A Rising Concern: Failures of Adhered Rigid Insulation

A RISING CONCERN:
FAILURES OF ADHERED RIGID INSULATION
DAVID S. SLICK, PE, CFM
NICHOLAS A. PITEO, PE
ALEX J. KOSIS, PE
PHILIP K. FREDERICK, PE
SIMPSON GUMPERTZ & HEGER, INC.
2101 Gaither Road, Suite 250, Rockville, MD 20850
Phone: 301-417-0999 • Fax: 301-417-9825 • E-mail: NAPiteo@sgh.com
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1 P I T E O • 4 1
4 2 • P I T E O S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1
ABSTRACT
As part of the building enclosure, the roofing assembly must resist numerous challenges
to effectively separate the interior building environment from the exterior. When properly
designed and constructed, roofing assemblies can provide many years of service.
Conversely, poorly secured roofing assemblies are unable to resist significant wind loads
and are likely to become detached or damaged from wind events well below the design wind
pressures for the assembly. Simpson Gumpertz & Heger’s (SGH) recent investigation experience
at several buildings has shown that defects related to the design and installation of
adhered rigid insulation are often major contributors to failure of the roofing assembly. We
will describe several adhesive failure mechanisms identified by our recent roofing investigations
and discuss the lessons to be learned from these failures.
SPEAKER
NICHOLAS A. PITEO, PE — SIMPSON GUMPERTZ & HEGER, INC., ROCKVILLE, MD
NICHOLAS PITEO is a member of SGH’s technical staff in the Washington, DC, area
office with experience in both building technology and structural projects. His experience
includes investigation of existing structures and building envelopes, rehabilitation design,
and new design encompassing a variety of materials, including glass, metals, sealants,
masonry, wood, stone, concrete, and many additional materials related to building envelope
construction. He received his bachelor’s and master’s degrees in engineering from Penn
State and is a registered professional engineer in Maryland, New York, and Washington, DC.
ABSTRACT
As part of the building enclosure, the
roofing assembly must meet numerous
challenges to effectively separate the interior
building environment from the exterior.
Design and construction professionals have
many options to consider when selecting a
roofing assembly, including a variety of
materials and attachment methods.
Considering benefits that include competitive
costs, a large pool of available applicators,
and available light colors to meet the
growing demand for “cool” roofing assemblies,
roofing designers frequently prescribe
a single-ply membrane installed outboard of
the insulation for low-slope roofing applications.
While single-ply roofing assemblies
can be mechanically fastened or ballasted, a
large portion of these assemblies are
adhered systems. When properly designed
and constructed, single-ply roofing assemblies
can provide many years of service.
Conversely, poorly adhered single-ply roofing
assemblies are unable to resist significant
wind loads and are likely to be further
damaged or destroyed by wind events well
below the design wind pressures for the
assembly. Recent investigations of failed
single-ply low-slope roofing assemblies
revealed several common component adhesion
defects, as well as moisture within the
roofing assemblies. This paper discusses
several common adhesive failure mechanisms
and the detrimental effects of excess
moisture within adhered roofing assemblies
and presents case studies as examples of
these conditions.
ADHERED ROOFING ASSEMBLIES
Traditional roofing assemblies frequently
rely on mopped hot-applied asphalt to
secure faced rigid cellular polyisocyanurate
thermal insulation board to monolithic substrates.
Installation requires that roofing
mechanics heat the asphalt in a kettle,
spread a continuous layer of asphalt onto
the substrate, place insulation into the
rapidly cooling asphalt, and “walk” the
insulation into place. After only a few seconds,
the hot-applied asphalt bonds sufficiently
to hold the insulation in place.
Hot-applied asphalt also provides a
qualitative assessment of the moisture level
in a concrete deck. When applied to a “wet”
concrete deck, hot-applied asphalt foams,
which is an indication to the applicator that
the concrete substrate contains excessive
moisture and should be allowed additional
time to dry before installing the roofing
assembly. Each layer of hot-applied asphalt
also helps to retard the flow of moisture
vapor within the roofing assembly. Hot
fluid-applied asphalt is infrequently
installed in contemporary roofing assemblies
due to limitations that include management
of a hot kettle and a strong
asphaltic odor during installation, which
many people find objectionable.
A foamed adhesive installed in “beads”
or “ribbons” is frequently the insulation and
protection board adhesive of choice for contemporary
adhered roofing assemblies.
Insulation and protection-board installation
typically requires roofing mechanics to
apply “beads” or “ribbons” of consistent
width and spacing to the substrate, place
insulation and protection board into the
adhesive, and maintain contact between the
insulation or protection board and the substrate
until the adhesive develops adequate
bond strength to secure the insulation or
protection board. Such assemblies also typically
include a contact adhesive that is
used to adhere the roofing membrane to the
uppermost substrate. Properly designed
and constructed, contemporary adheredroofing
assemblies can provide reliable roofing
performance. However, improperly
designed or poorly constructed adheredroofing
assemblies cannot resist significant
wind loads and may incur damage or failure
from wind events well below the design wind
pressures for the assembly. Additionally,
foamed adhesive used as a component of
contemporary adhered-roofing assemblies
provides no indication of excessive concrete
substrate moisture content and does not
retard the flow of moisture vapor within the
roofing assembly.
In the following section, we discuss several
common adhesive defects in contemporary
adhered-roofing applications identified
in roofing investigation case studies, as well
as the effect of moisture found within the
A RISING CONCERN:
FAILURES OF ADHERED RIGID INSULATION
Photo 1 – Widely spaced, missing, and irregular insulation board adhesive
patterns within the roofing assembly.
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1 P I T E O • 4 3
roofing assemblies on roofing component
adhesion. These defects are independent of
the type of roofing membrane installed.
COMMON ADHESIVE DEFECTS IN
CONTEMPORARY ADHEREDROOFING
APPLICATIONS
Common Defects With the Application
of Foamed Insulation Adhesive
Foamed adhesive spacing and pattern.
Widely spaced beads of insulation adhesive
that exceed the manufacturer’s recommended
maximum spacing, as well as missing
and irregular or “randomly” placed
insulation adhesive beads that exceed the
manufacturer’s target pattern and spacing
(Photo 1) often contribute to failure of roofing
assemblies. Manufacturers and independent
testing facilities determine wind
uplift ratings with laboratory testing, which
forms the basis of building code compliance
and other approvals. Foamed adhesive
installation must comply with the spacing
parameters specified by the designer and
manufacturer to achieve the intended wind
uplift resistance. Corners and perimeter
areas of roofs require more closely spaced
foamed adhesive than the field of the roof to
resist larger design wind uplift pressures in
these areas.
Foamed adhesive quantity. Insufficient
quantity of foamed adhesive, or adhesive
beads narrower than the minimum width
recommended by the manufacturer, frequently
contribute to failure of roofing
assemblies (Photo 2). Testing by manufacturers
and independent test
agencies determine the
required width of foamed insulation
adhesive beads, and this
testing forms the basis of
building code compliance and
other approvals. The manufacturer’s
recommended minimum
bead width typically provides
sufficient adhesive that,
when compressed, spreads out
to engage both contact surfaces,
considering reasonable
variations in workmanship.
Common Defects With
Insulation Installation
Inadequate insulation boardto-
adhesive contact. Insulation
boards must be properly
installed to achieve desired
bonding strength between the
adhesive and insulation board.
Inadequate contact between
insulation boards and adhesive
results in unbonded insulation,
which can allow insulation
displacement, reduces
uplift resistance of the roofing
assembly, and increases the likelihood of
roofing assembly detachment. Cured adhesive
(frequently with a shiny surface) bonded
to the lower substrate surface that never
was adhered to the upper surface (Photo 3)
Photo 2 – Insulation adhesive beads narrower than the manufacturer’s minimum prescribed
3/8-in width.
Photo 3 – Cured insulation adhesive not compressed by the roofing assembly
above.
4 4 • P I T E O S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1
is indicative of a lack of initial
contact between bonding surfaces
or tardy placement of successive
layers of insulation. In
some instances, cured adhesive
can indent the underside of the
upper insulation board. Manufacturers’
literature typically prescribes
placement of insulation
boards in foamed adhesive within
a specific condition or state during
the adhesive curing process.
The time period to reach this condition
varies with ambient weather
conditions. Individual manufacturers
have different requirements
for insulation placement,
so one installation method does
not apply to all foamed adhesive
applications. Insulation adhesive
that is unbonded or weakly bonded
between insulation layers or
between the substrate and bottom
insulation layer will not provide
the specified roofing assembly uplift
performance.
Failure to maintain insulation board contact
with the substrate. Foamed adhesive
that contacts both bonding surfaces during
installation but separates prior to curing,
can leave thin “strings” of adhesive connecting
both surfaces (Photo 4) or completely
separate beads of adhesive on each surface
(Photo 5). Manufacturers’ literature typically
requires sustained pressure to hold insulation
boards in contact with the substrate
until the foamed adhesive reaches sufficient
strength to hold the insulation boards in
place. Roofing mechanics often maintain
contact between insulation and the substrate
by placing weighted containers, often
filled with roofing material, over the newly
installed insulation as temporary ballast of
uniform weight. This installation method
may provide adequate cure time for the
adhesive but is dependent on many factors,
including installation speed, ambient air
temperature, and relative humidity. With
this method, temporary ballast is usually
abundant near the beginning of a project;
however, it becomes more scant as construction
materials are installed on the
building. The roofing mechanic must provide
sufficient ballast material throughout
the project to ensure proper installation of
insulation with foamed adhesives. Failure
to hold insulation boards in place until the
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1 P I T E O • 4 5
Photo 4 – Insulation adhesive
that had contacted both upper
and lower insulation surfaces
during roofing installation, but
had separated and cured,
leaving thin “strings” of
adhesive connecting the upper
and lower insulation facers.
Photo 5 – Insulation
adhesive that cured
separately on upper and
lower insulation surfaces.
adhesive gains sufficient strength frequently
results in inadequately bonded insulation
layers, which can allow insulation movement
and roofing assembly detachment
under wind force. This issue is especially
prevalent with insulation boards that are
curled or otherwise out of plane prior to
installation.
Moisture in a Roofing Assembly
Moisture trapped within a roofing
assembly is another potential cause of insulation
and membrane detachment in
adhered-roofing assemblies. Commonly
installed roof insulation boards consist of
polyisocyanurate cores with organic facers.
Organic facers are sensitive to moisture and
deteriorate or lose cohesive strength when
exposed to moisture (Photo 6). The organic
facers form the majority of the bonding surface
for insulation adhesive, and their deterioration
can result in loss of adhesion
between roofing layers and failure of the
roofing assembly. Moisture in a roofing
assembly can also exacerbate the aforemen-
Summary of Observations for Case Studies 1 and 2
Table 1 – Observations of design or installation defects for Case Studies 1 and 2.
Table 2 –Observations of moisture within the roofing assembly for Case Studies 1 and 2.
Percentage of openings
with observed condition
Description of condition Case 1 Case 2
Sample openings with “randomly” placed beads of insulation that were not applied according 28% 72%
to the roofing or adhesive system manufacturer’s instructions, recommended adhesive patterns,
or recommended maximum spacing
Sample openings with adhesive beads narrower than the manufacturer’s minimum prescribed widths 41% 71%
Sample openings with cured adhesive on the lower insulation board or roof deck that was not 59% 75%
bonded to the upper insulation board surface. Some locations had adhesive that was not
compressed, suggesting that the adhesive cured before the next layer of insulation was installed.
Sample openings with adhesive that had contacted and then separated from upper and lower 62% 75%
insulation board surfaces during installation, leaving thin “strings” of adhesive connecting the
upper and lower insulation board facers or that had completely disconnected between the upper
and lower insulation board surfaces
Percentage of openings
with observed condition
Description of condition Case 1 Case 2
Sample openings with moisture between the top layer of insulation and the roofing membrane 79% Not Recorded
Sample openings with moisture on insulation board facers 83% 47%
Sample openings with staining on insulation board facers 86% 53%
Sample openings with apparent microbial growth on insulation board facers 97% 50%
Sample openings with partial cohesive failure of organic insulation facer during removal 100% 83%
Sample openings with partial adhesive failure of roofing membrane from insulation facer Not Recorded 57%
during removal
Sample openings with bowed or curled insulation (typically occurred at openings with 28% 23%
evidence of previous moisture)
4 6 • P I T E O S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1
Photo 6 – Wet organic insulation facer between the roofing membrane and the top
layer of insulation indicated by hydration paper that turns pink in the presence
of water.
tioned defects with foamed adhesive or
insulation board installation. Recently
placed concrete is wet and slowly releases
moisture as it cures. Roofing materials
placed over insufficiently cured concrete
substrates can trap excess moisture within
a roofing assembly, and this practice is a
common source of such moisture. Similarly,
concrete decks exposed to rain absorb and
store moisture, which slowly dries to the
environment. Wet roofing materials, ineffective
temporary roofing tie-offs during rain,
and moisture-laden air from occupied
spaces below that flows unrestricted into
the roofing assembly are other common
sources of excessive roofing assembly moisture.
ADHESIVE ROOFING ASSEMBLY
FAILURE CASE STUDIES
The case studies below briefly describe
two roofing assemblies that SGH investigated.
We list our observations of defects with
foamed adhesive application and insulation
installation in Table 1. We list our observations
of evidence of moisture within the
roofing assemblies in Table 2.
Case 1: TPO Roofing Membrane
Installation
The typical roofing assembly for Case 1
consisted of the following components, from
exterior to interior: TPO roofing membrane,
reportedly adhered with a synthetic polymer-
based contact adhesive; multiple layers
of paper-faced polyisocyanurate insulation,
reportedly adhered with a one-component
foamed adhesive; a 2- to 3-in-thick,
cast-in-place concrete topping slab; and
approximately 6-in-thick precast concrete
roof deck planks. The building owner
observed ballooning and billowing roofing
membrane at two buildings on the property
approximately three to four years after construction.
SGH investigated the roofing assembly
of each building and observed displaced
insulation and insulation board deformation
reflected through the TPO membrane
prior to making any openings in the roofing
assembly. We also observed occasional roofing
membrane ballooning and billowing at
all roof surfaces that we later determined
were caused by internal building pressure
and uplift pressure created by wind blowing
across the roofs. Emergency retention bars
installed over the roof surfaces reduced, but
did not eliminate, the extent of membrane
ballooning and billowing. We made 29 sample
openings in the roofing assembly during
our investigation and summarize our field
observations in Tables 1 and 2.
During our investigation, we measured
the temperature and relative humidity of
the concrete topping slab using a modified
version of ASTM F2170 (Photo 7). We had to
limit the time between readings due to facility
constraints. After 24 hours, the average
relative humidity of the topping slab was
75% (relative humidity ranged between
58.6% and 91.1%). We also tested an insulation
board sample in accordance with
ASTM C1616 and calculated a sample moisture
content of 66.29% by weight (i.e., the
sample was wet).
Case 2: CSPE Roofing Membrane
Installation
The typical roofing assembly for Case 2
consisted of the following components, from
exterior to interior: white chlorosulfonated
polyethylene CSPE roofing membrane,
reportedly adhered with a one-component
bonding adhesive; multiple layers of paperfaced
polyisocyanurate insulation, reportedly
adhered with a one-component foamed
adhesive or a two-component polyurethane
foamed adhesive; an approximately 3-inthick
composite cast-in-place concrete roof
deck; and a steel roof deck with 3-in-deep
flutes. Building facilities personnel
observed inflated and detached roofing
membrane and an irregular roof surface
Photo 7
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1 P I T E O • 4 7
4 8 • P I T E O S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1
approximately one year after construction.
We investigated the roofing assembly
and observed displaced insulation and
insulation board deformation reflected
through the CSPE membrane prior to making
any openings in the roofing assembly.
The membrane was also debonded from the
substrate in numerous locations prior to
making sample openings. Workers had covered
large portions of the roof with sand
bags as temporary ballast. We made 30
sample openings in the roofing assembly
during our investigation and summarize
our field observations in Tables 1 and 2.
During our investigation, we measured
the temperature and relative humidity of
the concrete topping slab using a modified
version of ASTM F2170. We had to limit the
time between readings due to facility constraints.
After four to 48 hours, the average
relative humidity of the concrete topping
slab was 73% (relative humidity ranged
between 66% and 85%). We also tested five
insulation board samples in accordance
with ASTM C1616 and calculated sample
moisture content that ranged from 3% to
42% by weight, with an average of 10% (i.e.,
the samples ranged from slightly wet to very
wet).
Summary of Hygrothermal Analyses for
Case Studies
For Cases 1 and 2, we modeled the
existing roofing assembly and performed
hygrothermal analyses with the WUFI Pro
4.2 computer program to simulate the
impact of residual moisture within the castin-
place concrete roof deck or topping slab,
considering seasonal, cyclic water vapor
transmission through the assemblies. We
used typical built-in moisture levels defined
in the WUFI program for the various components
of the roofing assembly for Case 1.
We input moisture levels similar to those
that we measured in the field and during
laboratory testing into the WUFI program
for the various components of the roofing
assembly for Case 2. We used material
property data, surface heat transfer coefficients,
and boundary condition temperatures
from the American Society of Heating,
Refrigeration, and Air-Conditioning
Engineers (ASHRAE) 2009 Handbook of
Fundamentals and the WUFI program database.
The Case 1 simulation covered a fouryear
time frame to model conditions from
building completion through the period
during which the roofing insulation and
membrane detachment occurred. The Case
2 simulation covered both a five- and tenyear
time frame from the time of our investigation
forward.
For both case studies, the hygrothermal
modeling showed that diffusive vapor transfer
from residual moisture in the roofing
assembly, primarily from the concrete substrate
and insulation, would drive toward
the building exterior during the initial heating
season. The moisture drive would
reverse and drive toward the building interior
during the cooling season. The low permeability
of the roofing membrane on the
exterior side of the assembly and the concrete
substrates on the interior side of the
assemblies prevent moisture from entering
or exiting both assemblies. This trapped
moisture cycles through the roofing assemblies
with changes in vapor drive between
heating and cooling seasons. Temperatures
within the roofing assemblies also vary with
the heating and cooling season and frequently
reach, or drop below, the dew point
temperature. The cyclical moisture drive,
combined with temperatures that are below
the dew point, result in condensation and
repeated wetting of the various roofing components.
These modeling results are consistent
with our field observations of moisture,
stains, and apparent mold growth within
the roofing assemblies.
Frequent condensation and wetting of
roofing components cause reduced cohesive
strength of the insulation facers, delamination
of the insulation facers, biological
growth within the roofing assembly, and
weakened adhesive bond strength among
the concrete substrate, insulation layers,
and roofing membrane. Additionally, moisture
trapped within a roofing assembly,
combined with roofing thermal cycles, also
contributes to dimensional changes and
bowing of the insulation boards, which
places additional stress on the adhesive
bond and wet insulation facers. Our field
observations corroborated such insulation
board bowing and loss of adhesive bond.
Case Study Conclusions
We concluded that the insulation board
adhesion defects and moisture in the roofing
assemblies were independently sufficient
to cause detached roofing insulation
and membrane for both Case 1 and Case 2.
As a result, the roofing assemblies could
not resist reasonably anticipated wind
forces. Permanent repairs for Case 1 and 2
required removal of the existing roofing
assemblies and replacement with similar
roofing assemblies.
DESIGN AND CONSTRUCTION
CONSIDERATIONS
The successful design of a roofing
assembly is contingent upon the designer’s
understanding of the specified roofing
materials and the anticipated load placed
on the assembly by occupants, climate, and
construction activity. Each project has
unique conditions that require consideration
during the design process. Additionally,
the proper installation of the design during
construction is paramount to the overall
success of a roofing assembly. We describe
several lessons learned from our investigations
in the sections below.
Foamed Adhesive and Insulation
Installation
The use of foamed adhesive to attach
insulation boards requires different techniques
than those used with hot-mopped
asphalt. Foamed adhesives are more sensitive
to imperfections in the installation
process than hot-applied asphalt, and failure
to follow proper installation procedures
may result in defects like those noted above
and contribute to roofing failures. We summarize
general recommended foamed-adhesive
installation procedures below:
• Determine the wind uplift rating of
the considered roofing assemblies,
and compare them to the projectspecific
design wind load requirements.
• Clean and prepare the substrate as
recommended by the adhesive manufacturer.
• Deliver, prepare, and apply foamed
adhesive as recommended by the
designer and manufacturer. Apply
foamed adhesive with the bead spacing
and bead size specified for the
project. Common application techniques
typically utilize handheld
applicators or carts manufactured
specifically for this purpose.
• Install an air barrier within the roofing
assembly if air passing from the
building into the roofing assembly
will be detrimental to roofing assembly
performance. A vapor retarder
may also function as an air barrier,
depending on product selection and
application.
• Install insulation within the specific
condition or state during the foamed
adhesive-curing process as recommended
by the manufacturer. Cure
time for foamed adhesive is dependent
on both temperature and relaS
Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1 P I T E O • 4 9
tive humidity and will influence the
installation of insulation. The
foamed adhesive manufacturer
should visit the project at the commencement
of the adhesive application
and insulation installation
process to verify acceptable application
and installation techniques
consistent with the manufacturer’s
recommendations.
• Provide intimate contact between
the insulation boards and the substrate
when placing the upper insulation
boards into the adhesive, and
maintain intimate contact until the
adhesive has adequately cured.
Insulation boards often will not conform
to the substrate and will
require continuous pressure until
the adhesive has gained sufficient
strength to hold the insulation in
place. Contemporary foamed adhesives
require longer time to develop
appropriate bond strength than traditional
hot-mopped asphalt. The
National Roofing Contractors Association
(NRCA) Roofing Manual recommends
against adhering insulation
boards thicker than 2 in and
larger than 4 ft by 4 ft, in part to
facilitate adequate adhesive contact.
• ASTM C1289-10 allows dimensional
variations in manufactured insulation
boards. Insulation boards that
are warped but within allowable
dimensional standards, and insulation
boards applied over decks that
are not sufficiently planar or otherwise
uneven, may need to be scored
and compressed to establish contact
with the roof deck or substrate.
Insulation boards that exceed
dimensional tolerances defined by
ASTM C1289-10 can contribute to
insulation adhesion problems and
should not be used.
Moisture Control in Roofing Assemblies
Excess moisture often contributes to
failure of roofing assemblies. Wet or insufficiently
cured concrete decks are a common
source of moisture within the roofing
assembly. The roofing industry (including
roofing assembly and roofing adhesive manufacturers,
roofing designers, and roofing
contractors) recognizes a concrete substrate
as a potential source of moisture in roofing
assemblies and generally specifies installation
of roofing materials over a properly
cured and “dry” substrate.
Common methods of concrete roof deck
moisture testing, such as the “plastic sheet
test” (ASTM D4263), can falsely indicate a
“dry” concrete substrate and are unreliable
as the sole evaluation of concrete moisture
levels. Unlike the flooring industry, which
continues to refine maximum moisture
threshold criteria and test protocols for
flooring substrates to help prevent adhesive
failures, the roofing industry does not provide
similar moisture criteria and test protocols
to evaluate concrete roof decks. This
is driven, in part, by unpredictable weather
conditions and schedule pressures that
force roofing contractors to quickly provide
a watertight facility and limit the amount of
time that a roof substrate can be left
exposed.
During our investigations of roofing failures,
we have attempted to develop more
sophisticated and reliable concrete moisture
testing for concrete roof decks based
on moisture testing protocols common in
the flooring industry. However, our data collection
is limited and frequently constrained
by the need to install or reinstall roofing
assemblies to provide a watertight facility.
The roofing industry must develop a guideline
or standard to practically and reliably
evaluate the moisture content of concrete
substrates along with criteria to define
appropriate concrete moisture content prior
to the installation of a roofing assembly to
help roofing professionals reduce the risk of
roofing failures due to wet substrates.
In some cases, scheduling or other considerations
may not allow a concrete substrate
to achieve appropriate moisture levels
prior to roofing installation. Elevated
interior relative humidity in occupied
spaces below the roof may also introduce
moisture into the roofing assembly through
air leakage and/or moisture vapor diffusion.
Such cases may require a vapor
retarder, air barrier, or both installed on the
concrete substrate as part of the roofing
assembly to limit moisture vapor diffusion
and/or air leakage into the roofing assembly,
both of which can contribute to moisture
within the roofing assembly and cause
moisture-related issues. An industry guideline
or standard should recommend installation
of a vapor retarder and/or air barrier
to address such conditions.
CONCLUSIONS
Improper application of foamed adhesive,
improper roofing design, and poorly
executed foamed-adhesive application and
insulation board installation can result in
catastrophic failure of roofing assemblies.
Instructions for foamed-adhesive application
vary between manufacturers and products
and can be affected by ambient weather
conditions. There is no uniform application
method, and relying on jobsite knowledge
obtained from experience with traditional
roofing adhesives to construct a roofing
assembly with foamed adhesive comes
with considerable risk. The roofing and
roofing adhesive industries must develop
guidelines or standards that provide more
reliable and practical foamed-adhesive and
insulation board installation protocol to
help roofing designers, roofing contractors,
and building owners reduce the risk of
installed roofing assemblies that include
inadequate insulation adhesive bond that
may lead to roofing assembly failure well
below the project-specific design wind load
requirements.
Additionally, other factors, such as
moisture trapped within a roofing assembly,
can also trigger or exacerbate roofing failures.
The roofing industry must develop
more reliable and practical roof substrate
moisture testing protocols to help designers,
contractors, and building owners
reduce the risk of roofing and adhesive failures
due to wet substrates. Until then, roofing
professionals must consider installation
of a continuous vapor retarder, and possibly
an air barrier, directly over the substrate
when installing roofing over concrete decks,
wet substrates, or interior spaces with elevated
relative humidity levels.
REFERENCES
American Society of Heating, Refrigerating,
and Air-Conditioning Engineers
(ASHRAE) 2009 Handbook of
Fundamentals, American Society of
Heating, Refrigerating, and Air-
Conditioning Engineers, Inc.,
Atlanta, GA, 2009.
ASTM Standard C1289-10: “Standard
Specification for Faced Rigid Cellular
Polyisocyanurate Thermal Insulation
Board,” Annual Book of ASTM
Standards, ASTM International,
West Conshohocken, PA, 2010.
ASTM Standard C1616-07e01: “Standard
Test Method for Determining
the Moisture Content of Organic and
Inorganic Insulation Materials by
Weight,” Annual Book of ASTM Standards,
ASTM International, West
Conshohocken, PA, 2010.
ASTM Standard D4263-83 (2005):
“Standard Test Method of Indicating
5 0 • P I T E O S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 1
Moisture in Concrete by the Plastic
Sheet Method,” Annual Book of
ASTM Standards, ASTM International,
West Conshohocken, PA,
2010.
ASTM Standard F2170-09: “Standard
Test Method for Determining Relative
Humidity in Concrete Floor
Slabs Using in Situ Probes,” Annual
Book of ASTM Standards, ASTM
International, West Conshohocken,
PA, 2010.
The NRCA Roofing Manual: Membrane
Roof Systems, National Roofing Contractors
Association, Rosemont, IL,
2011.
D.S. Slick, N.A. Piteo, D.A. Rutila,
“Adhesive Failures Due to Moisture
in Roofing Assemblies,” Fourth Symposium
on Durability of Building &
Construction Sealants & Adhesives.
D.S. Slick, N.A. Piteo, D.A. Rutila, “Case
Studies of Low-Rise Foam Adhesive
for Rigid Roofing Insulation,” Seventh
Symposium on Roofing Research
and Standards Development.