c)
where m is moisture content (fraction of ovendry mass) and
r is density, dry mass volume at moisture content m (kg
m–3).
A nonlinear char rate model has been found useful. This al-
ternative model is
(18–3)
where m is char rate coefcient (min mm–1.23).
A form of Equation (18–3) is used in the NDS Method for
calculating the re resistance rating of an exposed wood
member. Based on data from eight species (Table 18–3), the
following equation was developed for the char rate
coefcient:
m = -0.147 + 0.000564r + 1.21m + 0.532fc (18–4)
where r is density, ovendry mass and volume, and fc is char
contraction factor (dimensionless).
The char contraction factor is the thickness of the residual
char layer divided by the original thickness of the wood
layer that was charred (char depth). Average values for the
eight species tested in the development of the equation are
listed in Table 18–3. These equations and data are valid
when the member is thick enough to be a semi-innite slab.
For smaller dimensions, the charring rate increases once the
temperature has risen above the initial temperature at the
center of the member or at the unexposed surface of the pan-
el. As a beam or column chars, the corners become rounded.
Charring rate is also affected by the severity of the re ex-
posure. Data on charring rates for re exposures other than
ASTM E 119 have been limited. Data for exposure to con-
stant temperatures of 538, 815, and 927 °C are available in
Schaffer (1967). Data for a constant heat ux are given in
Table 18–3.
The temperature at the innermost zone of the char layer is
assumed to be 300 °C. Because of the low thermal conduc-
tivity of wood, the temperature 6 mm inward from the base
of the char layer is about 180 °C. This steep temperature
18–14
General Technical Report FPLGTR190
Table 18
3. Charring rate data for selected wood species
Wood exposed to ASTM E 119 exposurea
Wood exposed to a constant heat flux
Linear charring ratee
(min mm–1)
Thermal penetration
depth dg
( mm)
Average mass
loss rate
(g m–2 s–1)
Species
Densityc
(kg m–3)
Char
con-
traction
factord
Linear
charring
ratee
(min
mm–1)
Non-
linear
charring
ratef
(min
mm–1.23)
Thermal
penetra-
tion
depthg
(mm)
18-
kW m–2
heat
flux
55-
kW m–2
heat
flux
18-
kW m–2
heat
flux
55-
kW m–2
heat
flux
18-
kW m–2
heat
flux
55-
kW m–2
heat
flux
Softwoods
Southern
Pine
509 0.60 1.24 0.56 33 2.27 1.17 38 26.5 3.8 8.6
Western
redcedar
310 0.83 1.22 0.56 33
Redwood 343 0.86 1.28 0.58 35 1.68 0.98 36.5 24.9 2.9 6.0
Engelmann
spruce
425 0.82 1.56 0.70 34
Hardwoods
Basswood 399 0.52 1.06 0.48 32 1.32 0.76 38.2 22.1 4.5 9.3
Maple, hard 691 0.59 1.46 0.66 31
Oak, red 664 0.70 1.59 0.72 32 2.56 1.38 27.7 27.0 4.1 9.6
Yellow-
poplar
504 0.67 1.36 0.61 32
aMoisture contents of 8% to 9%.
bCharring rate and average mass loss rate obtained using ASTM E 906 heat release apparatus. Test durations were 50 to 98 min for 18-kW m–2 heat
flux and 30 to 53 min for 55-kW m–2 heat flux. Charring rate based on temperature criterion of 300 °C and linear model. Mass loss rate based on
initial and final weight of sample, which includes moisture driven from the wood. Initial average moisture content of 8% to 9%.
cBased on weight and volume of ovendried wood.
dThickness of char layer at end of fire exposure divided by original thickness of charred wood layer (char depth).
eBased on temperature criterion of 288 °C and linear model.
fBased on temperature criterion of 288 °C and nonlinear model of Equation (18–3).
gAs defined in Equation (18–6). Not sensitive to moisture content.
gradient means the remaining uncharred cross-sectional area
of a large wood member remains at a low temperature and
can continue to carry a load. Once a quasi-steady-state char-
ring rate has been obtained, the temperature prole beneath
the char layer can be expressed as an exponential term or a
power term. An equation based on a power term is
( )
2
ii
1300
--+= d
x
TTT
(18–5)
where T is temperature (°C), Ti initial temperature (°C), x
distance from the char front (mm), and d thermal penetra-
tion depth (mm).
In Table 18–3, values for the thermal penetration depth pa-
rameter are listed for both the standard re exposure and the
constant heat ux exposure. As with the charring rate, these
temperature proles assume a semi-innite slab. The equa-
tion does not provide for the plateau in temperatures that
often occurs at 100 °C in moist wood. In addition to these
empirical data, there are mechanistic models for estimating
the charring rate and temperature proles. The temperature
prole within the remaining wood cross section can be used
with other data to estimate the remaining load-carrying ca-
pacity of the uncharred wood during a re and the residual
capacity after a re.
Fire-Retardant-Treated Wood
Wood products can be treated with re retardants to improve
their re performance. Fire-retardant treatments results
in delayed ignition, reduced heat release rate, and slower
spread of ames. HRRs are markedly reduced by re-re-
tardant treatment (Fig. 18–4). In terms of re performance,
re-retardant treatments are marketed to improve the ame
spread characteristics of the wood products as determined
by ASTM E 84, ASTM E 108, or other ammability tests.
Fire-retardant treatment also generally reduces the smoke-
developed index as determined by ASTM E 84. A re-
retardant treatment is not intended to affect re resistance of
wood products as determined by an ASTM E 119 test in any
consistent manner. Fire-retardant treatment does not make
a wood product noncombustible as determined by ASTM E
136 nor does it change its potential heat as determined by
NFPA 259.
Because re-retardant treatment does reduce the amma-
bility of the wood product, FRT wood products are often
used for interior nish and trim in rooms, auditoriums, and
corridors where codes require materials with low surface
ammability. Although FRT wood is not a noncombustible
material, many codes have specic exceptions that allow
the use of FRT wood and plywood in re-resistive and non-
combustible construction for framing of non-load-bearing
partitions, nonbearing exterior walls, and roof assemblies.
Fire-retardant-treated wood is also used for such special
purposes as wood scaffolding and for the frame, rails, and
stiles of wood re doors.
To meet specications in building codes and various stan-
dards, FRT lumber and plywood is wood that has been
pressure treated with chemicals to reduce its ame spread
characteristics. In the case of other composite wood prod-
ucts, chemicals can be added during the manufacture of the
wood product. Fire-retardant treatment of wood generally
improves the re performance by reducing the amount of
ammable volatiles released during re exposure or by
reducing the effective heat of combustion, or both. Both
results have the effect of reducing HRR, particularly during
the initial stages of re, and thus consequently reducing the
rate of ame spread over the surface. The wood may then
self-extinguish when the primary heat source is removed.
FRT products can be found in the Underwriters Laborato-
ries, Inc., “Building Materials Directory,” evaluation reports
of ICC Evaluation Service, Inc. (ICC–ES), and other such
listings.
Pressure Treatments
In impregnation treatments, wood is pressure impregnated
with chemical solutions using pressure processes similar to
those used for chemical preservative treatments. However,
considerably heavier absorptions of chemicals are necessary
for re-retardant protection. Penetration of chemicals into
the wood depends on species, wood structure, and moisture
content. Because some species are difcult to treat, the
degree of impregnation needed to meet the performance re-
quirements for FRT wood may not be possible.
Inorganic salts are the most commonly used re retardants
for interior wood products, and their characteristics have
been known for more than 50 years. These salts include
monoammonium and diammonium phosphate, ammonium
sulfate, zinc chloride, sodium tetraborate, and boric acid.
Guanylurea phosphate is also used. Chemicals are combined
in formulations to develop optimum re performance yet
still retain acceptable hygroscopicity, strength, corrosivity,
machinability, surface appearance, glueability, and
18–15
Chapter 18 Fire Safety of Wood Construction
0
50
100
150
200
250
100 200 300 400 500 600 700 800
Time (s)
Rate of heat release (kW m-2)
Untreated Douglas fir
plywood, 12.5 mm thick
FRT treated
Figure 18–4. Heat release curves for untreated and fire-
retardant-treated (FRT) Douglas-fir plywood, 12.5 mm
thick.
paintability. Cost is also a factor in these formulations. Ac-
tual formulations of commercial re-retardant treatments
are generally proprietary. For the two interior re-retardant
treatments listed in American Wood Protection Association
(AWPA) (formerly American Wood-Preservers’ Association)
standards, the chemicals listed are guanylurea phosphate
and boric acid for FR-1 and phosphate, boric acid, and am-
monia for FR-2. Species-specic information on the depth
of chemical penetration for these two formulations can be
found in Section 8.8 of AWPA Standard T1. Traditional re-
retardant salts are water soluble and are leached out in exte-
rior applications or with repeated washings. Water-insoluble
organic re retardants have been developed to meet the
need for leach-resistant systems. Such treatments are also
an alternative when a low-hygroscopic treatment is needed.
These water-insoluble systems include (a) resins polymer-
ized after impregnation into wood and (b) graft polymer
re retardants attached directly to cellulose. An amino resin
system based on urea, melamine, dicyandiamide, and related
compounds is of the rst type.
There are AWPA standards that describe methods for testing
wood for the presence of phosphate or boron. Such tests can
be used to determine the presence of re-retardant treat-
ments that contain these chemicals. AWPA Standard A9 is a
method for analysis of treated wood and treating solutions
by x-ray spectroscopy. The method detects the presence
of elements of atomic number 5 or higher including B(5)
and P(15). AWPA Standard A26 has a method for analysis
of re retardant FR1 solutions or wood by titration for the
percentages of boric acid and guanylurea phosphate. AWPA
Standard A3 describes methods for determining penetra-
tion of re retardants. Included are two methods for boron-
containing preservatives and re retardants and one method
for phosphorus-containing re retardants. The compositions
of commercial re-retardant treatments are proprietary. In
the case of boron, tests for its presence cannot distinguish
between treatments for preservation and those for re re-
tardancy. Such chemical tests are not an indicator of the
adequacy of the treatment in terms of re retardancy. Small-
scale re tests such as the cone calorimeter (ASTM E 1354),
oxygen index (ASTM D 2863), re tube (ASTM E 69), and
various thermal analysis methodologies can also be used to
determine the presence of re retardant treatment.
Performance Requirements
The IBC has prescriptive language specifying performance
requirements for FRT wood. The re performance require-
ment for FRT wood is that its FSI is 25 or less when tested
according to the ASTM E 84 ame spread test and that it
shows no evidence of signicant progressive combustion
when this 10-min test is continued for an additional
20 min. In addition, it is required that the ame front in the
test shall not progress more than 3.2 m beyond the center-
line of the burner at any given time during the test. In the
IBC, FRT wood must be a wood product impregnated with
chemicals by a pressure process or other means during
manufacture. In applications where the requirement being
addressed is not for “re-retardant-treated wood” but only
for Class A or B ame spread, the treatment only needs to
reduce the FSI to the required level in the ASTM E 84 ame
spread test (25 for Class A, 75 for Class B).
In addition to requirements for ame spread performance,
FRT wood for use in certain applications is required to meet
other performance requirements. Wood treated with inor-
ganic re-retardant salts is usually more hygroscopic than is
untreated wood, particularly at high relative humidities. In-
creases in equilibrium moisture content of this treated wood
will depend upon the type of chemical, level of chemical re-
tention, and size and species of wood involved. Applications
that involve high humidity will likely require wood with low
hygroscopicity. Requirements for low hygroscopicity in the
IBC stipulate that interior FRT wood shall have a moisture
content of not more than 28% when tested in accordance
with ASTM D 3201 procedures at 92% relative humidity.
Exterior re-retardant treatments should be specied when-
ever the wood is exposed to weather, damp, or wet condi-
tions. Exterior type treatment is one that has shown no in-
crease in the listed ame spread index after being subjected
to the rain test of ASTM D 2898. Although the method of
D 2898 is often not specied, the intended rain test is usu-
ally Method A of ASTM D 2898. Method B of D 2898 in-
cludes exposures to UV bulbs in addition to water sprays, is
described in FPL publications, and is an acceptable method
in AWPA Standard U1 for evaluating exterior treatments.
The ASTM D 2898 standard practice was recently revised to
include Methods C and D. Method C is the “amended rain
test” described in the acceptance criteria for classied wood
roof systems (AC107) of the ICC Evaluation Service, Inc.
Method D is the alternative rain test described in ASTM E
108 for roof coverings.
Fire-retardant treatment generally results in reductions in the
mechanical properties of wood. Fire-retardant-treated wood
is often more brash than untreated wood. For structural ap-
plications, information on mechanical properties of the FRT
wood product needs to be obtained from the treater or chem-
ical supplier. This includes the design modication factors
for initial strength properties of the FRT wood and values
for the fasteners. Adjustments to the design values must take
into account expected temperature and relative humidity
conditions. In eld applications with elevated temperatures,
such as roof sheathings, there is the potential for further
losses in strength with time. Fire-retardant-treated wood
that will be used in high-temperature applications, such as
roof framing and roof sheathing, is also strength tested in
accordance with ASTM D 5664 (lumber) or ASTM D 5516
(plywood) for purpose of obtaining adjustment factors as
described in ASTM D 6841 (lumber) and ASTM D 6305
(plywood). The temperatures used to obtain the adjustment
18–16
General Technical Report FPLGTR190
factors also become the maximum temperature that can be
used in kiln drying of lumber or plywood after treatment.
Corrosion of fasteners can be accelerated under conditions
of high humidity and in the presence of re-retardant salts.
For re-retardant treatments containing inorganic salts,
the types of metal and chemical in contact with each other
greatly affect the rate of corrosion. Thus, information on
proper fasteners also needs to be obtained from the treater or
chemical supplier. Other issues that may require contacting
the treater or chemical supplier include machinability, glu-
ing characteristics, and paintability.
Fire-retardant treatment of wood does not prevent the wood
from decomposing and charring under re exposure (the rate
of re penetration through treated wood approximates the
rate through untreated wood). Fire-retardant-treated wood
used in doors and walls can slightly improve re resistance
of these doors and walls. Most of this improvement is asso-
ciated with reduction in surface ammability rather than any
changes in charring rates.
There are specications for FRT wood issued by AWPA
and NFPA. In terms of performance requirements, these
specications are consistent with the language in the codes.
The AWPA standards C20 and C27 for FRT lumber and
plywood have recently been deleted by AWPA. They have
been replaced by AWPA “Use Category System Standards”
for specifying treated wood. The specic provisions are
Commodity H of Standard U1 and Section 8.8 of Standard
T1. The re protection categories are UCFA for interior
applications where the wood is protected from exterior
weather and UCFB for exterior applications where any
water is allowed to quickly drain from the surface. Neither
category is suitable for applications involving contact with
the ground or with foundations. Commodity Specication
H is re-retardant treatment by pressure processes of solid
sawn and plywood. The performance requirements for Com-
modity Specication H treatments are provided in Standard
U1. Section 8.8 of Standard T1 provides information on the
treatment and processing (that is, drying) of the products.
There is also NFPA standard 703 for FRT wood and re-
retardant coatings. In addition to the performance and
testing requirements for FRT wood products impregnated
with chemicals by a pressure process or other means during
manufacture, this NFPA standard provides separate speci-
cations for re-retardant coatings.
For parties interested in developing new re-retardant treat-
ments, there are documents that provide guidelines on the
data required for technical acceptance. In the AWPA Book
of Standards, there is “Appendix B: Guidelines for evaluat-
ing new re retardants for consideration by the AWPA.” The
ICC–ES has issued an “Acceptance criteria for re-retar-
dant-treated wood” (AC66), which provides guidelines for
what is required to be submitted for their evaluation reports.
There is also “Acceptance criteria for classied wood roof
systems” (AC107). Because of the relative small size of the
specimen, FPL uses the cone calorimeter in its research and
development of new FRT products.
Fire-Retardant Coatings
For some applications, applying the re-retardant chemical
as a coating to the wood surface may be acceptable to the
authorities having jurisdiction. Commercial coating prod-
ucts are available to reduce the surface ammability charac-
teristics of wood. The two types of coatings are intumescent
and nonintumescent. The widely used intumescent coatings
“intumesce” to form an expanded low-density lm upon
exposure to re. This multicellular carbonaceous lm insu-
lates the wood surface below from high temperatures. In-
tumescent formulations include a dehydrating agent, a char
former, and a blowing agent. Potential dehydrating agents
include polyammonium phosphate. Ingredients for the char
former include starch, glucose, and dipentaerythritol. Po-
tential blowing agents for the intumescent coatings include
urea, melamine, and chlorinate parans. Nonintumescent
coating products include formulations of the water-soluble
salts such as diammonium phosphate, ammonium sulfate,
and borax.
NFPA standard 703 includes specications for re-retardant
coatings. Because coatings are not pressure impregnated or
incorporated during manufacture, re-retardant coated wood
is not FRT wood as dened in most codes or standards in-
cluding NFPA 703. In NFPA 703, a re-retardant coating is
dened as a coating that reduces the ame spread of Doug-
las-r and all other tested combustible surfaces to which it
is applied by at least 50% or to a ame spread classication
value of 75 or less, whichever is the lesser value, and has a
smoke developed rating not exceeding 200 when tested in
accordance with ASTM E 84, NFPA 255, or UL 723. There
is no requirement that the standard test be extended for an
additional 20 min as required for FRT wood. NFPA 703 dif-
ferentiates between a Class A coating as one that reduces
ame spread index to 25 or less and a Class B coating as
one that reduces ame spread index to 75 or less.
Fire-retardant coatings for wood are tested and marketed to
reduce ame spread. Clear intumescent coatings are avail-
able. Such coatings allow the exposed appearance of old
structural wood members to be maintained while providing
improved re performance. This is often desirable in the
renovation of existing structures, particularly museums and
historic buildings. Studies have indicated that coatings sub-
jected to outdoor weathering are of limited durability and
would need to be reapplied on a regular basis.
Although their use to improve the resistance ratings of wood
products has been investigated, there is no general accep-
tance for using coatings to improve the re resistance rating
of a wood member. There is a lack of full-scale ASTM E
119 test data to demonstrate their performance and validate
a suitable calculation methodology for obtaining the rating.
18–17
Chapter 18 Fire Safety of Wood Construction
Literature Cited
Atreya, A. 1983. Pyrolysis, ignition, and ame spread on
horizontal surfaces of wood. Cambridge, MA: Harvard
University. Ph.D. dissertation.
Babrauskas, V. 2003. Ignition Handbook. Issaquah, WA:
Fire Science Publishers.
Benjamin, I.A.; Davis, S. 1979. Flammability testing for
carpets. Fire Technology. 15(3): 189–194.
Browne, F.L. 1958. Theories of the combustion of wood
and its control—a survey of the literature. Rep. No. 2136.
Madison, WI: U.S. Department of Agriculture, Forest
Service, Forest Products Laboratory.
Bryan, K.M. 1998. Computational modeling of wood
combustion. Madison, WI: University of Wisconsin–
Madison. Ph.D. dissertation.
Dietenberger, M.A. 1994. Protocol for ignitability, lateral
ame spread, and heat release rate using LIFT apparatus.
In: Nelson, G.L., ed. Fire and polymers II. Materials and
tests for hazard prevention: Proceedings of 208th national
meeting of the American Chemical Society; 1994 August
21–26; Washington, DC. ACS symposium series 599.
Washington, DC: American Chemical Society, 1995.
Chapter 29.
Dietenberger, M.A. 1996. Ignitability analysis using the
cone calorimeter and LIFT apparatus. Proceedings of
the international conference on re safety, Vol. 22; 1996
July 22–26. Columbus OH: Product Safety Corporation.
Sissonville, WV: 189–197.
Dietenberger, M.A. 2002. Update for combustion properties
of wood components. Fire and Materials Journal. (26):
255–267.
Dietenberger, M.A. 2004. Ignitability of materials
in transitional heating regimes. Proceedings of 5th
international scientic conference on wood & re safety;
2004 April 18–22. Slovak Republic. Zvolen, Slovakia:
Faculty of Wood Sciences and Technology, Technical
University of Zvolen, 2004: 31–41.
Dietenberger, M.A.; Grexa, O. 2000. Correlation of smoke
development in room tests with cone calorimeter data for
wood products. Proceedings of 4th international scientic
conference on wood & re safety; 2000 May 14–19. Slovak
Republic: The High Tatras, Hotel Patria, Strbske Pleso.
Dietenberger, M.A.; White, R.H. 2001. Reaction-to-re
testing and modeling for wood products. In: Proceedings
of 12th annual BCC conference on ame retardancy;
2001 May 21–23. Stanford, CT. Norwalk, CT: Business
Communications Co., Inc. 54–69.
Östman, B.A.-L.; Mikkola, E. 2006. European classes for
the reaction to re performance of wood products. Holz als
Roh- und Werkstoff. 64: 327–337.
Schaffer, E.L. 1967. Charring rate of selected woods-
transverse to grain. Res. Pap. FPL 69. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory.
Shazadeh, F. 1984. The chemistry of pyrolysis and
combustion. In: Rowell, R.M., ed. The chemistry of solid
wood, advances in chemistry series 207. Washington, DC:
American Chemical Society. 489–530.
Stamm, A.J. 1955. Thermal degradation of wood and
cellulose. Presented at the symposium on degradation
of cellulose and cellulose derivatives. Sponsored by the
Division of Cellulose Chemistry, 127th national meeting
of the American Chemical Society. Cincinnati, OH.
(April 4–7).
Tang, W.K. 1967. Effect of inorganic salts on pyrolysis of
wood, alpha-cellulose, and lignin determined by dynamic
thermogravimetry. Res. Pap. FPL 71. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory.
Tsantaridis, L.; Östman, B. 2004. European classes for
the reaction to re performance of wood oorings. Trätek
Rapport 0411026. Stockholm: Institute för Träteknisk
Forskning.
White, R.H. 2000. Fire performance of hardwood species.
NTIS #PB2001-107912WBT. In: XXI IUFRO World
Congress, Kuala Lumpur, Malaysia; 2000 August 7–12.
White, R.H.; Dietenberger, M.A.; Tran, H.; Grexa, O.;
Richardson, L.; Sumathipala, K.; Janssens, M. 1999.
Comparison of test protocols for the standard room/corner
test. Fire and Materials. 23: 139–46.
Additional References
General
APA. [Current edition]. Fire-rated systems. Tacoma, WA:
APA–The Engineered Wood Association. www.apawood.
org.
CWC. 1996. Fire safety design in buildings. Ottawa, ON,
Canada: Canadian Wood Council. www.cwc.ca.
ICC. [2006 or current edition]. International building code.
Country Club Hills, IL: International Code Council, Inc.
www.iccsafe.org.
International wildland-urban interface code
International re code
International residential code for one- and two-family
dwellings
International Code Council performance code for
buildings and facilities
NFPA. [Current edition]. Fire protection handbook. Quincy,
MA: National Fire Protection Association. www.nfpa.org.
18–18
General Technical Report FPLGTR190
NFPA 1144. Standard for reducing structure ignition
hazards from wildland re.
NFPA 101. Life safety code.
Society of Fire Protection Engineers. [Current edition].
The Society of Fire Protection Engineers handbook of
re protection engineering. Quincy, MA: National Fire
Protection Association. www.nfpa.org.
Fire Test Standards
ASTM. [Current edition]. West Conshohocken, PA: ASTM
International. www.astm.org.
ASTM D 2863. Measuring the minimum oxygen
concentration to support candle-like combustion of
plastics (oxygen index).
ASTM D 2898. Accelerated weathering of re-retardant
treated wood for re testing.
ASTM D 6513. Calculating the superimposed load on
wood-frame walls for standard re-resistance tests.
ASTM E 69. Combustible properties of treated wood
by re-tube apparatus.
ASTM E 84. Surface burning characteristics of building
materials.
ASTM E 108. Fire tests of roof coverings.
ASTM E 119. Fire tests of building construction and
materials.
ASTM E 136. Behavior of materials in a vertical tube
furnace at 750 °C.
ASTM E 162. Surface ammability of materials using a
radiant heat energy source.
ASTM E 648. Critical radiant ux of oor-covering
systems using a radiant heat energy source.
ASTM E 662. Specic optical density of smoke
generated by solid materials.
ASTM E 814. Fire tests of through-penetration re
stops.
ASTM E 906. Heat and visible smoke release rates
from materials and products.
ASTM E 970. Critical radiant ux of exposed attic oor
insulation using a radiant heat energy source.
ASTM E 1321. Determining material ignition and ame
spread properties.
ASTM E 1354. Heat and visible smoke release rates for
materials and products using an oxygen consumption
calorimeter.
ASTM E 1623. Determination of re and thermal
parameters of materials, products, and systems using an
intermediate scale calorimeter (ICAL).
ASTM E 1678. Measuring smoke toxicity for use in re
hazard analysis.
ASTM E 2257. Room re test of wall and ceiling
materials and assemblies.
ASTM E 2579. Specimen preparation and mounting of
wood products to assess surface burning characteristics.
NFPA. [Current edition]. Quincy, MA: National Fire
Protection Association. www.nfpa.org.
NFPA 251. Fire resistance of building construction and
materials.
NFPA 253. Critical radiant ux of oor covering
systems using a radiant heat energy source.
NFPA 255. Surface burning characteristics of building
materials.
NFPA 259. Potential heat of building materials.
NFPA 286. Evaluating contribution of wall and ceiling
interior nish to room re growth.
UL. [Current edition]. Northbrook, IL: Underwriters
Laboratories, Inc. www.ul.com.
ANSI/UL 723. Surface burning characteristics of
building materials.
Ignition
Atreya, A. 1998. Ignition of res. London, England:
Philosophical Transactions of the Royal Society of London.
(356): 2787–2813.
Brenden, J.J.; Schaffer, E. 1980. Smoldering wave-front
velocity in berboard. Res. Pap. FPL 367. Madison, WI:
U.S. Department of Agriculture, Forest Service, Forest
Products Laboratory.
Dietenberger, M.A. 1994. Ignitability analysis of siding
materials using modied protocol for LIFT apparatus.
In: Proceedings, 3rd re and materials conference; 1994
October 27–28; Crystal City, VA. London: Interscience
Communications Limited: 259–268.
Kubler, H. 1990. Self-heating of lignocellulosic materials.
In: Nelson, G.L., ed. Fire and polymers—hazards
identication and prevention. ACS symposium series 425.
Washington, DC: American Chemical Society: 429–449.
LeVan, S.; Schaffer, E. 1982. Predicting weight loss from
smoldering combustion in cellulosic insulation. Journal of
Thermal Insulation. 5: 229–244.
Matson, A.F.; Dufour, R.E.; Breen, J.F. 1959. Part II. Survey
of available information on ignition of wood exposed to
moderately elevated temperatures. In: Performance of
type B gas vents for gas-red appliances. Bull. of Res. 51.
Chicago, IL: Underwriters Laboratories: 269–295.
18–19
Chapter 18 Fire Safety of Wood Construction
Schaffer, E.L. 1980. Smoldering in cellulosics under
prolonged low-level heating. Fire Technology. 16(1): 22–28.
Shazadeh, F.; Sekiguchi, Y. 1984. Oxidation of chars
during smoldering combustion of cellulosic materials.
Combustion and Flame. 55: 171–179.
Flame Spread
AWC. [Current edition]. Flame spread performance of wood
products. Design for code acceptance No. 1. www.awc.org.
Bengelsdorf, M.F. [Current edition]. Fire-hazard
classication of construction plywood panels. Report 128.
Tacoma, WA: APA–The Engineered Wood Association.
www.apawood.org.
Gardner, W.D.; Thomson, C.R. 1988. Flame spread
properties of forest products. Comparison and validation of
prescribed Australian and North American ame spread test
methods. Fire and Materials. 12: 71–85.
Holmes, C.A.; Eickner, H.W.; Brenden, J.J.; White, R.H.
1979. Fire performance of structural akeboard from
forest residue. Res. Pap. FPL–RP–315. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory.
Sumathipala, K.; White, R.H. 2006. Flammability tests
for regulation of building and construction materials. In:
Vivek, B.A., ed. Flammability testing of materials used in
construction, transport and mining. Cambridge, England:
Woodhead Publishing, Ltd. 217–30. Chapter 10.
UL. 1971. Wood-re hazard classication. Card Data
Service, Serial No. UL527. Chicago, IL: Underwriters
Laboratories, Inc.
UL. [Current edition]. Building materials directory.
Northbrook, IL: Underwriters Laboratories, Inc.
www.ul.com.
Yarbrough, D.W.; Wilkes, K.E. 1997. Thermal properties
and use of cellulosic insulation produced from recycled
paper. In: The use of recycled wood and paper in building
applications. Proc. 7286. Madison, WI: Forest Products
Society: 108–114.
Flashover and Room/Corner Tests
Bruce, H.D. 1959. Experimental dwelling—room res. Rep.
1941. Madison, WI: U.S. Department of Agriculture, Forest
Service, Forest Products Laboratory.
Dietenberger, M.A.; Grexa, O.; White, R.H. [and others].
1995. Room/corner test of wall linings with 100/300 kW
burner. In: Proceedings, 4th international re and materials
conference; 1995 November 15–16. Crystal City, MD.
London: InterScience Communications Limited: 53–62.
Holmes, C. 1978. Room corner-wall re tests of some
structural sandwich panels and components. Journal of Fire
& Flammability. 9: 467–488.
Holmes, C.; Eickner, H.; Brenden, J.J. [and others]. 1980.
Fire development and wall endurance in sandwich and
wood-frame structures. Res. Pap. FPL 364. Madison, WI:
U.S. Department of Agriculture, Forest Service, Forest
Products Laboratory.
Schaffer, E.L.; Eickner, H.W. 1965. Effect of wall linings on
re performance within a partially ventilated corridor. Res.
Pap. FPL 49. Madison, WI: U.S. Department of Agriculture,
Forest Service, Forest Products Laboratory.
Tran, H.C. 1991. Wall and corner re tests on selected wood
products. Journal of Fire Sciences. 9: 106–124. March/
April.
Heat Release and Heat of Combustion
Babrauskas, V.; Grayson, S.J., eds. 1992. Heat release in
res. New York: Elsevier Applied Science.
Chamberlain, D.L. 1983. Heat release rates of lumber
and wood products. In: Schaffer, E.L., ed. Behavior of
polymeric materials in re. ASTM STP 816. Philadelphia,
PA: American Society for Testing and Materials: 21–41.
Hagge, M.J.; Bryden, K.M.; Dietenberger, M.A. 2004.
Effect of backing board materials on wood combustion
performance. In: Proceedings, 5th international scientic
conference on wood & re safety; 2004 April 18–22.
Tran, H.C. 1990. Modications to an Ohio State University
apparatus and comparison with cone calorimeter results. In:
Quintiere, J.G.; Cooper, L.Y., eds. Heat and mass transfer
in res. Proceedings, AIAA/ASME thermophysics and heat
transfer conference; 1990 June 18–20; Seattle, WA. New
York: The American Society of Mechanical Engineers.
(141): 131–139.
Tran, H.C. 1992. (B) Experimental data on wood materials.
In: Babrauskas, V.; Grayson, S.J., eds. Heat release in res.
New York: Elsevier Applied Science: 357–372. Chapter 11,
Part B.
White, R.H. 1987. Effect of lignin content and extractives
on the higher heating value of wood. Wood and Fiber
Science. 19(4): 446–452.
White, R.H.; Dietenberger, M.A. 2004. Cone calorimeter
evaluation of wood products. In: 15th annual conference on
recent advances in ame retardancy of polymeric materials.
Stamford, CT. Norwalk, CT: Business Communications
Company, Inc.: 331–342.
White, R.H.; Dietenberger, M.A.; Stark, N. 2007. Cone
calorimeter tests of wood-based decking materials. In:
Proceedings, 18th annual conference on recent advances
in ame retardancy of polymeric materials. Stamford, CT.
Norwalk, CT: BCC Research. (2): 326–337.
Combustion Products
Brenden, J.J. 1970. Determining the utility of a new
optical test procedure for measuring smoke from various
18–20
General Technical Report FPLGTR190
wood products. Res. Pap. FPL 137. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory.
Brenden, J.J. 1975. How nine inorganic salts affected
smoke yield from Douglas-r plywood. Res. Pap. FPL
249. Madison, WI: U.S. Department of Agriculture, Forest
Service, Forest Products Laboratory.
Hall, J.R., Jr. 1996. Whatever happened to combustion
toxicity. Fire Technology. 32(4): 351–371.
Tran, H.C. 1990. Correlation of wood smoke produced from
NBS smoke chamber and OSU heat release apparatus. In:
Hasegawa, H.K., ed. Characterization and toxicity of smoke.
ASTM STP 1082. Philadelphia, PA: American Society for
Testing and Materials: 135–146.
Fire Resistance
AITC. [Current edition]. Calculation of re resistance of
glued laminated timbers. Tech. Note 7. Englewood, CO:
American Institute of Timber Construction. www.aitc-
glulam.org.
APA. [Current edition]. Calculating re resistance of glulam
beams and columns. Publ. No. EWS Y245A. Tacoma, WA:
APA–The Engineered Wood Association. www.apawood.
org.
ASCE. 1982. Evaluation, maintenance and upgrading of
wood structures—a guide and commentary. New York:
American Society of Civil Engineers. 428 p.
AWC. [Current edition]. Calculating the re resistance of
exposed wood members. Tech. Rep. 10. Washington, DC:
American Forest & Paper Association—American Wood
Council. www.awc.org.
AWC. [Current edition]. Chapter M16: Fire design. In:
ASD/LRFD manual for engineered wood construction.
Washington, DC: American Forest & Paper Association—
American Wood Council. www.awc.org.
AWC. [Current edition]. Component additive method
(CAM) for calculating and demonstrating assembly re
endurance. Design for code acceptance No. 4. Washington,
DC: American Forest & Paper Association—American
Wood Council. www.awc.org.
AWC. [Current edition]. Design of re-resistive exposed
wood members. Design for code acceptance No. 2.
Washington, DC: American Forest & Paper Association—
American Wood Council. www.awc.org.
AWC. [Current edition]. Fire rated wood oor and wall
assemblies. Design for code acceptance No. 3. Washington,
DC: American Forest & Paper Association—American
Wood Council. www.awc.org.
Gypsum Association. [Current edition]. Fire resistance
design manual. Washington, DC: Gypsum Association.
www.gypsum.org.
HUD. 1980. Guideline on re ratings of archaic materials
and assemblies. Rehabilitation guidelines, part 8.
Washington, DC: U.S. Department of Housing and Urban
Development, Superintendent of Documents. http://www.
huduser.org/Publications/PDF/re.pdf.
Intertek. [Current edition]. Directory of listed product.
www.intertek-etlsemko.com.
Janssens, M. 1994. Thermo-physical properties for
wood pyrolysis models. In: Proceedings, Pacic timber
engineering conference; 1994 July 11–15; Gold Coast,
Australia. Fortitude Valley MAC, Queensland, Australia:
Timber Research and Development Advisory Council:
607–618.
Janssens, M.L.; White, R.H. 1994. Short communication:
temperature proles in wood members exposed to re. Fire
and Materials. 18: 263–265.
Ross, R.J.; Brashaw, B.K.; Wang, X.; White, R.H.; Pellerin,
R.F. 2004. Wood and timber condition assessment manual.
Madison, WI: Forest Products Society.
Schaffer, E.L. 1966. Review of information related to the
charring rate of wood. Res. Note FPL–145. Madison, WI:
U.S. Department of Agriculture, Forest Service, Forest
Products Laboratory.
Schaffer, E.L. 1977. State of structural timber re
endurance. Wood and Fiber. 9(2): 145–170.
Schaffer, E.L. 1984. Structural re design: Wood. Res. Pap.
FPL 450. Madison, WI: U.S. Department of Agriculture,
Forest Service, Forest Products Laboratory.
Tran, H.C.; White, R.H. 1992. Burning rate of solid wood
measured in a heat release calorimeter. Fire and Materials.
16: 197–206.
UL. [Current edition]. Fire resistance directory. Northbrook,
IL: Underwriters Laboratories, Inc. www.ul.com.
White, R.H. 2003. Fire resistance of engineered wood rim
board products. FPL Res. Pap. 610. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory. 22 p.
White, R.H. 2004. Fire resistance of exposed wood
members. In: Proceedings, 5th international wood & re
safety conference. Zvolen, Slovak Republic: Technical
University of Zvolen, Faculty of Wood Science and
Technology. 337–44.
White, R.H. 2006. Fire resistance of structural composite
lumber products, Res. Pap. FPL–RP–633. Madison, WI:
U.S. Department of Agriculture, Forest Service, Forest
Products Laboratory. 28 p.
White, R.H. 2008. Analytical methods for determining re
resistance of timber members. In: SFPE handbook of re
protection engineering. 4th ed. Quincy, MA: National Fire
Protection Association.
18–21
Chapter 18 Fire Safety of Wood Construction
White, R.H.; Nordheim, E.V. 1992. Charring rate of wood
for ASTM E 119 exposure. Fire Technology. 28: 5–30.
White, R.H.; Tran, H.C. 1996. Charring rate of wood
exposed to a constant heat ux. In: Proceedings, wood &
re safety, 3rd international conference. The High Tatras;
1996 May 6–9; Zvolen, Slovakia. Zvolen, Slovakia: Faculty
of Wood Technology, Technical University: 175–183.
Fire-Retardant-Treated Wood
ASTM. [Current edition]. West Conshohocken, PA: ASTM
International. www.astm.org.
ASTM D 3201. Hygroscopic properties of re-retardant
wood and wood-base products.
ASTM D 5516. Evaluating the mechanical properties
of re-retardant treated softwood plywood exposed to
elevated temperatures.
ASTM D 5664. Evaluating the effects of re-retardant
treatments and elevated temperatures on strength
properties of re-retardant treated lumber.
ASTM D 6305. Calculating bending strength design
adjustment factors for re-retardant-treated plywood
roof sheathing.
ASTM D 6841. Calculating design value treatment
adjustment factors for re-retardant-treated lumber.
AWPA. [Current edition]. Gransbury, TX: American
Wood Protection Association (formerly American Wood-
Preservers’ Association). www.awpa.com.
Standard A2. Analysis of waterborne preservatives and
re-retardant formulations.
Standard A3. Determining penetration of preservative
and re retardants.
Standard A9. Analysis of treated wood and treating
solutions by X-ray spectroscopy.
Standard A26. Analysis of re retardant solutions and
wood by titration.
Standard T1. Use category system: processing and
treatment standard.
Standard U1. Use category system: user specication
for treated wood.
Holmes, C.A. 1977. Effect of re-retardant treatments on
performance properties of wood. In: Goldstein, I.S., ed.
Wood technology: chemical aspects. Proceedings, ACS
symposium series 43. Washington, DC: American Chemical
Society.
ICC-ES. [Current edition]. Whittier, CA: ICC Evaluation
Service, Inc. www.icc-es.org.
AC66. Acceptance criteria for re-retardant-treated
wood.
18–22
General Technical Report FPLGTR190
AC107. Acceptance criteria for classied wood roof
systems.
Lebow, P.K., Winandy, J.E. 2003. Using kinetic models
to predict thermal degradation of re-retardant-treated
plywood roof sheathing. Paper 048. In: Proceedings, 31st
annual conference of North American Thermal Analysis
Society; 2003 September 22–24; Albuquerque, NM.
LeVan, S.L. 1984. Chemistry of re retardancy. In: Rowell,
R.M., ed. The chemistry of solid wood. Advances in
Chemistry series 207. Washington, DC: American Chemical
Society.
LeVan, S.L.; Tran, H.C. 1990. The role of boron in ame-
retardant treatments. In: Hamel, M., ed. Proceedings 47355,
1st international conference on wood protection with
diffusible preservatives; 1990 November 28–30; Nashville,
TN. Madison, WI: Forest Products Research Society: 39–41.
LeVan, S.L.; Winandy, J.E. 1990. Effects of re-retardant
treatments on wood strength: a review. Wood and Fiber
Science. 22(1): 113–131.
LeVan, S.L.; Ross, R.J.; Winandy, J.E. 1990. Effects of re
retardant chemicals on the bending properties of wood at
elevated temperatures. Res. Pap. FPL–RP–498. Madison,
WI: U.S. Department of Agriculture, Forest Service, Forest
Products Laboratory.
NAHB. 1990. Home builders guide to re retardant treated
plywood. Evaluation, testing, and replacement. Upper
Marlboro, MD: National Association of Home Builders,
National Research Center.
NFPA. [Current edition]. Quincy, MA: National Fire
Protection Association. www.nfpa.org.
NFPA 703. Fire retardant-treated wood and re-
retardant coatings for building materials.
Winandy, J.E. 1995. Effects of re retardant treatments
after 18 months of exposure at 150 °F (66 °C). Res.
Note FPL–RN–0264. Madison, WI: U.S. Department of
Agriculture, Forest Service, Forest Products Laboratory.
Winandy, J.E.; LeVan, S.L.; Ross, R.J.; Hoffman, S.P.;
McIntyre, C.R. 1991. Thermal degradation of re-retardant-
treated plywood—development and evaluation of a test
protocol. Res. Pap. FPL–RP–501. Madison, WI: U.S.
Department of Agriculture, Forest Service, Forest Products
Laboratory.
... Lignin, on the other hand, has the least stability and degrades at a lower temperature, where its decomposition results in the emission of combustible, volatile chemicals that may ignite and spread a fire ( Dorez et al., 2014 ;Mauranen et al., 2015 ;Gan et al., 2019 ). In addition to cellulose and lignin, hemicellulose, which is present between them, also affects how wood ignites, but to a lesser degree ( Dietenberger and White, 2010 ;Dorez et al., 2014 ). Additionally, variations in cellulose, hemicellulose, and lignin composition and ratios can occur between different wood species and even within the same tree, which can affect the characteristics of how quickly a fire starts, spreads, and how severe it is overall ( Poletto et al., 2012 ;Fayçal et al., 2022 ). ...
... Depending on the circumstances, the combustion of wood is a highly dynamic process involving various physical and chemical processes. The behaviour of fire is influenced by several variables, including temperature, oxygen concentration, moisture content, and others, which can have an impact on the fire intensity, duration, and spread ( Dietenberger and White, 2010 ;Martinka et al., 2012 ;Wang et al., 2017b ;Jayasuriya et al., 2022 ). The differences in properties between wood and wood-based composites, e.g., various thermal response, mechanical strength, and chemical properties, can influence how they respond to fire ( Bo ž iková et al., 2021 ;Kristak et al., 2021 ). ...
... Moreover, using untreated wood in construction can raise the fire risk since it is more prone to burning and decomposing ( Popescu and Pfriem, 2020 ;Gazizov et al., 2022 ). Wood fires cause an enormous financial risk, whereby restoring or replacing destroyed buildings can be expensive ( Dietenberger and White, 2010 ;Mustafa, 2020 ;Vicente et al., 2020 ;Gazizov et al., 2022 ). In addition to the direct expenses of repairing the fire's damage, there can also be indirect costs from lost productivity, disruption of corporate operations, and eviction of fire-affected individuals or groups ( Henry et al., 2018 ;Walls et al., 2020 ). ...
Article
Full-text available
Due to their durability, versatility, and aesthetic value, wood and wood-based composites are widely used as building materials. The fact that these materials are flammable, however, raises a major worry since they might cause fire hazards and significant loss of life and property. The article investigates the variables that affect fire performance as well as the various fire-retardant treatments and their mechanisms. The current developments and challenges in improving the fire performance of wood and wood-based composites treated with fire-retardant materials are summarized in this paper. Nanoparticles, organic chemicals, and densification are some recent developments in fire-retardant treatments that are also emphasized. Key points from the review are summarized, along with potential areas for further research and development.
... Wood is widely available, durable, and easy to process material. However, it is also highly combustible, leading to major fire safety concerns notably in terms of flame spread (R. H. White & Dietenberger, 2010a). Regulations limit the use of wood interior finishing in critical areas such as emergency exits, corridors and stairways in high-rise buildings (Canadian Wood Council, 1996). ...
Thesis
Full-text available
Wood is a natural composite material used in interior finishing as flooring. It is appreciated for its appearance, availability, and low environmental impact. However, its use is limited in non-residential construction because of the risk of fire propagation. Fireproofing of wood considers all treatments applied to wood to make it less combustible. Traditional approaches to fireproof wood, such as impregnation, are fossil fuel, energy, and time consuming. Surface treatment approaches have been proposed for textiles and have shown very promising results limiting the amount of used chemicals and thus its impact on the environment. Indeed, surface treatments aim at concentrating the fireproofing action on the surface exposed to the fire. In this project, two surface treatments were studied. First, a new method for the deposition of polyelectrolyte complexes was developed using surface impregnation at reduced pressure. The performance of a polyelectrolyte deposit was studied on the freeze-dried polyelectrolyte complexes. This approach allowed us to highlight the effect of the ratio between two polyelectrolytes on the fire performance of yellow birch (Betula alleghaniensis, Britt). Mass gain was identified as a limiting factor to improve the fire performance and several approaches were studied to increase it either by activating the wood surface by delignification or by increasing the wettability of the solution by adding wetting agents. Nanoparticles have also been added to the formulation, but no improvement of the fire performance was noticed. As a second approach, surface treatment by atmospheric jet plasma deposition has been studied. Several precursors were deposited on sugar maple (Acer saccharum, Marsh.) virgin or pretreated with a photopolymerized primer. This comparison highlighted the importance of the preparation method of the substrate in fire performance. Better performance was obtained on samples pretreated with a light-cured primer since in that case a homogenous deposit was obtained and could act as a fire protective barrier.
... A higher HRR can trigger rapid fire escalation, increased temperatures, and faster exposure of structural elements to intense heat. Consequently, this accelerated exposure may cause structural components to reach critical temperatures sooner than they would in scenarios with lower heat release rates [23]. ...
Article
Full-text available
This study delved into the combustion properties of combined glulam bonded using polyurethane (PUR) and resorcinol-phenol-formaldehyde (RPF) adhesives. The experiment involved three distinct wood species, namely, spruce, alder, and beech, which were combined in homogeneous, non-homogeneous symmetrical, and non-homogeneous asymmetrical arrangements. These species were selected to represent a spectrum, namely, softwood (spruce), low-density hardwood (alder), and high-density hardwood (beech). The varying combinations of wood species illustrate potential compositions within structural elements, aiming to optimize mechanical bending resistance. Various parameters were measured during combustion, namely, the heat release rate (HRR), peak heat release rate (pHRR), mass loss rate (MLR), average rate of heat emission (ARHE), peak average rate of heat emission (MARHE), time to ignition (TTI), and effective heat of combustion (EHC). The findings indicate that incorporating beech wood into the composite glulam resulted in an increase in heat release, significantly altering the burning characteristics, which was particularly evident at the second peak. Conversely, the use of spruce wood exhibited the lowest heat release rate. Alder wood, when subjected to heat flux at the glued joint, displayed the highest heat emission, aligning with the results for EHC and MARHE. This observation suggests that wood species prone to early thermal decomposition emit more heat within a shorter duration. The time to ignition (TTI) was consistent, occurring between the first and second minute across all tested wood species and combinations. Notably, when subjected to heat flux, the glulam samples bonded with PUR adhesive experienced complete delamination of the initial two glued joints, whereas those bonded with RPF adhesive exhibited only partial delamination.
... This slight increase in modulus may be due to the coating increasing the density of the wood by filling some pores [58]. This is a promising result as flame retardant treatments generally impair the mechanical properties of wood [59]. In a practical application, the treatment would likely be applied to much larger pieces of wood, significantly reducing the surface area-to-volume ratio. ...
Article
Full-text available
Biobased and biodegradable polymers represent a valid and sustainable alternative to oil-based plastics, as they are renewable and address the issue related to the end-of-life of non-compostable materials. However, the poor gas barrier of biopolymers limits their use in several applications, including food packaging. In this work, chitosan/graphene oxide (CS/GO) nanocomposite coatings were successfully deposited by ultrasonic spray on a compostable polybutylene succinate (PBS) film. The moisture resistance of the chitosan coatings was improved by crosslinking with polyethyleneglycol diglycidyl ether (PEGDE). The resulting coatings were transparent, with thickness in the 1–2.5 μm range, and exhibited good adhesion to the PBS film and mechanical and scratch resistance due to the presence of GO nanofiller. In detail, the PEGDE-crosslinked CS/GO (CS/PEGDE/GO) nanocomposite coating containing 1 wt% GO allowed to reduce O2 and CO2 transmission rates by 85 % and 93 %, respectively, compared to uncoated PBS film. The permeability reduction is ascribed to the formation of compact coatings with GO nanoplates oriented parallel to the PBS substrate. Furthermore, the improvement in CO2 barrier properties was up two-time more than that related to oxygen, suggesting the use of CS/PEGDE/GO coatings in applications where gas permselectivity is required. This research demonstrates the potential of the ultrasonic spray technique for producing bionanocomposite barrier coatings with improved gas barrier performance.
... Wood impregnated with various inorganic salts is usually more hygroscopic than untreated wood, especially at high RH. The increase in EMC of such wood depends on the chemicals used, retention, and wood species (White & Dietenberger, 2010). The EMC of impregnated wood and the effects of preservative retention on the equilibrium point are still unknown. ...
Article
Full-text available
The sorption properties of wood have a characteristic influence on some of its properties, such as the mechanical properties and susceptibility to fungal decay. Moist wood is more susceptible to fungal decay, and wood is often impregnated in order to protect it from fungal decomposition, photodegradation or fire. In particular, inorganic salts affect the sorption properties of wood. For this purpose, the sorption properties of Norway spruce wood impregnated with Burnblock refractory (uptake 38 kg/m³) were investigated. The microscopic analysis confirmed the presence of crystals of this in the cell lumina of wood tissue. Sorption properties were determined using an instrument capable of dynamic vapour sorption (DVS) assessment. DVS analysis confirmed that the sorption properties of impregnated spruce wood are comparable to those of non-impregnated spruce wood. However, the higher hysteresis at higher relative humidity is probably due to the presence of crystals in the cell lumina.
Article
Full-text available
Research in historical timber assessment is hindered by the limited availability of samples, yet understanding the fire resistance of historic wood is crucial for preservation efforts. There is an opinion that historic wood behaves similarly to contemporary wood in terms of fire resistance. The aim of this paper is to observe the rate of charring of historical pine wood during the experiment, the color changes in the sample that occurred due to thermal loading, and the changes in the chemical composition of pine wood. Test samples made from historic pine wood were loaded with a 50 kW∙m−2 radiation panel for 60 min. The charring process was faster at the beginning of the charred layer formation. The charring rate at the beginning of the test at a depth of 10 mm from the exposed side reached values from 1.28 mm∙min−1 to 3.16 mm∙min−1. At a depth of 30 mm from the exposed side, the individual charring rates approached a value of 1 mm∙min−1 (0.99 mm∙min−1 to 1.08 mm∙min−1). Observations during medium-scale testing revealed distinct layers forming on the exposed side: a charred layer, charring base, pyrolysis layer, and intact wood. The chemical composition of the wood changed under the influence of the thermal load. The relative contents of extractives and holocellulose decreased with the increasing temperature while the lignin content increased. The highest value of combustion heat was measured in the charred layer of the sample. Correlation analysis demonstrated a negative relationship between the combustion heat and holocellulose, while a positive correlation was found with the lignin content. Chemical changes were also monitored using the FTIR method. These findings provide valuable insights into the behavior of historic pine wood under thermal loading, which is essential for understanding and preserving historical structures.
Article
Application of nanomaterials is gaining tremendous interest in the field of wood science and technology for value addition and enhancing performance of wood and wood-based composites. This review focuses on the use of nanomaterials in improving the properties of wood and wood-based materials and protecting them from weathering, biodegradation, and other deteriorating agents. UV-resistant, self-cleaning (superhydrophobic) surfaces with anti-microbial properties have been developed using the extraordinary features of nanomaterials. Scratch-resistant nano-coatings also improve durability and aesthetic appeal of wood. Moreover, nanomaterials have been used as wood preservatives for increasing the resistance against wood deteriorating agents such as fungi, termites and borers. Wood can be made more resistant to ignition and slower to burn by introducing nano- clays or nanoparticles of metal-oxides. The use of nanocellulose and lignin nanoparticles in wood-based products has attracted huge interest in developing novel materials with improved properties. Nanocellulose and lignin nanoparticles derived/synthesized from woody biomass can enhance the mechanical properties such as strength and stiffness and impart additional functionalities to wood-based products. Cellulose nano-fibres/crystals find application in wide areas of materials science like reinforcement for composites. Incorporation of nanomaterials in resin has been used to enhance specific properties of wood-based composites. This review paper highlights some of the advancements in the use of nanotechnology in wood science, and its potential impact on the industry.
Chapter
The article is devoted to the study of the flammability of wood impregnated with fire retardants: melamine polyphosphate, melamine poly(magnesium phosphate), aluminum hydroxide, and boric acid. A serious drawback of wood, limiting its use, is its fire hazard. One of the methods for increasing the fire protection of wood is the application of fire retardants to the surface of wooden structures. To determine the effect of wood treatment with fire retardants on the fire performance properties of wood, the method for determining the ash content and the method of the quality control of fire-retardant treatment were used in this work. The study showed that the fire retardant efficiency of wood with a double application of the solution provides only melamine polyphosphate coating of wood. The treatment of wood with aqueous solutions of boric acid, aluminum hydroxide, and melamine poly(magnesium phosphate) requires multiple coating of these fire retardants. the ash content of the wood samples coated with fire retardants is higher than that of a sample of untreated wood.
Book
Full-text available
This is the only authoritative treatise to encompass the entire field of ignition. The 1116-page handbook was published in cooperation with the Society of Fire Protection Engineers, under whose auspices the peer review was performed.
Article
Full-text available
The irradiance plotted as function of time to ignition for wood materials tested in the Cone Calorimeter (ASTM E1354) differs signiticantly from that tested in the Lateral Ignition and Flame spread Test (LIFT) apparatus (ASTM E1321). This difference in piloted ignitabilty is primarily due to the difference in forced convective cooling of the specimen tested in both apparatus. By calibrating for the convective heat transfer coefficient and using appropriate heat conduction analysis of the test data for both apparatus, similar values were derived for the following thermo-physical properties: surface emissivity, surface ignition temperature, thermal conductivity, and thermal diffusivity. This paper reports on results with tongue & groove boards of clear grade redwood.
Chapter
Description 12 peer-reviewed papers on smoke toxicity, smoke chemistry, and smoke characterization. Studies address smoke hazard assessment, smoke spread modeling, appropriateness of toxicity tests, appropriateness of animal species as human substitutes, effect of materials on nuclear winter, and more. For researchers, regulators, producers, and those involved in the field of smoke research.
Article
New technologies in building materials have resulted in the use of a wide variety of materials in decks. As part of our effort to address fire concerns in the wildland-urban interface, the Forest Products Laboratory has been examining the fire performance of decking products. In addition to preservative-treated wood, decking products include wood-plastic composites and imported wood species. Requirements for wood-plastic composites are often based on expectation for untreated wood. Claims for imported wood species have included Class A flame spread ratings. This paper discusses results of cone calorimeter tests conducted on a variety of wood-based decking materials.
Article
We modified an Ohio State University (OSU) heat release apparatus to obtain piloted ignition data and heat release rate (HRR) using the oxygen Consumption method. To monitor deviation from the baseline heat flux to the sample during a test, an auxiliary heat flux meter was added to the OSU apparatus. The heat flux measured was the incident flux from the radiant panel and did not include flame flux. To obtain time to ignition, the gas phase was ignited with a pilot above the specimen. Six wood materials used in a series of room fire tests were tested at three different heating flux levels. The same materials are being tested in a Cone Calorimeter and the LIFT apparatus, but only data on Douglas-fir plywood were available for this paper. Ignition data from the three methods were in good agreement. Comparison of HRR between the OSU apparatus using the oxygen consumption technique and the Cone Calorimeter showed encouraging results. Good agreement was obtained for the first and second peaks of heat release. In the intermediate phase, the OSU apparatus gave higher HRR. Average heat of combustion was also slightly higher with the OSU apparatus. Experiments with a Cone Calorimeter holder in the OSU apparatus also showed higher heat release in the intermediate phase than in the Cone Calorimeter. The backing affected only the second peak of heat release. The difference in the intermediate phase was most likely due to the heating environment of each apparatus.
Article
Fire retardancy of wood involves a complex series of simultaneous chemical reactions, the products of which take part in subsequent reactions. Most fire retardants used for wood increase the dehydration reactions that occur during thermal degradation so that more char and fewer combustible volatiles are produced. The mecha-nism by which this happens depends on the particular fire retardant and the thermal-physical environment. This chapter presents a literature review of the inves-tigations into the mechanisms, a discussion of test methods used for determining fire retardancy, the var-ious formulations used to make wood fire retardant, and the research needs in the field of fire retardancy. wOOD WAS FIRST TREATED FOR FIRE RETARDANCY in the first century A. D. when the Romans used solutions of alum and vinegar to protect their boats against fire. In 1820, Gay-Lussac advocated the use of ammonium phosphates and borax for treating cellulosic material. Many of the promising inorganic chemicals used today were identi-fied between 1800 and 1870. Since then, the development of fire retardants for wood has accelerated. Commercially treated wood be-came available after the U.S. Navy (1895) specified its use in ship construction, and New York City (1899) required its use in buildings over twelve stories tall (1). production reached over 65 million board feet in 1943, but by 1964 only 32 million hoard feet was treated annually (1). Increased efforts to expand the use of wood products in insti-tutional and commercial structures may require wood to be treated with fire retardants. Therefore, research on fire-retardant treatments for wood has accelerated. Early Studies One of the earliest studies on fire-retardant treatments for wood was conducted between 1930 and 1935 (Forest Products Laboratory).