Ductile Concrete

Fracture Mechanics of Concrete and Concrete Structures High Performance, Fiber Reinforced Concrete, Special Loadings and Structural
Applications- B. H. Oh, et al. (eds)

2010 Korea Concrete Institute, ISBN 978-89-5708-182-2
Blast loading response of ultra high performance concrete and reactive

powder concrete slabs
J.H.J. Kim, N.H. Yi, I.S. Oh & H.S. Lee
Yonsei University, Seoul, Korea
J.K. Choi & Y.G. Cho
Hyundai Institute of Construction Technology, Yongin, Korea
ABSTRACT: In recent years, there have been numerous explosion-related accidents due to military and terrorist activities. Such incidents caused not only damages to structures but also human casualties, especially in urban areas. To protect structures and save human lives against explosion accidents, better understanding of the
explosion effect on structures is needed. In an explosion, the blast load is applied to concrete structures as an
impulsive load of extremely short duration with very high pressure and heat. Generally, concrete is known to
have a relatively high blast resistance compared to other construction materials. However, normal strength
concrete structures require higher strength to improve their resistance against impact and blast loads. Therefore, a new material with high-energy absorption capacity and high resistance to damage is a better material for
blast resistance design. Recently, Ultra High Strength Concrete (UHSC) and Reactive Powder Concrete (RPC)
have been actively developed to significantly improve concrete strength. UHSC and RPC can improve concrete
strength, member size and weight reductions and workability improvement. High strength concrete usages in
better earthquake resistance and increase a building height and bridge span. Also, UHSC and RPC can be implemented for blast resistance design of infrastructure due to terror or impact such as 9.11 terror attack. Therefore, in this study, the blast tests are performed to investigate the behavior of UHSC and RPC slab subjected to
blast load. Blast wave characteristics, including incident and reflected pressures as well as maximum and residual displacements and strains in steel and concrete surface are measured. Also, blast damages and failure modes
were recorded for each specimen. From these tests, UHSC and RPC are shown to effectively resist blast explosions compare to normal strength concrete.
1 INTRODUCTION
In recent years, there have been numerous explosionrelated accidents due to military and terrorist activities. Such incidents caused not only damages to
structures but also human casualties. Especially, in
metropolitan areas which are exposed to terror attack, these severe loading related accidents can cause
great human causalities, economical losses, and public infrastructure destructions, and civilian structure
collapses. To protect structures and save human lives
against explosion accidents, better understanding of
the explosion effect on structures is needed. In an
explosion, the blast load is applied to structures as an
impulsive load of extremely short duration with very
high pressure and heat.
Generally, concrete is known to have a relatively
high blast resistance compared to other construction
materials. However, normal strength concrete structures require higher strength to improve their resistance against impact and blast loads. Therefore, a
new material with high-energy absorption capacity
and high resistance to damage is a better material for

blast resistance design. Recently, Ultra High Strength
Concrete (UHSC) and Reactive Powder Concrete
(RPC) have been actively developed to significantly
improve concrete strength. UHSC and RPC can improve concrete strength, reduce member size and
self-weight, and improve workability. Commonly,
UHSC and RPC produce compressive strength
greater than 150MPa and sometime up to
180~200MPa. High strength concrete are used to
improve earthquake resistance as well as constructions of high-rises and long span bridges. Also,
UHSC and RPC can be implemented to blast resistance design of infrastructure against terror or impact
(ASCE 1999, Baker 1973).
The Korean building code has been modified in
year 2009 where any high-rises located in the city of
Seoul with the height of over 50 above ground floors
or 200m, the terror resistant design has to be incorporated. This code regulation reflects the keen public
interest on blast resistance and protective design
concepts. However, since UHSC or RPC has been
assume that the evaporable water is a function of

relative humidity, h, degree of hydration, c, and
2 LITERATURE LEVIEW
degree of silica fume reaction, s, i.e. we=we(h,c,s)
= age-dependent sorption/desorption isotherm
2.1 Characteristic of blast load
(Norling Mjonell 1997). Under this assumption and
An
is a Equation
very fast chemical
reaction producby explosion
substituting
1 into Equation
2 one
ing
transient air pressure waves called blast waves.
obtains
For a free-air burst, the blast wave will travel away
from
source as a spherical
wave front as shown
w
w the
e & + we &and
e h +1(a).
&the durain Figure
The
overpressure
) =
(D
hpeak
(3)
c
s + wn
h
h
t
tion of the overpressure vary
from the
c with distance
s
explosives. The magnitude of these parameters also
depends
theis explosive
from which the
the slope materials
of the sorption/desorption
where won
e/h
explosive
compound
is
made.
Usually
the size of The
the
isotherm (also called moisture capacity).
explosive
is given 3)in must
termsbeof
a TNT
governing compound
equation (Equation
completed
weight.
Explosive
behavior
on a number of
by appropriate
boundary
anddepends
initial conditions.
factors:
ambient between
temperature,
ambient of
pressure,
exThe relation
the amount
evaporable
plosive
composition,
explosive
material
properties,
water and relative humidity is called adsorption
and
the nature
the ignition
type. Additional
isotherm
if of
measured
withsource
increasing
relativity
factors
type, energy,
and duration
of the
humidityinclude
and desorption
isotherm
in the opposite
events
as
well
as
geometry
of
surroundings
(i.e.,
concase. Neglecting their difference (Xi et al. 1994),
in
fined
or
unconfined).
When
a
condensed
high
explothe following, sorption isotherm will be used with
sive
is initiated,
reaction
generates
several
reference
to bothexplosion
sorption and
desorption
conditions.
additional
characteristics
such as blast
of very
By the way,
if the hysteresis
of wave
the moisture
high
pressure,
the explosive
case
isotherm
wouldfragmentation
be taken intofrom
account,
two different
or
structural
elements,
hot
gas
with
a
pressure
from
relation, evaporable water vs relative humidity, must
100
up to
300 kilobar,
temperature
of about
be used
according
to theand
signa of
the variation
of the
3,000~4,000.
The main
is impulsive
relativity humidity.
The blast
shapeeffect
of the
sorption
pressure
the blastbywave
shown in
isotherm loading
for HPCfrom
is influenced
manyasparameters,
Figure
1(b)
(Baker
1973,
Mays
&
Smith
1995).
especially those that influence extent and rate of the
After a reactions
short time,and,
the inoverpressure
behind pore
the
chemical
turn, determine
shock
front
and becomes
smaller than
structure
anddrops
pore rapidly
size distribution
(water-to-cement
that
of
the
surrounding
atmosphere
as
shown
in Figratio, cement chemical composition, SF content,
ure
1(b).
This
pressure
domain
is
known
as
the
negacuring time and method, temperature, mix additives,
tive
phase.
The
front
of
the
blast
wave
weakens
it
etc.). In the literature various formulations canas be
progresses
outward the
and sorption
its velocity
drops toward
the
found to describe
isotherm
of normal
velocity
of
sound
in
the
undisturbed
atmosphere.
concrete (Xi et al. 1994). However, in the present
The the
characteristics
of a blast
wave resulting
paper
semi-empirical
expression
proposedfrom
by
an
explosion
depend
mainly
on
the
physicalit
Norling Mjornell (1997) is adopted because
Proceedings of FraMCoS-7, May 23-28, 2010
explicitly accounts for the evolution of hydration

reaction and SF content. This sorption isotherm
reads
we h c s = G c
Path of shock propagation to point

(other
)
(
,
,
than
the normal point ,
1
incidence,
1
W Angle of
s )1
10(g
c R)hA
Normal
1 c distance,
e
Horizontal
distance
( ,
K c
1
10(g
1 c
)Shelter
s e
(4)
Path of shock propagation
normal
) to surface
c h
(a) Spherical free air blast
surface
Ground
1
where the first term (gel isotherm) represents the

physically bound (adsorbed) water and the second
term (capillary isotherm) represents the capillary
water. This expression is valid only for low content
of SF. The coefficient G1 represents the amount of
water per unit volume held in the gel pores at 100%
relative humidity, and it can be expressed (Norling
Mjornell 1997) as
Pso
Positive specific impulse, is
Pressure
J = D ( h,developed,
T ) h
recently
their blast resistant capacities
(1)
have never been studied. In order to properly and efficiently
incorporate UHSC
and RPCD(h,T)
into protective
The proportionality
coefficient
is called
design
scheme,
an
in-depth
research
on blast
resismoisture permeability and it is a nonlinear
function
tance
behaviorhumidity
on UHSC
RPC is Turgently
of the relative
h andand
temperature
(Baant
needed
at
this
time
(Kim
2009,
Zineddin
et
al.
2007).
& Najjar 1972). The moisture mass balance requires
Therefore,
in
this
study,
the
blast
tests
are
that the variation in time of the water mass per perunit
formed
to concrete
investigate
the content
behavior
volume of
(water
w) of
be UHSC
equal toand
the
RPC
slabs subjected
to blast
load.
divergence
of the moisture
flux
J Blast wave characteristics including incident and reflected pressures
as well
as maximum and residual displacements and
(2)
w = in steel
J
strains
and concrete surface are measured.
t
Also, blast damages and failure modes were recorded
for The
eachwater
specimen.
From
these
UHSCasand
RPC
content
w can
betests,
expressed
thecomsum
are
shown
to
effectively
resist
blast
explosions
of
the
evaporable
water
we (capillary water, water
pared
toand
normal
strength
concrete.
on these
vapor,
adsorbed
water)
and
the Based
non-evaporable
test
results,
the
blast
design
procedure
will
sug(chemically
bound)
water
wn (Mills be 1966,
gested.
Pantazopoulo & Mills 1995). It is reasonable to
Negative specific impulse, is
Ps
tA
Ambient, Po
tA+to+to
tA+to
Pso
Ps
phase
c Positive
G ( c s ) = k vg
c
+ k s t s
Duration,
c vg s
,
(5)
Negative phase
Duration, to
(b) Pressure-time history

Figure 1. cSphericalsfree air blast (TM5-1300 1990, Kim et al.
where k vg and k vg are material parameters. From the
2007).
maximum amount of water per unit volume that can
fill all poresof(both

pores
gel pores),
one
properties
the capillary
source and
theand
medium
through
obtains To create reference
can calculate
K1 as one
which
blast waves
propagate.
blast experiments, some controlled explosions have
To relate
been conducted under ideal conditions.
g h
other explosions
toc the
e
w with
s + non-ideal
s G conditions
refc
s
em
erence explosions, blast scaling
laws
can
be
(6)
( ) =
K
ployed.
used approach to blast
c s The most widely
g h
c Hopkinson,
wave scaling is that formulated
which
e c by
is commonly described as the cube-root scaling law.
TheThe
scaled
distance,
Z, is defined
ksvg the
and Hopking1 can
material
parameters
kcvg andusing
son-Cranz's
cube
root
law
as
(ASCE
1999):
be calibrated by fitting experimental data relevant to
10
0.188
0.22
10
free (evaporable)
water
content in concrete at
1/ 3
1/ 3
(1)
Z = R / Eages
or(Di
Z Luzio
= R / W&
various
Cusatis 2009b).
where, Z is scaling distance; R is stand-off distance

2.2 Temperature evolution
from the target structure; E is total explosive thermal
Note that,
early W
age,is since
chemical
reactions
amount
of at
energy;
chargetheweight
of equivalent
associated
with The
cement
hydration
and isSFused
reaction
TNT
amount.
scaling
distance
for
are
exothermic,
the
temperature
field
is
not
uniform
evaluation of blast wave characteristics.
for non-adiabatic systems even if the environmental

temperature
istrends
constant. Heat conduction can be
2.2
Research
described
in concrete,
at least for temperature not
exceeding
100C
(Baant
& have
Kaplan
1996),high
by
Concrete is generally known to
a relatively
Fouriers
law, which
reads
blast
resistance
capacity
compared to other construction materials. However, concrete structures, which
q = not
T designed to have blast protective capacity,
were
(7)
require retrofitting during their service life to improve
resistance
against
where their
q is
the heat
flux,blastT loads.
is theRetrofitting
absolute
method
of
attaching
extra
structural
members
or
temperature, and is the heat conductivity;
in this
supports to increase the blast resistance is inefficient
= D ( h , T ) h
in the perspective of additional construction cost and

eliminating useable space. Also, since this method
does not greatly improve the overall structural resistance against blast load, a more feasible method of
retrofitting to improve blast resistance would be to
use Ultra High Strength Concrete (UHSC) or Reactive Powder Concrete (RPC). UHSC and RPC
would also be very effective in new constructions
since they can be used for concrete materials in reinforced concrete members.
In fact, beams and plates constructed using high
strength concrete (HSC) showed better impact resistance capacity than ones made using normal strength
concrete (NSC) in past researches. However, due to
social and governmental constraints, this type of
comparison study has not been carried over to blast
resistance capacity study, resulting in insufficient database of HSCs role as blast resisting material (Kim
2009).
Recently, several researchers have pursued static
and impact capacity studies on fiber reinforced concrete members under time-dependent loading conditions. The reference study has shown that the impact
and blast loaded UHSC or RPC study results are
non-existing and blast loaded HSC study results are
scarcely existing at best (Habel et al. 2008).
Figure 2. Overview blast setup.
3 BLAST TEST DETAILS
( D details
+ and
3.2 Specimen manufacturing

h h) =
h t
In this paper, the failure behaviors of reinforced

UHSC and RPC slabs under blast loading are studied. The tests were performed as 2 step process of
preliminary and main tests at Agency for Defense
Development of Koreas testing sight. In the preliminary test stage, TNT 35lbs was used as blast load on
control specimens (NSC specimen). After the trial
tests, ANFO 35lbs was selected as the blast explosive charge to be used for the main test stage.
3.1 Blasting test setup
In this study, in order to eliminate the 3-D effect, RC
slab specimens are placed at a same level as ground
surface(Razaqpur et al. 2007). A steel frame is constructed and buried in the ground as shown in Figure
2(a). For preventing the supporting frame distortion
during blast loading, the stiffeners with 250mm spacing are installed on wall surface of supporting frame.
Rubber pads of the same width and length as the
steel angle legs were placed between the angles and
test specimen to ensure uniform support conditions.
The explosive used for the test was spherical ANFO,
which was held by wooden horizontal bar. Figure
2(b) shows the test specimen setup with the 35lbs
ANFO (28.7lbs TNT) explosive charge. The 1.5m
standoff from specimens to explosive middle point is
consistently maintained.
The proportionality coefficient D(h,T)

moisture permeability and it is a nonlinea
of the relative humidity h and temperature
& Najjar 1972). The moisture mass balanc
that the variation in time of the water mas
volume of concrete (water content w) be eq
divergence of the moisture flux J
= supporting
J
(a) Buried
frame
t
w
The water content w can be expressed a

of the evaporable water we (capillary wa
vapor, and adsorbed water) and the non-e
(chemically bound) water wn (Mil
Pantazopoulo & Mills 1995). It is reas
assume that the evaporable water is a fu
relative humidity, h, degree of hydration
degree of silica fume reaction, s, i.e. we=w
= age-dependent sorption/desorption
(Norling Mjonell 1997). Under this assum
by substituting Equation 1 into Equati
(b) Explosive
obtainscharge and specimen
e & + we & + w
c
s
c
s
For the relative and absolute comparisons between
the specimens casted with UHSC, RCP, and NSC
where w /h is the slope of the sorption/
RC slabs with the edimensions of 1,000
w h
e
isotherm (also called moisture capac
1,000150mm and D10 (71.33mm2) mesh type reingoverning equation (Equation 3) must be
forcements with 82mm spacing are used. The steel
by appropriate boundary and initial conditi
ratio of the reinforced NSC and UHSC specimens is
The relation between the amount of e
same as the 2 volume % of short steel fibers used in
water
and relative humidity is called
RPC specimen. The mix proportions for NSC,
isotherm
if measured with increasing
UHSC, and RPC are tabulated in Table 2, 3, and 4,
humidity
and
desorption isotherm in th
respectively. The 100200mm cylindrical specimens
case. Neglecting their difference (Xi et al.
are prepared for compressive and tensile strength
the following, sorption isotherm will be
tests performed at Hyundai Institute of Construction
reference to both sorption and desorption c
Technology. The number of specimens tested for
By the way, if the hysteresis of the
NSC, UHSC, and RPC are 2, 4, and 4 specimens, reisotherm would be taken into account, two
spectively. The average compressive strength of
relation, evaporable water vs relative humi
NSC, UHSC, and RPC are 25.6, 202.0, and 203.0
be used according to the sign of the varia
MPa, respectively. The compressive strengths with a
relativity humidity. The shape of the
deviation over 15% are eliminated from consideraisotherm for HPC is influenced by many p
tion. The tensile strength of RPC is approximately
especially those that influence extent and
2.3 times greater (21.4MPa) than NSC (2.2MPa) and
chemical reactions and, in turn, determ
UHSC, (9.21MPa), respectively, due to the addition
structure and pore size distribution (waterof 2 vol.% of short steel fibers in RPC.
ratio, cement chemical composition, SF

curing time and method, temperature, mix
In the literature various formulatio
3.3 Measurementetc.).
outline
found
to describe
sorption
isotherm
The free field incident pressure
was the
measured
at 5m
concrete
al. 1994).where
However,
from the center of
the test(Xi
slabetspecimens
rein th
paper
the
semi-empirical
expression
pro
flected pressure on concrete specimen was measure
Norling
Mjornell
(1997)
is
adopted
b
at the center of the top surface of the specimen and
J = 2.
D (Mix
h, T )proportion
h
Table
of normal strength concrete (NSC).(1)
Max. Size of Target
reaction
isotherm
Binder
(kg) and SF content.
FA (kg) This sorption
Slump
W/B S/a
CA
AE
Water (kg) reads
Coarse Aggre- Strength
(mm)
(%)
(%)
(kg)
(kg)
The
proportionality
coefficient
D(h,T)
is
called
Cement
Fly-ash
S1
S2
(MPa)
gate (mm)
moisture permeability
and
nonlinear
25
24
100 it is a49.8
47.7 function
163
294
33
616
264
957
2.45
of the relative humidity h and temperature T (Baant
& Najjar
Therange
moisture
mass
Table
3. Mix1972).
proportion
of Ultra
Highbalance
Strengthrequires
Concrete (UHSC).
we (h c s ) = G ( c s )
+
that (%)
the variation
water(kg)mass perBinder
unit (kg)
W/B
S/ain
(%)time of theWater
FA (kg)
CA (kg) (g AE
(%)
c )h
c 1~3
e
of concrete
w) be equal <to1300
the
<volume
20
< 39.1(water content
< 140
< 450
< 700
(4)
divergence of the moisture flux J
)h
(g
1
460 mm
1000 mm
30
D10
460 mm
30
150
LVDT (Behavior displ.)

LVDT (Residual displ.)
Pressure transducer
Strain Gauge
Accelerometer
230 mm
1000 mm
w h
e + ( D h) = we
h
h t
500 mm
w
&c + e &s + w&n
c
s
100 mm
100 mm
1000 mm
(3)
500 mm
230 mm
where we/h is the slope of the sorption/desorption

isotherm (also called moisture capacity). The
governing equation (Equation 3) must be completed
Gauges on and
concrete
surface
by appropriate(b)boundary
initial
conditions.
Figure
3.
Location
of
measuring
sensor.
The relation between the amount of evaporable
230mm
from
center. (e.g.,
point of specimen
isotherm
if the
measured
with 1/3
increasing
relativity
diagonal
length).
To
measure
strain,
6mm
strain
humidity and desorption isotherm in the
opposite
gauges
are attached
on difference
reinforcing(Xi
steel
case. Neglecting
their
et atal.tensile
1994),rein
gion
and
30mm
strain
gauges
are
attached
on conthe following, sorption isotherm will be used
with
crete
top and
bottom
surfaces
shown Figure
3. In
reference
to both
sorption
and as
desorption
conditions.
case
of
retrofitted
specimen,
FRP
strain
gauges
are
By the way, if the hysteresis of the moisture
attached
gauges
bottom
isotherm instead
would of
be concrete
taken intostrain
account,
twoondifferent
surface.
Also,
LVDTs
on
the
specimen
center
are
used
to
measure
the
maximum
and
residual
displacebe used according to the sign of the variation of the
ments.
relativity humidity. The shape of the sorption
isotherm for HPC is influenced by many parameters,
4chemical
BLAST reactions
TEST RESULTS
and, in turn, determine pore
structure and pore size distribution (water-to-cement
UHSC
and RPCchemical
RC slabs composition,
are blast loadedSFto content,
analyze
ratio, cement
their
resistance
performance.
In
the
preliminary
testcuring time and method, temperature, mix additives,
ing
stage,
NSC
RC
slab
was
tested
to
estimate
etc.). In the literature various formulations can the
be
blast
behavior
and the isotherm
required of
explosive
foundcracking
to describe
the sorption
normal
charge
main tests.
concreteweight
(Xi etforal.the1994).
However, in the present
paper the semi-empirical expression proposed by
Norling Mjornell (1997) is adopted because it
10
c c+ ks s
G ( c s ) = k vg
c vg s
(5)
where kcvg and ksvg are material parameters. From the

fill all4.pores
(bothscene
capillary
pores
and gel pores), one
Figure
Explosive
by ANFO
35lbs.
can calculate K1 as one obtains
0.30
Free field pressure (MPa)
0.25
500 mm
<BOTTOM>

Mjornell 1997) as
0.20
g c c h
ePressure_ RPC2
s GFreefield
s
Incident
pressure_ConWEP
Freefield Pressure_UHSC2
1
10
s + 0.22
0.188
K ( c s )0.15
=
10
0.10
g c c h
(6)
The material parameters kcvg and ksvg and g1 can

free (evaporable) water content in concrete at
various ages (Di Luzio & Cusatis 2009b).
0.05
0.00
-0.05
10
20
30
Time (msec)
40
(a) Free field pressure

Note that, at early age, since the chemical reactions
associated with cement hydration and SF reaction
are exothermic, the temperature field is not uniform
temperature is constant. Heat conduction can be
described in concrete, at least for temperature not
exceeding 100C (Baant & Kaplan 1996), by
Fouriers law, which reads
30
25
Reflect pressure (MPa)
<TOP>
230 mm
1~3
200kg ~
230 mm
230 mm 230 mm
c
c
K ( c s )e
Filler (2.2~200)
(%)
Admixture (%) Steel Fiber
Table 4. Mix proportion range of Reactive Powder Concrete (RPC).

W/B
Cement (kg) Water (kg) Silica Fume (%) FA (kg)
w(%)
(2) 800~1000
= J< 800
<20
> 200
10~30
t
The water content w can be expressed as the sum

of the evaporable water we (capillary water, water
vapor, and adsorbed water) and the non-evaporable
(chemically bound) water wn (Mills 1966,
(Norling Mjonell(a)1997).
Undergauge
this assumption and
Steel strain
by substituting Equation 1 into Equation 2 one
obtains
10
Pressure_ConWEP
20
Reflect Pressure_UHSC2
Reflect Pressure_ RPC2
15
10
5
0
= T-5
0.0
0.5
1.0
1.5
Time (msec)
2.0
2.5
3.0
(7)
where q is the(b)heat
T is the absolute
Reflectflux,
pressure
temperature,
andpressure
is the
heat conductivity;
in this
Figure
5. Measured
on specimens
(ANFO 35lbs).
4.1 Blasting tests

When ANFO 35lbs was used as the explosive charge,
extreme wave of high pressure, temperature, noise,
and energy dispersed out radially. The photos in Figure 4 are ANFO 35lbs detonation photos. Since
ANFO detonation produces debrisless explosion, giving a more of pure pressure type of explosion loading, ANFO explosive charge is used for the main
tests.
4.2 Measured blast pressure results
Due to the exploded metal debris of TNT steel container impacting and damaging the pressure gauge installed in the centertop surface of the specimen, the
compressive blast pressure data was not obtained in
the preliminary stage. The measured free field and reflected pressures of ANFO 35lbs are shown in Figure
5. And the other data are tabulated in Table 6. The
measured data are inconsistent due to the variations
in experimental and environmental conditions (i.e.,
charge shape, charge angle, wind velocity, humidity,
etc.). However, the obtained blast pressure data seem
to agree well with ConWEP data.
4.3 Tested specimen examination
When the testing is completed and the safety is insured for the inspectors, the surface examination of
the specimen was performed. Figure 6(a), 6(b), and
6(c) are the schematic drawings of NSC, UHSC, and
RPC slab bottom surface crack distributions after
ANFO 35lbs blasting, respectively. The NSC speci-
J = D ( h,turtle
T )h back type crack patmens had a well dispersed
tern. The crack lines followed the cone prism type of
plastic yield line from
the proportionality
center to the 4 corners,
in- D(h,T)
The
coefficient
dicating a 2D membrane
plastic
failure
mode.
Howmoisture permeability and it is a nonlinea
ever, UHSC specimens
crack pattern
showed
of the relative
humidity
h andmostly
temperature
macro-cracks concentrated
near
or
on
the
yield
& Najjar 1972). The moisture lines.
mass balanc
The RPC specimens
predominantly
di- mas
that showed
the variation
in time ofone
the water
rectional, center bisecting
type,
macro-cracks.
Since
volume of concrete (water content w) be eq
RPC specimen isdivergence
made using
cement
mortar
of the
moisture
fluxwith
J
short fibers, it tended to be brittle but the crack
bridging effect of short
fibers resisted crack propaga w = J form only in the direction where the macro-cracks
t
tion perpendicular to the principle tensile strain direction as shown in Figure
The6(c)
water content w can be expressed a
of the evaporable water we (capillary wa

vapor, and adsorbed
water)
4.4 Deflection measurements
from blast
tests and the non-e
(chemically
bound)
water
wn (Mil
The incidental and
residual deflections
are1995).
measured
Pantazopoulo
&
Mills
It
is reas
from the blast test.
Both that
deflection
results of water
maxi- is a fu
assume
the
evaporable
mum and residualrelative
measurements
are htabulated
inofTahumidity,
, degree
hydration
ble 7. In the preliminary
tests
using
TNT
35lbs,
the
degree
of
silica
fume
reaction,
s, i.e. we=w
maximum measured
at the sorption/desorption
center of the
= deflection
age-dependent
specimen was beyond
25mm
measurement
capacity
(Norling
Mjonell
1997).
Under
this assum
of the LVDT. The
specimen
center
deflection-time
by substituting
Equation 1 into Equati
histories for NSC-TNT
35lbs, which exceeded
obtains
LVDT measuring capacity, and NSC-ANFO 35lbs
are shown in Figure 7. As shown in Table 7, the

w
w
w h
maximum and residual
deflections
from ANFO
e & 35lbs
e & + w
+ ( D h ) =
e
c +
s
h 18.57mm
for NSC, UHSC, hand
and
t RPC are
c
s
9.03mm, 12.83mm and 3..86mm, and 11.91mm and
4.31mm, respectively. In case of maximum retrofit
/h ismore
the slope
of thethan
sorption/
where wehave
effect, the RPC (35.85%)
effective
isotherm
(also
called
moisture
UHSC (30.90%). But in case of residual effect, the capac
equation
(Equation
3) must be
UHSC (57.23%)governing
have more
effective
than RPC
by appropriate boundary and initial conditi

The relation between the amount of e
Table 6. Measured blast pressure.
water
andRPC1
relative humidity
SPECIMEN
ConWEP
NSC2
UHSC1
UHSC2
RPC2 is called
isotherm
if measured
with35lbsincreasing
Charge
ANFO 35lbs
ANFO 35lbs ANFO 35lbs ANFO
35lbs ANFO
35lbs ANFO
Environment Temp.
5
8
NR humidity -9
NR isotherm in th
and desorption
Humid (%)
up 51
56
NR case. Neglecting
39
their NR
difference (Xi et al.
Reflect Pres- Center (MPa)
17.02
NR
NR
16.92the following,
NR sorption
21.99
isotherm will be
sure
Impulse (MPa-msec) 2.42
NR
NR
3.87
NR
2.83
reference
to
both
sorption
230mm (MPa)
16.53
26.58
NR
18.76
22.62
22.1and desorption c
By
the
way,
if
the
hysteresis of the
3.26
NR
3.02
2.03
22.41
be taken
into account, two
Free Field
Peak overpressure 0.170
0.161
0.249
0.191isotherm would
0.16
0.191
Pressure
0.23
0.191
0.23relation, evaporable
0.229
0.21 vs relative humi
water
* NR : Not Record
* NSC : Normal strength concrete(control specimen)be used according to the sign of the varia
* UHSC : Ultra High Strength Concrete
* RPC : Reactive Powder Concrete
relativity humidity. The shape of the
isotherm for HPC is influenced by many p
especially those that influence extent and
chemical reactions and, in turn, determ
structure and pore size distribution (waterratio, cement chemical composition, SF
curing time and method, temperature, mix
etc.). In the literature various formulatio
found to describe the sorption isotherm
concrete (Xi et al. 1994). However, in th
paper the semi-empirical expression pro
Norling Mjornell
(a) NSC
(b) UHSC
(c) RPC (1997) is adopted b
NSC2
UHSC1
RPC2
Figure 6. The crack pattern of blasted specimens (bottom side).
= D ( h , T ) h
(1)
The proportionality coefficient D(h,T) is called

moisture permeability and it is a Exceed
nonlinear
function
Measurement
Capacity
of the relative humidity h and temperature
T (Baant
& Najjar 1972). The moisture mass balance requires
that the variation in time of the water mass per unit
volume of concrete (water content w) be equal to the
divergence of the moisture
flux J
NSC TNT, ANFO
0

reaction and SF content. This sorption isotherm
reads
NSC_ANFO 35 lb
2000
1500
Displacement (mm)
-5
-10
-15
N SC_T N T 35lbs
Acceleration (g)
1000
we (h c s ) = G c s
,
,500
-25
-30
50
10 0
15 0
T im e (m sec )
(a) NSC

relative humidity, h, degree
hydration, c, and
Residualof
Displacement
degree of silica fume
reaction,
s, i.e. we=we(h,c,s)
Maximum
Displacement
= age-dependent sorption/desorption
isotherm
ANFO 35lb, UHSC
obtains
(b) UHSC
10
5
UHSC_ 1
Displacement (mm)
UHSC_ 2
-5
-10
-15
-20
50
100
150
(2)
2 00
200
10
(a) NSC
(4)
115

UHSC 2the capillary
term (capillary isotherm) represents
water. This expression
is
valid
only
for low content
ANFO 35lb, UHSC
amount of
of SF. The coefficient
G1 represents
Maximum acceleration
, UHSC1 = Not the
Reported
Maximum acceleration,
= 1420pores
g
water per unit volume
held inUHSC
the 2gel
at 100%
Mjornell 1997) as
2000
1600
Acceleration (g)
= J
10
-500
c )h
(g
c
e
)h
(g
c
c
K 3( c s )6 e
9
12
Time (msec)
1
N SC_A N F O 35lbs
-20
ANFO 35lb, NSC

1
( Maximum
) 1acceleration = 1815.5 g
,
1200
800
400
0
c c+ ks s
G ( c s ) = k vg
c vg Time
s (msec)
-400
12
(5)
15
Time (ms ec)
3500
&c + e &s + w&n
c
s
RP C_1
RP C_2
(3)
3000
2500
Acceleration (g)
Displacement (mm
w
w h 10
e
( D h ) =
+
e
h
h t 5
UHSC
parameters. From the
where kcvg and ksvg are(b)material
fill all pores (both capillary pores andRPC
gel1 pores), one
RPC 2

by appropriate boundary andANFO
initial
35lb,conditions.
RPC
(msec) increasing relativity
isotherm if measured Time
with
(c) RPC
humidity and desorption
isotherm in the opposite
Figure
7. Displacement
of concrete
center
case. Neglecting
theirbehavior
difference
(Xi etspecimen
al. 1994),
in
point under blast loading.
the following, sorption isotherm will be used with
reference to both sorption and desorption conditions.
(52.29%).
ByAstheshown
way,in ifTable
the 7,hysteresis
the concrete
moisture
the bottomofcenter
isothermwere
would
be taken
into account,
strains
over
16,000
for NSCtwo
anddifferent
UHSC
relation,
evaporable
water
vs
relative
humidity,
must
specimens. However, when the strain measurements
be used according to the sign of the variation of the
Table
7. Measured
blast
results. by many parameters,
isotherm
for HPC
is test
influenced
SPECIMEN
NSC2
especially those that influenceNSC1
extent and rate
of the
Charge
TNT
35lbs
ANFOpore
35lbs
chemical
reactions
and,
in
turn,
determine
Max. displacement (mm)
Over 25
18.565
structureof and
pore size distribution
Average
max disp.(mm)
18.565 (water-to-cement
ratio, Effect
cement
SF content,
Retrofit
(%) chemical composition,
curing time
and method,
mix 5.790
additives,
Residual
displacement
(mm) temperature,
12.260
Average
residual
disp.(mm)
etc.). Inof the
literature
various9.025
formulations can be
Retrofit
(%)
- isotherm of normal
found Effect
to describe
the sorption
Strain
Steel
up
16012
concrete (XiSteel
et al.
1994). However,
in the5964
present
bottom
15998
28113
paper the semi-empirical
expression
proposed
Concrete up
NR
11848 by
Norling Mjornell
(1997)
is
adopted
because
it
Concrete bottom 16007
NR
-5
-10
-15
50
100
150
* NR : Not Record
200
ANFO 35lb, RPC
1500
1000
500
Maximum acceleration
g
, RPC1=
2844.6
10

RPC
Maximum acceleration,
2=1 2465.2 g
s + 0.22 s G
0.188
10
-1000
-1500
0
1
1
)=
K ( c s -500
-20
2000
g c c h
g c
c h
(6)
c
s
Timek(msec)
The material parameters
vg and k vg and g1 can
RPC
be calibrated by fitting(c)experimental
data relevant to
Figure
8.
Specimen
acceleration
under
blast
loading.
and displacements for NSC, UHSC, and RPC specimens are compared, RPC data at the specimen center
2.2 Temperature
evolution
tend
to be less than
those of NSC and UHSC specimens.
It
means
RPC
specimens
have
more blast
resisNote that, at early age,
since the
chemical
reactions
tance
capacity
than
others.
This
result
is
probably

UHSC1
UHSC2
RPC1 conduction
RPC2can be
temperature is
constant. Heat
ANFO 35lbs ANFO 35lbs ANFO 35lbs ANFO 35lbs
described in 15.14
concrete, at least
for temperature
not
10.517
10.73
13.09
exceeding
100C
(Baant
&
Kaplan
1996),
by
12.829
11.910
Fouriers law, which reads 35.85
30.90
1.860
5.86
3.202
5.41
3.860
4.306
q = T
(7)
57.23
52.29
2796
2832
where q is 7553.6
the heat flux,
T is the
absolute
6711
4502
24214 in this
temperature, 12821
and is the 11198
heat conductivity;
16025
18081
NR
4903
* - : Non-attached gauge
due to the short steel fiber reinforcing in RPC specimen where the fibers restrained crack opening by
crack bridging and controlling effect.
4.5 Acceleration measurements from blast test
Generally, specimen blast behavior can be analyzed
based on data obtained from LVDT and accelerometer. If LVDT data are unusable or imprecise, specimen acceleration data can be alternatively used. Figures 8(a)~(c) show the specimen acceleration
measurements for NSC, UHSC, and RPC specimens.
As shown Figures 8(a)~(c), the accelerations ranging
from 1,000~2,500g have occurred. However, these
acceleration measurements are combined values of
both specimen acceleration as well as the impulse acceleration. Also, for UHSC 1 specimen, the sensor
was detached from concrete surface when blast pressure was applied, causing imprecise noise. Therefore,
the data were considered unfit for analysis. Also,
RPC specimens have extreme high frequency vibration due to no reinforcement. Therefore, the reinforcement can be affect to specimen behavior under
blast load.
4.6 Blast design and analysis process
Based on the blast tests for NSC, UHSC, and RPC in
this study, the blast design and analysis process are
suggested. Most importantly the building and owner
requirements are needed for determination of blast
resistance capacity of a targeted structure. To evaluate the building requirements, the blast loading on
each component and resistance capacity can be derived from test results or research reports. If materials and structural system are selected, determination
of deformation limit using analysis method such as
HFPB (High Fidelity Physics Based) and SDOF,
MDOF, etc. is selected for the blast analysis. The details of design will be accompanied with satisfaction
of the deformation limit based on the analysis results.
Owner Specs & Standards
Explosion Hazard
Identification
Standard
Practice
Building Requirements :
Criticality, Occupancy, Layout,
Siting, etc.
Site
Study
New
Building Performance
Requirements
Free Field Blast

Parameter
Component Loads
Analysis Method
HFPB (High Fidelity
Physics Based)
SDOF
Existing
MDOF
Select Materials & Structural

System
Material
Properties
Preliminary Member Sizes
Deformation
Limits
Structural Analysis
Is
Deformation Limit
Satisfied?
Comparing with
BLAST TEST results
Foundation Design
Yes
Details & Documentation
Figure 9. Blast design and analysis process.
No
Ancillary Design
5 CONCLUSIONJ = D ( h, T )h
From this study, Ultra

High Strengthcoefficient
Concrete D(h,T)
The proportionality
(UHSC) and Reactive
Powder
Concrete
(RPC)
moisture permeability and it is RC
a nonlinea
slabs response induced
by explosive
of hblast
of the relative
humidity
and wave
temperature
pressure are evaluated
to understand
blast resis& Najjar
1972). The the
moisture
mass balanc
tance capacity blast
resisting
repair
materials
that the variation in time ofand
theretwater mas
rofitted structure.volume
The reflected
blast
pressure
andw) be eq
of concrete
(water
content
impulse values calculated
using
the moisture
ConWEPflux
were
divergence
of the
J in
reasonable agreement with the experimental data.
The performance comparison
of UHSC and RPC
wcontrol
= J specimens subjected to
specimens to NSC
t
blast loads of ANFO 35 lbs has shown the high blast
resistance capacity The
of about
30.9~35.9%
water
content
w canincrease
be expressed
a
with respect to average
maximum
displacement.
An
wa
of
the
evaporable
water
we (capillary
average of residual
displacements
waswater)
smaller
vapor,
and adsorbed
andthan
the non-e
normal strength (chemically
concrete specimens
residual
disbound)
water
wn (Mil
placement, even Pantazopoulo
though there &was
no 1995).
consistent
Mills
It is reas
trend due to variations
in
environmental
conditions.
assume
that
the
evaporable
water
is a fu
Therefore, to evaluate
the
damage
under
blast
load,
relative
humidity,From
h, degree of hydration
failure mode mustdegree
be considered.
the test re, i.e. we=w
of
silica
fume
reaction,
sults, the failure patterns
of both UHSCsorption/desorption
and RPC sin=
age-dependent
dicate that they (Norling
are muchMjonell
more resistant
to blast
1997).
Under
this assum
loading and have by
higher
blast
resistance
capacity
than
substituting Equation 1 into Equati
NSC.
obtains
w
w h
ACKNOWLEDGMENTS
e & + we & + w
+ ( D h ) =
e
c
s
h t
c
s
The research was supported the financial support
provided KOSEF (Korea Science and Engineering
/h High
is theToughness
slope of theand
sorption/
where weof
Foundation, Development
isotherm
(also and
called
moisture capac
High Ductility Polyurea
for Repair
Rehabilitation
governing
equation
(Equation
3) must be
of Structure of General Concrete and be Impressed
by
appropriate
boundary
and
initial
by Load of Impact and Explosion) and KOGAS fromconditi
The relation
theAffairs
amount of e
Ministry of Land, Transport
andbetween
Maritime
water
and
relative
humidity
called
(Design standard of extremely large storage tankisand
isotherm
if
measured
with
increasing
optimum analysis technique).
humidity and desorption isotherm in th

case. Neglecting their difference (Xi et al.
REFERENCES the following, sorption isotherm will be
reference to both sorption and desorption c
By Design
the way,
if theSecurity:
hysteresis
ASCE, 1999. Structural
for Physical
State ofof the
isotherm
would beontaken
into
account, two
the Practice Report.
Task Committee
Physical
Security,
American Societyrelation,
of Civil Engineers,
Newwater
York.vs relative humi
evaporable
Baker, W.E., 1973. be
Explosions
in Air. Wilfred
Engiused according
to theBaker
sign of
the varia
neering, San Antonio.
relativity
humidity.
The
shape
of the
Harbel, K., and Gauvreau, P., 2008. Response of Ultra-High
isotherm for HPC is influenced by many p
Performance Fiber Reinforced Concrete (UHPFRC) to Imespecially
those
influence
extent and
pact and Static Loading.
Cement
andthat
Concrete
Composites
30 : 938-946. chemical reactions and, in turn, determ
Kim, H.J., Nam, J.W.,
Kim, S.B.,
Byun, K.J., (waterstructure
andKim,
poreJ.H.J.,
sizeand
distribution
2007. Analytical ratio,
Evaluations
of
the
Retrofit
Performances
cement chemical composition, SF
of Concrete Wallcuring
Structures
Subjected
to Blast temperature,
Load, Jour- mix
time
and method,
nal of the Korea Concrete Institute 19(2) : 241-250.
etc.). In theEvaluation
literatureforvarious
Kim, J.H.J., 2009. Experimental
the Blast formulatio
Refound
to
describe
the
sorption
isotherm
sistance Capacity of Concrete Structure using Ultra High
concrete
et al. 1994).
Strength Concrete.
Report (Xi
of Hyundai
InstituteHowever,
of Con- in th
struction Technology.
paper the semi-empirical expression pro
Mays, G.C., and Smith,
P.D., 1995.
Blast effect
on Buildings:
Norling
Mjornell
(1997)
is adopted b
Design of Buildings to Optimize Resistance to Blast LoadProceedings of FraMCoS-7, May 23-28, 2010
J =ing.
DThomas
( h, T )hTelford.
(1)
Razaqpur, A.G., Tolba, A., and Contestabile, E., 2007. Blast
Loading Response of Reinforced Concrete Panels ReinThe proportionality
coefficient
D(h,T) isComposcalled
forced
with Externally Bonded
GFRP Laminates.
moisture
it is a nonlinear function
ite Part permeability
B : engineering and
38: 535-546.
TM5-1300/AFR
P-39,
1990. Structures
to Reof the relative88-2/NAVFAC
humidity h and
temperature
T (Baant
sist
the
Effects
of
Accidental
Explosions.
Joint
Depart& Najjar
1972). The moisture mass balance requires
ments of the Army, Air Force and Navy Washington, DC.
that
the
variation
in time of the water mass per unit
Yi, N.H., Kim, S.B., Kim, J.H.J, and Cho, Y.G., 2009. Behavvolume
of
concrete
(waterStructure
contentunder
w) beBlast
equal
to the:
ior Analysis of Concrete
Loading
divergence
of
the
moisture
flux
J
(I) Experiment Procedures, Journal of the Korea Society
Civil Engineering 29(5A) : 557-564.
= J

obtains
w h
e + ( D h) = we
h
h t
w
&c + e &s + w&n
c
s
(3)

by appropriate boundary and initial conditions.
isotherm if measured with increasing relativity
humidity and desorption isotherm in the opposite
case. Neglecting their difference (Xi et al. 1994), in
the following, sorption isotherm will be used with
reference to both sorption and desorption conditions.
By the way, if the hysteresis of the moisture
isotherm would be taken into account, two different
be used according to the sign of the variation of the
isotherm for HPC is influenced by many parameters,
chemical reactions and, in turn, determine pore
structure and pore size distribution (water-to-cement
ratio, cement chemical composition, SF content,
curing time and method, temperature, mix additives,
etc.). In the literature various formulations can be
found to describe the sorption isotherm of normal
concrete (Xi et al. 1994). However, in the present
paper the semi-empirical expression proposed by
Norling Mjornell (1997) is adopted because it
(4)
10(g
K1 ( c , s )e 1 c
(2)
Yi,
N.H., Kim,
S.B., Kim,
Cho, Y.G.,of2009.
Behavexplicitly
accounts
forJ.H.J,
theand
evolution
hydration
ior
Analysis
of
Concrete
Structure
under
Blast
Loading
reaction and SF content. This sorption isotherm:
(II) Blast Loading Response of Ultra High Strength Conreads
crete and Reactive Powder Concrete Slabs, Journal of the
Korea Society Civil Engineering 29(5A) : 565-575.
Zineddin, M., and Krauthammer,
T., 2007. Dynamic
Re
sponse and Behavior of Reinforced

Concrete
Slabs
1
under
(h, , ) = G ( , ) 1
weImpact
+ Engiof Impact
c Loading.
s
cInternational
s 10Journal
1
(g
c )h
neering 34 : 1517-1534.
1 c
c )h

Mjornell 1997) as
c c+ ks s
G ( c s ) = k vg
c vg s
(5)
where kcvg and ksvg are material parameters. From the

fill all pores (both capillary pores and gel pores), one
w
K ( c s ) =
,
1
1
s + 0.22 s G
0.188
10
10
g c c h
g c c h
1
(6)
The material parameters kcvg and ksvg and g1 can

Note that, at early age, since the chemical reactions
temperature is constant. Heat conduction can be
described in concrete, at least for temperature not
exceeding 100C (Baant & Kaplan 1996), by
Fouriers law, which reads
q
= T
(7)
where q is the heat flux, T is the absolute

temperature, and is the heat conductivity; in this

Ductile Concrete

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ductile Concrete

Uploaded by

Copyright:

Available Formats

Fracture Mechanics of Concrete and Concrete Structures High Performance, Fiber Reinforced Concrete, Special Loadings and Structural

Applications- B. H. Oh, et al. (eds)

Blast loading response of ultra high performance concrete and reactive

J.K. Choi & Y.G. Cho

Hyundai Institute of Construction Technology, Yongin, Korea

and high resistance to damage is a better material for

assume that the evaporable water is a function of

explicitly accounts for the evolution of hydration

Path of shock propagation to point

Path of shock propagation

(a) Spherical free air blast

where the first term (gel isotherm) represents the

Positive specific impulse, is

Negative specific impulse, is

(b) Pressure-time history

maximum amount of water per unit volume that can

fill all poresof(both

where, Z is scaling distance; R is stand-off distance

for non-adiabatic systems even if the environmental

in the perspective of additional construction cost and

Figure 2. Overview blast setup.

3 BLAST TEST DETAILS

3.2 Specimen manufacturing

In this paper, the failure behaviors of reinforced

The proportionality coefficient D(h,T)

The water content w can be expressed a

isotherm (also called moisture capac

ratio, cement chemical composition, SF

of the relative humidity h and temperature T (Baant

LVDT (Behavior displ.)

where we/h is the slope of the sorption/desorption

Proceedings of FraMCoS-7, May 23-28, 2010

where kcvg and ksvg are material parameters. From the

Free field pressure (MPa)

where the first term (gel isotherm) represents the

The material parameters kcvg and ksvg and g1 can

(a) Free field pressure

2.2 Temperature evolution

Reflect pressure (MPa)

Table 4. Mix proportion range of Reactive Powder Concrete (RPC).

The water content w can be expressed as the sum

4.1 Blasting tests

of the evaporable water we (capillary wa

are shown in Figure 7. As shown in Table 7, the

by appropriate boundary and initial conditi

Figure 6. The crack pattern of blasted specimens (bottom side).

Proceedings of FraMCoS-7, May 23-28, 2010

The proportionality coefficient D(h,T) is called

explicitly accounts for the evolution of hydration

The water content w can be expressed as the sum

where the first term (gel isotherm) represents the

ANFO 35lb, NSC

Time (ms ec)

&c + e &s + w&n

where we/h is the slope of the sorption/desorption

Proceedings of FraMCoS-7, May 23-28, 2010

ANFO 35lb, RPC

are exothermic, the temperature field is not uniform

Free Field Blast

Select Materials & Structural

Preliminary Member Sizes

Details & Documentation

Figure 9. Blast design and analysis process.

From this study, Ultra