pia BUILDING CONSTRUCTION INFORMATION FROM THE CONCRETE AND MASONRY INDUSTRIES

August 1994

ASSESSING THE CONDITION AND REPAIR

ALTERNATIVES OF FIRE-EXPOSED CONCRETE
AND MASONRY MEMBERS
addition, because of the superior inherent fire resis-
tance of concrete and masonry materials, insitu repairs
can often be utilized rather than resorting to removal
and replacement procedures.
Wood-and steel-framed construction do not offer
these advantages. In other than small fires, wood
framing members are often consumed as fuel, leaving
demolition and reconstruction as the only viable
alternative. Unprotected structural steel suffers equally
destructive consequences. Lightweight steel trusses
and bar joists can collapse after just 5-10 minutes of
fire exposure.( 1l Steel columns of small cross-sectional
area typically possess a fire endurance of no more than
10-20 minutes. (2 l Although steel regains strength upon
cooling, the amount of recovery is dependent upon
the maximum temperature reached within the mate-
rial. Members that are severely distorted during a fire
will cool in the deformed position making strength gain
useless. The result is irreparable damage to the steel.
Fig. 1 shows the yield strength of steel at elevated
temperatures as a percentage of the yield strength at
21°C (70°F).

PURPOSE
The sections that follow describe an approach for
assessing the condition of fire-exposed concrete and
masonry building construction. Various testing and
analytical methodologies are descri_bed and some
After a building fire, concrete and masonry members may general information is provided about restoration
be all that remains. An accurate assessment of the damage
procedures. Detailed repair techniques are beyond
is essential to the restoration process.
the scope of this report.

INTRODUCTION INVESTIGATION PROCEDURES

When a building is damaged by fire, an inspection to Past experience has shown that concrete and masonry
assess the structural damage should be conducted possess excellent structural and barrier capabilities,
as soon as possible after the fire. With concrete and even in severe building fires. Under certain condi-
masonry construction, repair options (insitu restoration tions, however, concrete and masonry can suffer
or removal and replacement) versus demolition and significant distress. Examples include: fires involving
rebuilding are feasible alternatives in most cases. In temperatures well above those that result from the
 TEMPERATURE, °F
70 200 400 600 800 1 000 1200 1400
The first phase of the preliminary investigation is to
i:L 100 inspect the structural elements in affected areas for
0 ['... physical appearance. Observations of cracking,
!::::- ......... ~
~ spalling, deflections, distortions, misalignment, and
80
C\i ~"""'-- exposed steel reinforcement should be noted.
~
~ Various measurements of geometry, deflections,
_J
<(
f= 60 ' r'\. deformations, etc., can be taken of suspect members
~
u..
0
<f. 40
'\ for comparison against undamaged members of the
same structure. During the investigation, it is useful to
:r:
I- \. document and categorize the type of damage and its
(')
severity by building element such as beams, slabs, col-
z
w
a:
20 "i'o....... umns and walls. Having a summarized schedule of
tn damage allows for a broader picture of damaged mem-
0
I I II bers in need of a more detailed investigation, and is
u:J 0
>= 21 200 400 600 760 helpful in evaluating the extent and nature of the repair
TEMPERATURE, °C
process. Table 1 provides guidance for assessing and
Fig. 1. Yield strength of hot-rolled structural grade steels at categorizing damage of individual concrete members.
elevated temperatures as a percentage of initial yield strength
at 21 ac (70°F). Source: Ref. 3.
Types of Distress
Excessive deflection, large extensive cracks in struc-
turally sensitive areas, misalignments, and distorted
burning of ordinary combustibles; and, fires involving members are indications that the load-carrying
heavy fuel loads that cause prolonged burning of capabilities may have been seriously impaired. Where
combustibles while in direct contact with concrete or these types of conditions are present, strong consider-
masonry. ation should be given to the removal and replacement
In the aftermath, an assessment of damaged build- of affected members. Other factors, however, such as
ing elements and the formulation of repair plans must a building's architectural features and the
be done. Unfortunately, there is no single procedure importance of continuing occupancy can often dictate
that applies in every case. The best approach is one the selection of the restoration process. If the physical
that includes considerations of the fire intensity, age of conditions described above are absent, it is likely that
the structure, the importance of the affected areas or concrete and masonry members can be repaired
members, and the potential savings that may be gained in place.
by conducting a detailed investigation.(4l Conducting
extensive testing is not always prudent, as it may prove Concrete
more costly than just proceeding with the removal and Spalled areas of concrete may or may not represent a
replacement of a damaged member. A solid under- serious problem, but in any event, can prove to be a
standing of both structural engineering principles and useful source of information. In cast-in-place flexural
the effects of fire on building materials is invaluable in members, local buckling of reinforcing bars exposed
the decision making process. by spalling usually indicates that the steel has been
subjected to direct fire exposure. As the steel
The Preliminary Investigation approaches and reaches a temperature of 600°C
After data on the structure and fire event have been (1112°F), the bars lose about 50% of their yield strength
collected and safe entry to the building has been and are unable to resist the axial thermal restraining
established, a preliminary investigation should be forces imposed by the surrounding construction. Thus,
performed. Conducting the preliminary inspection buckling occurs.
becomes the single most important factor in evaluat- The absence of buckled or distorted exposed bars
ing the building's rehabilitation potential. The goals of in flexural members may indicate that spalling has
this investigation are to provide information on the occurred after the fire. If this is the case, the steel is
condition of the structure, the type and severity of the not likely to have reached 600°C (1112°F). In general,
problems in affected areas, the feasibility of rehabili- reinforcing bars in flexural members that lack signs of
tating the structure, and the need for conducting a severe distortion are unlikely to have suffered signifi-
detailed investigation. American Concrete Institute cant permanent reduction of yield strength. Similarly, if
(ACI) Report document 364.1 R, "Guide for Evaluation spalling does not extend to the steel (cover protection
of Concrete Structures Prior to Rehabilitation,"(s) is a remains intact), the structural strength of the member
good source document for guidance on conducting is relatively unaffected.
both preliminary and detailed investigations of For columns, the circumstances are different. In
concrete structures. columns that contain numerous ties or spiral confine-
2
Table 1. Initial Assessment of Damage and Probable Treatment Required

Damage Class 1 Damage Class 2 Damage Class 3 Damage Class 4

Soot and Smoke
deposits Present - - -

Color Change - Pink to buff surface Buff surface -

Spalling - Only minor Local Extensive
Steel exposure - - Steel showing Considerable areas
Surface separation Peeling Substantial Surface mostly
COLUMNS
gone. Remainder -
sounds hollow
when struck
Number of main
bars buckled - - Not more than one One or more
Microcracking - Extensive - -
Distortion - - Possible -
Reinforced concrete
solid slabs
Soot and Smoke Present General coverage Completely covered, -
deposits or color changed

Color change - - Pink -

Spalling Minor Present Present -
Steel exposure - 10% or less Over 10% -

Adherence of
SLABS Adhering Adhering Fallen clear
steel to concrete -
Plank Some broken Substantial damage - -
Ribs Intact - - -
Soot and Smoke
. deposits Present - - -
Spalling None Present Extensive -
Steel exposure - Small areas - -
Adherence of steel
to concrete - - Generally adhering Fallen clear
Suspended ceiling Extensive damage - - -

Deflection - - Not severe Substantial

Soot and smoke Present Completely covered - -
deposits or color changed
Color change - Pink Buff Buff to gray
Spalling Minor Substantial, Substantial Extensive
but at edges only on soffit sides on soffit sides
Steel exposure Little or none Outer edges of Main bars Almost all
corner bars each about 50% lower main bars
Surface separation Cover concrete
BEAMS - of soffit sounds - -
hollow when struck
Number of main
bars buckled - - Not more than one Possibly several
Microcracking - Surface - -
Cracking - - Several cracks of -
the order of 1/4 in.
Deflection or fracture - - Deflection not Substantial
severe deflection or
fracture or both
Probable Cosmetic only Some replacement Examination in Removal and
COLUMNS,
treatment required greater detail. replacement, or
SLABS,
Considerable strengthening
BEAMS
replacement, or extensively with
reclassification as additional concrete
Class 2 or Class 4 and reinforcement

Source: Ref. 6.

3
ment, it is very possible for reinforcing bars to reach to non-prestressed reinforcing steel (bars). Where the
temperatures of 600°C (1112°F) without exhibiting depth of pink concrete is less than the cover
signs of severe distortion or buckling. Exposed steel thickness , the reinforcing steel is not likely to be
in columns warrants a more detailed investigation . seriously affected by temperature . If the pink discol-
One other point to consider that applies to exposed oration extends all the way to the reinforcement,
steel in all reinforced members is the possibility that further investigation of the steel's strength, and the
the heated steel may have been quenched during concrete's strength beyond the depth of reinforcement
firefighting operations .< 7l Quenching of steel results in is necessary. Concrete in the region of discoloration
a loss of ductility that can severely affect the load must be removed prior to making repairs.
carrying capabilities of reinforced members. Caution must be observed, however, in placing heavy
Spalling of prestressed concrete that exposes steel reliance on this visual technique alone since the pink
strand represents a serious problem. Exposed strand discoloration is not always apparent. Tovey attributes
can often be an indication of loss of prestress, resulting the pink discoloration phenomenon in heated concrete
in a reduced load carrying capacity of the member. to the presence of ferrous salts in the
As such, a structural evaluation must be made for cement paste, aggregate and/or sand He also
determination of shoring requirements and other safety observes that concretes containing siliceous aggregates
considerations. A visual inspection of the ends of the appear to be more susceptible to this reaction than
member should be made, if possible, to determine if those containing calcareous or igneous aggregates.(s)
any bond loss has occurred, accompanied by inward With some fire-exposed concrete, the discoloration that
movement or slippage of the strand. Buckling of the occurs may be so faint that it is not discernible to the
strand is seldom encountered because it generally naked eye. In other instances, the pink discoloration
remains in tension, even though substantial prestress doesn 't occur. The latter supports the premise that it is
may have been relaxed or lost (see Figs. Sb-Se). not just the presence of elemental iron in concrete that
Further explanation of this effect is given later in this leads to this reaction , but rather the stability of the iron-
report . If circumstances exist to warrant strong containing compounds that is important. Since the
consideration for performing insitu repairs to cause of the phenomenon is not fully understood , one
prestressed members, a more detailed investigation should not conclude that the concrete is undamaged,
is needed. based solely on the absence of the pink discoloration
The observation of a pink discoloration to a given Shear failures in normal weight concrete beams
depth of concrete (see photo) indicates that minimum exposed to fire are rare. This is supported by labora-
temperatures of 300°C (572°F) have occurred. This tory tests and field investigations through the years. In
discoloration is accompanied by a significant loss of a test program designed to investigate the shear and
concrete strength within the discolored region. Some- flexural behavior of concrete beams exposed to fire
times the discoloration can be seen without the need (Ref. 9), all of the test specimens exhibited shear crack-
of extracting cores, such as in areas of spalled ing prior to the development of flexural cracks when
concrete. If this phenomenon has occurred, it can be subjected to the standard ASTM E119(1a) fire condition.
used as a tool in determining the likelihood of damage Additionally, all of the beams failed by flexure rather
than shear, even though some of the specimens
showed considerable signs of shear distress . Thus ,
shear strength of beams at elevated temperatures does
not appear to be a problem unless shear strength is
inadequate at room temperature .
In continuous beams and slabs, it is not unusual to
observe flexural cracking in the negative moment
region (over the supports) . This behavior can be
explained by Fig. 2. Assuming that a sufficiently
severe fire exists, a redistribution of moments takes
place early in the fire exposure period. Redistributed
moments are generally limited by the nominal negative
moment strength near the supports and have the
effect of significantly reducing the applied positive
moment. In an ASTM E119 fire test, yielding of the
negative moment reinforcement (Fig 2c) typically
occurs within the first 30 minutes. At some po1nt from
the beginning of the fire exposure to the time of yielding,
Note the presence of cracks in the region of pink
the cracking moment strength in the negative moment
discoloration. Also observe the partial change of aggregate
color (upper right) from yellow gold to red. regions is reached and cracks develop. As the
4
 W (UNIFORM LOAD) usually the damage is confined to within 19 mm (3/4
f
in.) from the face of the fire-exposed wall. Weakened
STRUCTURE
mortar does not significantly affect the load-carrying
ability of concrete masonry walls, as evidenced by
Menzel's test results described later in this report (Ref.
12). Further information on the evaluation of concrete
masonry walls after a fire can be found in Ref. 13.
For the most part, the same applies to clay masonry
walls. In addition, research suggests that clay
NO FIRE
masonry walls can tolerate substantially wide cracks
without significantly affecting the compressive strength
of the wall.( 14l These findings assume that no major
M'n(NOMINAL NEGATIVE MOMENT STRENGTH) wall deformations or misalignments have occurred. A
more detailed discussion of this is provided in subse-

~
quent sections.
In general, if concrete or clay masonry exhibits no
DURINGFIRE ~ ~
excessive deformations or large extensive cracks,
insitu repairs are a likely remedy. For reinforced

L M·n 0 (REDUCED NOMINAL NEGATIVE

MOMENT STRENGTH)
masonry, the absence of exposed steel usually
indicates that the load-carrying ability of the wall is
relatively unaffected.
Fig. 2. Moment diagram for a symmetrical interior span of a
continuous one-way slab or beam before and during fire
exposure-neglecting axial restraint due to thermal Field Testing Techniques
expansion. A visual inspection alone is not always sufficient to
adequately assess the extent of fire damage and
recommend proper corrective action. Some common
duration of the fire exposure increases, the nominal field techniques and basic tools that can be used to
positive moment strength is influenced by elevated supplement the visual inspection are described
temperatures and continues to diminish. If it decreases herein.
to the point where it becomes equal to the applied Use of an ordinary hammer and chisel in testing for
(redistributed) positive moment, a third hinge forms resonance (sounding) and impact resistance can
and collapse occurs. reveal preliminary information about the hardness,
Since cracking in the negative moment regions integrity, depth of damage, and seriousness of cracking
occurs well before significant reductions in the nominal of concrete and masonry construction. This method
positive moment strength, the severity of the damage may permit an experienced investigator to determine
due to cracking cannot be readily determined from a whether the damage is cosmetic or structural, but is
visual inspection of the member. This is why it is primarily used to determine the need for additional test-
beneficial to be able to estimate the intensity and ing. It is commonly used in the evaluation of fire-dam-
duration of the fire in comparison to the standard E119 aged slabs, foundations, and walls. When concrete
fire condition. Conversely, if no cracks are visible in or masonry are struck with a small or medium sized
the negative moment regions, the fire was not suffi- hammer, good materials will give the impression of
ciently severe to cause the concrete's cracking being solid or hard, whereas damaged materials will
moment strength to be reached. In this case, the sound hollow or muffled. A screwdriver or chisel can
strength of the steel reinforcement can be considered be used to probe surface areas and mortar joints for
to be unaffected. A detailed investigation of these softened spots. Indications of other than surface
regions may become necessary if the precise value of damage suggest the need for a detailed investigation
the residual yield strength of the steel must be known using one or more of the methods and techniques
for subsequent structural evaluations. described later in this report.
For suspect concrete construction, an impact
Masonry rebound hammer adds an element of detail by providing
Masonry can exhibit fire distress similar to that of limited information regarding location, type, and
concrete. Small hairline cracks, pitting of aggregates, extent of fire damage. It can be useful in distinguishing
shallow spalling and other surface damage indicates concrete that has been exposed and affected by
a need for only cosmetic repairs. Cracks in excess of intense fire versus that which has not been exposed to
1.6 mm (1/16 in) wide deserve further investigation. high temperatures. Readings from impact rebound
Chalky or softened mortar joints in areas subjected to hammers give indications of surface hardness, but are
the most severe fire exposure are not uncommon, but generally not useful in accurately determining
5
Table 2. Melting Points of Some Common Materials (sonoscope) impact-echo, and impulse radar
technologtes. Destructive testing methods typically
Material Approximate Melting Temperature
oc (oF)
involve the extraction of concrete core samples or steel
reinforcement from existing construction for laboratory
Polyethylene 110 121 (230- 250) examination and testing . These methods and
Lead 327 (620) techniques are briefly described in the following para-
Zinc 421 (790) graphs. Additional information is provided in Ref. 16.
Aluminum alloys 482- 649 (900 - 1200) A sonoscope measures the speed of sound through
Aluminum 649 (1200) concrete and relates this to estimated compressive
Glass (softens) 593- 732 ( 11 00 - 1350) strength , modulus of elasticity, and quality of hardened
Silver 960 (1760) concrete. When modified, it is also used for detecting
Brass and Bronze 871 -982 (1600 - 1800)
cracks, although it does not distinguish between crack
Copper 1082 (1980)
Cast Iron 1149- 1371 (21 00 - 2500) size or the amount that are present. The signal-trans-
Steel 1399 (2550+) mitting transducer and signal-receiving transducer
must be held on opposite sides of the tested member
and precisely aligned to obtain accurate results.
Because of this, the sonoscope is often impractical for
use in testing walls and slabs. The test method is
compressive strengths. The results are highly sensitive
standardized as ASTM Designation C 597.( 17)
to proper calibration of the test equipment.
Impact-echo NOT in vo lves the use of an impact
hammer to send a low frequency stress wave into the
Estimating the Fire Severity concrete. The wave energy is reflected back and
The second phase of the preliminary investigation measured with a receiving transducer on the same side
involves the observation of building contents in areas of the member and the signals are recorded on an
most severely exposed by fire. Melting points of various oscilloscope. The collected data can be used to
items give indications of temperature ranges that have detect, locate, and classify discontinuities such as voids,
occurred in localized areas (see Table 2). This permits delaminations, cracks, and bond loss between cement
an estimation of the maximum temperatures that paste and aggregates within hardened concrete.
occurred during the fire and is useful in establishing Magnetic and microwave (radar) methods are used
the relative severity of the actual fire to the standard to locate reinforcing steel and other embedments in
E119 test fire. If the two fire exposures are compa- concrete, establish the thickness of structural compo-
rable, the type and amount of thermal distress suffered nents, and detect the presence of voids. They are
by concrete and masonry elements in a real-world fire particularly useful in determining the thickness of
should be similar to that which occurs under the undamaged concrete over steel reinforcement in cases
standard fire exposure. where fire damage has not fully extended to the steel.
While observing fire debris, one should check for This information is needed to restore adequate
the existence or remains of polyvinylchloride (PVC) concrete cover protection.
materials. When PVC burns, it emits vapors that form Destructive test methods are likely to provide the
hydrochloric acid in the presence of moisture. The most comprehensive and detailed information about
combustion of large quantities of PVC's resulting in high damaged areas. Their drawbacks include being
concentrations of hydrochloric acid can constitute a destructive by nature , and more time consuming due
hazard to reinforcing steel If there is suffictent to testing and analytical work that is typically conducted
evidence for this concern, testing for the chloride ion at off-site laboratories . When using destructiv e
content in reinforced concrete or masonry should be methods, extreme caution must be exercised during
recommended as part of the detailed investigation. The the removal of concrete cores or steel specimens. This
objective is to assess the potential for delayed long-term is especially germane to prestressed or post-tensioned
corrosion of the steel that can occur with the movement concrete so as not to further damage or weaken the
of chloride ions through the concrete or masonry. structural integrity of the affected member. In taking
specimens from prestressing strand, it is best to cut
Conducting a Detailed Investigation only one wire. Removal of cores and steel samples
should only be don e under the supervision of a
Concrete structural engineer.
When it becomes clear from the preliminary investiga- Estimates of concrete compressive strength ,
tion that a detailed investigation must be done , there modulus of elasticity, and Poisson's ratio, can be
are several methods available for this purpose . determined from testing a limited number of extracted
Non-destructive testing (NOT) systems employ mea- core samples. Petrographic analysis of extracted cores
surement techniques that utilize pulse-velocity can provide information on bond loss between the
6
 Temperature, °F
cement matrix and reinforcing steel, crack orientation
32 400 800 t200 t600
and their relationship to the aggregate, microcracking,
tOO ~~~~~--~--~~--~--~--~--~
extent of concrete dehydration, chemical compositional ........
.... .... "'----Stressed to 0.4 f,'
changes of cementitious materials and aggregates, .... ....

"'
80
and temperature distribution within a given concrete ....
........ Unstressed
depth. Unstressed Residual _j ' ' ,
60 .... ....
Temperature gradient is of particular interest in
fire-exposed columns, because once known, strength- ''
40
''
temperature relationships shown in Figs. 3a-3c can be Avg. Initial f, ~ 26.9t MPa (3900 psi) ''
used to estimate the post-fire residual compressive '
strength of the concrete. Using the "unstressed 20 ''
residual" curves leads to conservative residual strength
Siliceous Aggregate Concrete
'
estimates. The other curves in the figures may be used
200 400 600 800
with sufficient expertise and sound engineering Temperature, oc
judgment. Pre-fire load conditions of columns, the
redistribution of loads to surrounding construction, and Fig. 3a. Compressive strength of siliceous aggregate
stress reduction due to strength loss of columns at concrete at high temperatures and after cooling.
elevated temperatures are factors that can influence
the decision in using an alternate curve. All of the
curves assume that there is no significant internal
cracking of the concrete. Temperature, °F
Similarly, temperature-dependent strength loss and 32 400 800 t200 t600
recovery of reinforcing bars and prestressing steel can tOO
.""··,
be estimated from Figs. 4 and 5, respectively, provided
that no quenching has occurred. 'iii 80
·.,·.,
""~
Figs. 4a-4b indicate that conventional reinforcing 0
'if.
··,..
bars having not yet reached a temperature of 592°C
(1100°F), representing a 50% reduction in yield
.c
Ol
c:
60
Unstressed Residual / ' '-· •....._
~
/jj ·· ................................ .
strength, will recover a high percentage of their strength ">
40
·u; Avg. Initial f
'
~ 26.9t MPa (3900 psi) -··-··-.....
.......
upon cooling. In essence, bars in flexural members "'~
CL
that do not appear to be severely distorted are unlikely E
0 20 Carbonate Aggregate Concrete
()
to have suffered significant permanent reduction of
yield strength. 0
The effect of temperature on prestressing steel is 0 200 400 600 800

more complex. In conjunction with the strength loss Temperature, oc

associated with the limit of proportionality (LOP) Fig. 3b. Compressive strength of carbonate aggregate
decreasing with temperature, permanent relaxation concrete at high temperature and after cooling.
losses in the steel can occur well after the exposure to
fire has ended. As the elevated temperatures reduce
the LOP to a value less than the initial prestress of the
strand (at ambient conditions), the prestress is simi-
Temperature, °F
larly reduced. Relaxation losses in the steel based on
32 400 800 t200 t600
the maximum temperature and duration of exposure, tOO
further contribute to a reduction in prestress. This ef- ~ Stressed to
, ··-....... ~ 4 f, (sanded)
fect is illustrated in Figs. 5a-5e. These relaxation losses ~ 80 , Unstressed '-. '-
are not recovered upon cooling. E ' , (unsanded)---""-....··-....... \
0
Deflected prestressed flexural members indicate that 'if. Unstressed '}' ' ' Unstressed ' · · ..
.c 60 Residual___/ ' (sanded)~ ' \
the prestressing steel has been sufficiently heated to Ol
c: (sanded) ............. - . . . .
cause a reduction in the effective prestress force. If
precise values of the steel's residual yield strength are "
~
/jj
>
·u;
40
-- --................ ~
U)
~
'
"Unsanded" Concrete~ t7.9 MPa (2600 psi) ' ' ,
needed, various metallographic tests for microhardness CL
Avg. Initial f,· of
E
can be performed as described in Ref. 20. 0
()
20 Avg. Initial f ,·of "Sanded" Concrete ~ 26.9MPa (3900 psi)
Lightweight Aggregate Concrete
For guidance on other evaluation methods, ACI
Report 364.1 Rr 5l contains tables correlating appropri- 0
0 200 400 600 800
ate evaluation and testing procedures with the investi- Temperature, oc
gation of specific physical conditions associated with Fig 3c. Compressive strength of lightweight aggregate
fire-damaged concrete. Where unusual construction concrete at high temperature and after cooling.

7
 Temperature, oF Temperature, °F
32 400 800 1200 1472 32 400 800 1200 1472
140 140
,.........
// \ '
120
.
f/ /
i ... -- ........ \
I// ,,'.
' .
,,
120

'/ \
\
100
~.' ,,
•\
100
........

-·-· ~--·--
'
....

CL
ro
"\'. \ CL
ro
''
~
()
0
C\J
80
"\'.,, \
\,\
~
()
0
C\J
80

cr; Hot cr; After cooling

"0
·l "0
"ID
·:;;..
\ "ID
·:;;..
60
0
~
0 \ 0
~
0
60

"\'
~.
\\
40
,,,.
\'
,\
40

,\ ,,,.
20 ~. 20
'

0 200 400 600 800 0 200 400 600 800

Temperature, oc Temperature, oc
Hot rolled mild steel (rebars)
Hot rolled high yield (bars)
Cold worked high yield (wire mesh)

Fig. 4a. Yield strength of steel reinforcement while hot. Fig. 4b. Yield strength of steel reinforcement after cooling.

or abnormalities make analytical evaluations suspect, can be made by saw cutting and removing portions
load tests can be performed on structures or struc- of, or whole masonry units from the wall, and subjecting
tural elements in accordance with Chapter 20 of ACI coupons (square specimens) of the units or damaged
318-89.< 21 l Past experience has shown that load tests face shells to compressive strength tests. Prisms may
are seldom necessary. also be cut from the wall for compression tests,
although this test method tends to be less accurate.
Concrete Masonry Non-uniform fire damage on opposite face shells of
Unlike the evaluation process for concrete construe- the units and corresponding strength differences can
tion, NOT field tests are generally not employed for lead to unreliable test results. More often than not,
fire-damaged concrete masonry walls. The hollow strength testing is unnecessary. In addition, there is
geometry of concrete masonry units (emu's) typically always the possibility of doing more harm to the walls
negates the usefulness of most of the more sophisti- during the extraction process than if the walls were
cated NOT techniques previously described. In many left intact.
cases, a visual inspection of the wall is sufficient to Extensive fire tests conducted by Menzel in the
assess the extent of damage, if any. 1930's illustrates the excellent structural fire endurance
Should a more detailed investigation be warranted characteristics of concrete masonry walls.< 12l Three
for strength determinations, destructive test methods 8-inch hollow concrete masonry wall specimens were
are available for this purpose. Strength determinations subjected to subsequent E119 fire exposure periods
8
 Temperature °F
32 200 400 600 752
2000 .-----..-----~----.-----. 290
Tensile test on untreated 0.78% carbon steel wire

250

1600

200

ro
ll..
1200 '(ij
6
~ 150 Ul
Ul
Ul
Ul ~
~ U5
U5 800
100

400 UTS - Ultimate tensile strength

50 Fig. 5a. Temperature effects on untreated cold
LOP - Limit of proportionality drawn prestressing wire. (left)

Fig. 5b - 5e. Extended temperature effects on

untreated cold drawn prestressing wire. (below)
0 100 200 300 400
Temperature, oc

1100 900
Relaxation at 100°C (212°F) Relaxation at 200°C (392°F)

ro
ll..
ro
ll..
~ ~
l!5 1000 l!5 800
~ ~
U5 U5

900 L------.L.-----L...-----.1 700 L - - - - - - L - - - - - - - - ' - - - - - _ J

0 2 4 6 0 2 4 6
Duration, hours Duration, hours

700 300

Relaxation at 300°C (572°F) Relaxation at 400°C (752°F)

ro ro
ll..
ll..
~ ~

Ul
Ul
Ul
Ul
200 -
~ ~
U5 U5

500 100
0 2 4 6 0 2 4 6
Duration, hours Duration, hours

9
of 2 and 1/2 hours after being initially tested for periods of the wall thickness( 22l The direction in which the wall
of 2 and 1/2, 3, and 3 and 1/2 hours, respectively. Upon collapses will depend on the type and integrity of the
conducting strength tests , results showed no appre- lateral support system.
ciable differences in wall-to-unit strength ratios than Test results from over 200 full-scale fire tests and aux-
were experienced from conventional single fire exposure iliary tests in Australia comparable to ASTM E 119 fire
test procedures. The results suggest that concrete testing support the conclusion that concentrically
masonry walls can withstand one severe fire without loaded masonry walls (unreinforced and reinforced)
replacement, and still be able to perform structurally do not suffer sufficient strength loss at elevated
in the event of a second severe fire Menzel's fire test temperatures for walls to fail in compression(22l When
program of 215 concrete masonry walls additionally fire-exposed clay masonry walls fail, buckling is more
showed that mortar joints generally softened to a depth likely to control the mode of failure . This is largely due
of about 12-19 mm (one-half to three-fourths inch) from to the bond strength between the mortar and brick
the exposed face when subjected to ASTM E119 fires being substantially lower than the reduced compres-
for varying periods of up to 9 hours. The evidence sive strength of the wall. Adding reinforcement to clay
suggests that weakened mortar has little effect on the masonry walls virtually eliminates this buckling
axial load carrying ability of concrete masonry walls. potential.
Cracked loadbearing walls, within limits, still have
Clay Masonry substantial load carrying capabilities. Re sults of
For many of the same reasons expressed regarding structural tests conducted under non-fire
concrete masonry, post-fire inspection procedures for conditions at the Building Research Station in England
clay masonry walls are typically limited to a visual have shown that 229-mm (9-inch) brick walls with a
investigation. As with concrete masonry, signs of stepped or slanted crack up to 25 mm (1 inch) wide
deflection, cracking, deformation, and surface defects can still carry a minimum of 70% of its vertical load
should be observed and documented . Some behavioral capacity provided that the damage is not accompanied
information on clay masonry walls exposed to fire is by considerable transverse displacement. If walls are
provided below. out of plumb by not more than 25 mm (1 in), or bulge
The high fire resistance of fired clay brickwork is well no more than 12 mm (1/2 in) in a normal story height,b
known, but its ability to retain strength on cooling no repairs are usually necessary on structural grounds
compared with other materials is not sufficiently alone .< 14 l Because fire (up to about 1000°C (1852°F))
appreciated. Clay masonry units show little strength has very little effect on the compressive strength of
loss when heated to temperatures of up to 1ooooc clay masonry walls, it is reasonable to
(1832°F) , while mortars have virtually no strength at assume that cracks of the aforementioned width will
these temperatures and begin to lose substantial impact the heat-reduced load carrying capabilities of
strength at temperatures above 300-400°C (572- masonry walls to a similar extent. Cracks of the
7520F). However, mortar damage is usually confined to magnitude indicated above are significantly larger than
a shallow depth of approximately 12-19 mm (one-half those that would be expected due to fire exposure .
to three-fourths inch) from the surface. Buckling of clay masonry walls due to exterior fire
While clay masonry units can have compressive exposure is extremely rare. It is reasonable to attribute
strengths of up to about 138 MPa (20,021 psi), the this in part to the fastening system that gives the wall
allowable height for load bearing unreinforced masonry stability as it bows outward toward the fire. In the case
walls is limited by slenderness restrictions (height-to- of brick veneer walls, the wall is kept from buckling
thickness ratios). The greater the unsupported wall due to restraint provided by wall ties and framing
height for a given thickness (increasing slenderness elements. If the fire does not enter the building through
ratio), the more susceptible it is to buckling.a Expo- openings , thereby leaving the integrity of the connec-
sure to heat from interior fire expos ure accentuates this tions unaffected, brick masonry is capable of withstand-
behavior, as differential expansion between hot and ing even severe fire exposure for prolonged periods of
cool surfaces causes the wall to bow towards the fire. time . Some brick veneer walls (90 mm and 110 mm
If the deflection exceeds a critical amount, the wall (3.5 in. and 4.3 in.) masonry wythes) in the Australian
becomes unstable and experiences sudden collapse. test program were able to withstand the fire test for
Research suggests that this critical point occurs as over two hours .< 22l
the mid-height deflection of the wall reaches about 80%

bAithough most of the author's consulting work is done on commer-

aFor purposes of thi s section , the term "buckling " is used to cial/industrial projects, normal story height should be assumed to
describe the phenomenon of excessive bowing resulting in collapse ran ge between 2.4-3.6 m (8-12ft). A conservative approach should
due to tensile or shear bond failure at the brick-mortar interface. be taken where marginal conditions are present.

10
The Repair Process Removal and replacement -
Concrete Favorable characteristics of a removal and replace-
ment process can include:
Repair Options Versus Demolition and Rebuilding
• it is likely to be less expensive than insitu repairs for
As mentioned previously, concrete structures are more
severely damaged members
likely to be repairable after a fire than wood or steel-
• overcomes space or dimension limitation problems
framed structures . In assessing the repair alternatives,
sometimes experienced with in-place repair tech-
completion of the detailed investigation largely deter-
niques, e.g., shotcreting when space between steel
mines how much work is required. Depending on the
is less than 64 mm (2 .5 in.)
degree of damage, some concrete members may need
• it is more conducive to precast concrete construc-
no repair due to overdesign, some may only need
tion due to relative ease of removal, and ,
cosmetic repairs, and others may have to be strength-
• it is less dependent on the quality of the original
ened or removed and replaced
construction and antiquated design features.
Repair work should be supervised by a structural
engineer for numerous reasons. For example, unsafe
Drawbacks can include :
load transfers that may occur during shoring opera-
• greater debris removal costs
tions or during the removal and replacement of
• accessibility problems may necessitate the removal
structural members must be accounted for and
and replacement of sound walls and/or rqofing
guarded against. Loads that are to be supported by
• more shoring is required to carry the loads for-
new concrete must be temporarily supported by other
merly carried by removed members
means during the placement and curing periods .
• increased structural analytical work and supervision
Design standards and load requirements may have
is required to assure that excessive loads are not
also changed since the building was erected, requir-
transferred to other members during various stages
ing modified sections to comply with current building
of the removal and replacement process
codes. In addition , structural evaluations may become
• temporary walls and coverings may need to be
necessary at various stages of the project, should any
erected, and,
deviations in the repair process occur.
• undamaged members and supports may become
Most damaged structures can be repaired using the
damaged in the process of removing damaged sections.
same concrete placement techniques that are used
for new construction. Original building design and
Demolition and rebuilding- This is not an economically
space restrictions, however, can often have a signifi-
feasible alternative except in extremely rare cases.
cant influence on the selection of an appropriate
Particularly with cast-in-place concrete construction ,
repair method . Installation considerations and
demolition is usually a very difficult, time consuming,
economic factors that may not be readily apparent with
expensive, and disruptive process . This is largely due
respect to repair alternatives are described below.
to its exceptional durability and monolithic construction.

lnsitu restoration - Effecting Repairs

Favorable characteristics of insitu restoration can For concrete, the most common repair approach is to
include:
patch the existing member with concrete and install
• it can be done in a relatively short period of time reinforcing steel where necessary to restore the load
• occupancy of undamaged portions of the building carrying capacity Loose and damaged concrete
can continue should be chiseled away, being careful not to disturb
• business interruption losses are minimized the bond between the steel and the undamaged con-
• no large specialized equipment is necessary crete . Sandblasting should be done to clean steel and
• minimal or no shoring is required, and , concrete surfaces that are to receive fresh concrete.
• debris removal and disruption from construction traffic
Smooth concrete surfaces should also be sufficiently
are minimized.
roughened by bushhammering or sandblasting p:ior
to placing fresh concrete. Repair techniques for fire-
Drawbacks can include: damaged concrete are basically the same as those
• it can be more costly than removal and replacement used for repairing any other types of distressed
for severely damaged members concrete, i.e. , forming, shotcreting , etc. One possible
• it is less feasible for extensive damage to beams exception involves the use of epoxies and this is
and slabs, and , discussed later. Detailed repair procedures are outside
• repair work can become costly if the quality and the scope of th1s report. Ref. 23, however, provides an
design of original construction greatly differs from excellent source of information on repair techniques .
current code requirements. If structural members are required to possess fire

11
resistance by the governing building codes , epoxy • Economic considerations must be weighed with other
resins should not be used for the repair of large cracks factors to determine whether insitu repairs of concrete
or spalled areas that would result in inadequate construction should be performed, or removal and
concrete cover protection to reinforcement ( 8l Epoxy replacement is more feasible. Existing architectural
resins have low melting points that make them features and the importance of continuing
susceptible to run off in the event of a fire. Unless occupancy can often dictate the selection of the
appropriate test data is available indicating that a restoration process.
specific product can demonstrate this type of thermal • Nonprestressed concrete and reinforced masonry
resistance , epoxy resins should not be used for this members whose cover protection has remained in
application An exception would be for patching of place after fire exposure are unlikely to have suffered
cracks in areas of decreased fire exposure such as any significant loss of structural strength, provided
negative moment regions of flexural members that no major deformations or misalignments have
occurred.
Concrete Masonry • No field tests are generally performed in conducting
If a fire-distressed concrete masonry wall is free of large investigations to assess fire-damaged masonry walls .
deflections, it is likely that repairs will be minimal. Post-fire investigations typically consist only of visual
Damaged mortar can be removed and tuckpointed , inspections.
and cracks can be readily patched. For detailed infor- • If no severe distortion, cracking or displacement of
mation on crack repair, the following source material masonry walls is present, complete reinstatement of
can be referenced.( 14 ·24l the wall can usually be accomplished by patching
cracks and tuckpointing mortar joints.
Clay Masonry
As with concrete masonry walls , the absence of large
deflections or deformations usually indicates that
repairs will be minimal . For cracked walls , repair
methods will depend on the size and type of crack
and wall surface . Fine cracks (less than 1.6 mm (1/16
in)) are not very conspicuous and in brick masonry
would often be made more noticeable by repainting .
Such cracks can be filled by surface grouting that will
prevent water penetration and not greatly change the
wall appearance of relatively smooth walls . Clear
coatings intended to prevent water penetration of
masonry typically do not bridge cracks and, therefore,
will be ineffective in preventing the entry of water( 25l
Additional details on crack repair methods for masonry
are provided in Refs. 14 and 26 . If there are no severe
deflections or deformations and none of the cracks
are considered to be excessive, removal of the
distressed mortar and tuckpointing is usually sufficient
for complete restoration of the wall. (27 l

SUMMARY
• Concrete and masonry members damaged by fire
offer repair options that are not available to building
elements constructed of other materials. Even in
severe fires, complete demolition and rebuilding of
concrete and masonry structures is seldom necessary.
• The single most important item in evaluating the
rehabilitation potential of concrete after a building
fire is the preliminary investigation. Goals of this
investigation are to provide information on the
condition of the structure, the type and severity of
the problems in affected areas, the feasibility of
rehabilitating the structure, and the need for
conducting a detailed investigation.
12
REFERENCES Ratings, Concrete Reinforcing Steel Institute,
Chicago, IL, 1980, P. 8-4.
1. Fitzgerald, Robert W., "Structural Integrity During 12. Menzel, Carl A., Tests of the Fire Resistance and
Fire," Fire Protection Handbook, 16th edition, Strength of Walls of Concrete Masonry Units,
National Fire Protection Association, Quincy, MA, Portland Cement Association, Chicago, IL, 1934.
1986, P. 7- 103. 13. Evaluation of Concrete Masonry Walls After Being
2. Structural Fire Protection: Manual of Practice, T.T. Subjected to Fire, NCMA TEK 7-5 (formerly TEK
Lie, editor, prepared by the ASCE Committee on 117), National Concrete Masonry Association,
Fire Protection - Structural Division, American Herndon, Virginia, 1980.
Society of Civil Engineers, New York, NY, 1992, 14. Grimm, C.T., "Masonry Cracks: A Review of the
P. 70. Literature," Masonry: Materials, Design, Construc-
3. Reinforced Concrete Fire Resistance, First edition, tion, and Maintenance, ASTM STP 992, Harry A.
prepared by the CRSI Committee on Fire Ratings, Harris, editor, American Society for Testing and
Concrete Reinforcing Steel Institute, Chicago, IL, Materials, Philadelphia, PA, 1988, P. 257-280.
1980, P. 5-1. 15. Gustaferro, Armand H., and Martin, Leslie D.,
4. Gustaferro, Armand H., "Experiences From Design for Fire Resistance of Precast Prestressed
Evaluating Fire-Damaged Concrete Structures," Concrete, Second edition, PCI MNL-124,
Fire Damaged Structures, ACI SP 80-10, Prestressed Concrete Institute, Chicago, IL, 1989,
American Concrete Institute, Detroit, Ml, P. 276. P. 65.
5. ACI Committee 364, "Guide for Evaluation of 16. Muenow, Richard A., and Abrams, Melvin S.,
Concrete Structures Prior to Rehabilitation," "Nondestructive Testing Methods for Evaluating
Report 364.1 R, published in ACI Materials Damage and Repair of Concrete Exposed to Fire,"
Journal, Sept.-Oct. 1993, American Concrete Evaluation and Repair of Fire Damage to Concrete,
Institute, Detroit, Ml, 1993. T.Z. Harmathy, editor, ACI SP 92, American
6. "Concrete Structures After F'1re," Concrete Concrete Institute, Detroit, Ml, P. 63-86.
Construction, March 1972, Vol. 17, Number 3, 17. ASTM Designation C 597-83 (Reapproved 1991),
Concrete Construction Publications, Inc., Addison, "Standard Test Method for Pulse Velocity Through
IL, 1972, P. 101. Concrete, Section 4, Vol. 4.02, American Society
7. Krampf, L. and Haksever, A., "Possibilities of for Testing and Materials, Philadelphia, PA, 1993.
Assessing the Temperatures Reached by 18. Abrams, M.S., Behavior of Inorganic Materials in
Concrete Building Elements During a Fire, PCA Research and Development Bulletin
Fire,"Evaluation and Repair of Fire Damage to RD067.01 M, Portland Cement Association, Skokie,
Concrete, T.Z. Harmathy, editor, ACI SP 92, IL, 1980, P. 7.
American Concrete Institute, Detroit, Ml, 1986, 19. "Assessment and Reinstatement of Fire Damaged
P. 127. Concrete Structures," Concrete for Fire Resistant
8. Tovey, A.K., "Assessment and Repair of Fire- Construction- Cembureau report published by the
Damaged Concrete Structures-an Update," Cement and Concrete Association, Wexham
Evaluation and Repair of Fire Damage to Springs, Slough, 1979, P. 35-36.
Concrete, T.Z. Harmathy, editor, ACI SP 92, 20. Abrams, M.S., and Erlin, B., Estimating Post-Fire
American Concrete Institute, Detroit, Ml, 1986, Strength and Exposure Temperature of
P. 50. Prestressing Steel by a Metal/agraphic Method,
9. Lin, T.D., Ellingwood, Bruce, and Piet, Olivier, PCA Research Dept. Bulletin 219, Portland
Flexural and Shear Behavior of Reinforced Cement Association, Skokie, IL, 1967.
Concrete Beams During Fire Tests, PCA 21. ACI Committee318, "Building Code Requirements
Research and Development Bulletin RD091T, for Reinforced Concrete, (ACI 318-89) (Revised
Portland Cement Association, Skokie, IL, 1987. 1992)," American Concrete Institute, Detroit, Ml,
10. ASTM Designation E 119-88, "Standard Methods 1992.
of Fire Tests of Building Construction and 22. Gnanakrishnan, N. and Lawther, R., "Performance
Materials," Section 4, Vol. 4.07, American of Masonry Walls Exposed to Fire," Proceedings,
Society for Testing and Materials, Philadelphia, PA, Fifth North American Masonry Conference, Vol.
1993. Ill, University of Illinois at Champaign/Urbana,
11. Reinforced Concrete Fire Resistance, First June 3-6, 1990, P. 901-914.
edition, prepared by the CRSI Committee on Fire 23. "Reinstating Fire-Damaged Structures," Concrete
13
 Construction, May 1972, Vol. 17, Number 5,
Concrete Construction Publications, Inc., Addison,
IL, 1972.
24. Maintenance of Concrete Masonry Walls, NCMA
TEK 8-1 (formerly TEK 44 ), National Concrete
Masonry Association, Herndon, Virginia, 1972.
25. Colorless Coatings for Brick Masonry, BIA
Technical Notes on Brick Construction 7E, Brick
Institute of America, Reston, Virginia, Reissued
February 1987.
26. Moisture Resistance of Brick Masonry-
Maintenance, BIA Technical Notes on Brick
Construction 7F, Brick Institute of America, Reston,
Virginia, Reissued February 1987.
27. Bricks - Their Properties and Use, The Brick
Development Association, The Construction Press
Ltd., Lancaster, England, 1974, P. 204.

14
 Member organizations of the
NATIONAL CODES AND STANDARDS COUNCIL .

ACI American Concrete Institute

ASCC American Society for Concrete Construction
BIA Brick Institute of America
ESCSI Expanded Shale Clay & Slate Institute
NAA National Aggregates Association
NCMA National Concrete Masonry Association
NRMCA National Ready Mixed Concrete Association
NSA National Stone Association
PCA Portland Cement Association
PCI Precast/Prestressed Concrete Institute
TCA Tilt-up Concrete Association
WRI Wire Reinforcement Institute

This publication is intended for the use of professional personnel competent to evaluate the significance and limitations
of its contents and who will accept responsibility for the application of the material it contains. The National Codes and
Standards Council of the Concrete and Masonry Industries disclaims any and all responsibility for the application of the
stated principles or for the accuracy of the sources other than work performed or information developed by the Council.

National Codes and Standards Council of the Concrete and Masonry Industries
5420 Old Orchard Road, Skokie, Illinois 60077-1083

Printed in U.S.A. SR322.01 B