Materials and Structures/Matdriaux et Constructions, 2 3 , 4 5 7 - 4 6 0

RILEMDRAFTRECOMblENDATIONS
PROJETSDE RECOMMANDATIONDE LA RILEM
=.,

TC89-FMTFRACTUREMECHANICSOFCONCRETE-
TESTMETHODS
MI~EANIQUEDE LARUPTUREDU BI~TON-.
Mt~THODESD'ESSAI

Determination of fracture parameters and CTODc)

of plain concrete using three-pointbend tests
The text presented hereunder are drafts which are being submitted to enquiry. Comments
should be sentto Prof. S. P. Shah, The Technological Institute, Northwestern University,
Evanston, Illinois 60208-3109, USA, before 1 July 1991.

CONTENTS
1. Scope 5. Test results and calculations
2. Specimens 6. Report
3. Apparatus 7. Background
4. Test procedure 8. Applicable documents

1. SCOPE A saw-cut notch or a pre-cast notch shall be made,

with the width of the notch less than 5 mm. The initial
This recommendation covers the determination of the notch-to-depth ratio (ao/d) shall be equal to 1/3. A
critical stress intensity factor K~c and the critical crack tip minimum of four specimens are required for each type of
opening displacement CTOD~ of mortar and concrete, material.
using three-point bend tests on notched beams. The If the maximum size of aggregate is larger than
critical stress intensity factor is defined as the stress 50 ram, the specimen dimensions shall be increased
intensity factor calculated at the critical effective crack proportionally and the span-to-depth ratio shall be kept
tip, using the measured maximum load. The critical equal to 4.
crack tip opening displacement is defined as the crack tip
opening displacement calculated at the original notch tip
of the specimen, using the measured maximum load and 2.2 Fabrication of specimens
the critical effective crack length. After casting, the specimens shall be covered with wet
The critical stress intensity factor and the critical crack burlap or kept in the curing room with 100% relative
tip opening displacement, along with the Young's humidity at 23 + 2~ for the first 24 h. On the second day
modulus E, arc sufficient to characterize the fracture all the specimens shall be transferred to the curing room
resistance and energy dissipation of concrete and until about 4 h before testing.
mortar.

3. APPARATUS
2. SPECIMENS
A closed-loop testing machine with the crack mouth
opening displacement (CMOD) as the feedback signal
2.1 Dimensions
or a relatively stiff machine is required to achieve a
The specimen dimensions as indicated in Fig. 1 are given stable failure. The crack mouth opening displacement
in Table 1 for aggregates not larger than 50 mm (2 in.). and the applied load shall be recorded continuously
0025-5432/909 RILEM
 458 RILEM Draft Recommendations

Table 1 Dimensions of specimens

Dm~• Depth, d Width, b Length, L Span, s

(mm) (mm) (mm) (mm) (mm)

1 to 25 150 + 5 80 + 5 700 + 10 600 + 5

25.1to 50 250+5 150+5 1100+10 1000_+5

P
applied load is reduced to zero, reloading is applied. The
rate of unloading and reloading does not have to be the
same as the initial loading rate, and each loading and
unloading cycle shall be finished in about 1 min. Only
one cycle of loading-unloading is required for the test.

5. TEST RESULTS AND C A L C U L A T I O N S

A typical test result (i.e. a l o a d - C M O D curve with
several loading-unloading cycles) is shown in Fig. 2.
Based on the l o a d - C M O D curve, the Young's modulus
E, the critical stress intensity factor KT~, and the critical
crack tip opening displacement C T O D c can be deter-
mined.

5.1 Young's modulus, E

Fig. 1 Testing configuration and geometry of specimen.
P = load, L = specimen length, S = specimen loading span, The Young's modulus E is calculated from the equation
d = beam depth, b = beam thickness, HO = thickness of
holder of clip gauge, a0 = initial notch depth, CMOD = crack E = 6S aoVl(a)/[Cid2b] [N m -2] (1)
mouth opening displacement. in which Ci is the initial compliance calculated from the
l o a d - C M O D curve (Fig. 2) [m N - l ] ; also

during the test. A clip gauge is r e c o m m e n d e d to measure 0.66

the C M O D . H o w e v e r , if a clip gauge is not available, an Vl(O~ ) : 0.76 - 2.28a + 3.87cr z - 2.04a 3 -~ (1 - a)2 (2)
L V D T can be used as a replacement. To avoid possible
errors caused by the bending effect, the gauge length of
where a = (a0 + H O ) / ( d + H O ) and S, no, H O , d and b
the L V D T shall be kept as small as possible. The C M O D
are defined in Fig. 1.
shall be measured at the centre of the notch to minimize
possible errors caused by eccentricity.
The supports,and the loading arrangements shall be
such that the applied forces acting on the beam are 5.2 Critical effective crack length, ac
statically determinate.
The critical effective crack length ac [m] (ac = ao +
stable crack growth at peak load) is determined from the
4. TEST P R O C E D U R E Young's modulus E calculated from Equation 1 and the
unloading compliance Cu measured at the maximum
4.1 Control of loading load (Fig. 2). Using an iteration process, the critical
effective crack length ac is found when Equation 3 is
The rate of loading shall be controlled by a constant rate
satisfied:
of increment of crack mouth opening displacement (or
load-line deflection) such that the peak load is reached in
about 5 rain. E - 6S ac V,(a) (3)
CudZb
4.2 Loading and unloading requirement in which ac = the critical effective crack length to be
The applied load is manually reduced (also termed determined [m t, ~ = (a + H O ) / ( d + H O ) , and
unloading) when the load passes the maximum load and Cu = the unloading compliance at 95% of peak load [m
is at about 95% of the peak load (Fig. 2). When the N-'].
 Materials and Structures 459

1000
}p .~_~mm
?

750
z
a.
"I[3
- 500 /i ,,L clip gauqe
0 1/~/'/ //~N controls ~-/~.V.,~
0
J

250

,"&" I ~''- ~ ~ it''- I

oo 2.5 5.0 7.5 10.0 1 2 . 5 15.0 1 7 . 5 20.0
Crack mouth opening d i s p l a c e m e n t , CMOD (10-2 mrn)

Fig. 2 Typical experimental load-CMOD plot. Mortar mix = C:S:A:W = 1.0:2.6:0.0:0.65.

5.3 Critical stress intensity factor, K~c 2. Complete load-CMOD curve, peak load Pmax,
initial compliance Ci and unloading compliance C~ for
The critical stress intensity factor is calculated using the each specimen.
equation 3. Young's modulus E, critical effective crack length
ac, critical stress intensity factor K~c, and critical crack
K~ = 3(Pmax + 0.5W) S(rrac)U2F(~ IN m-3/2l tip opening displacement CTODc for each specimen.
2d2b 4. Special events occurring during each test.
(4)
in which 7. BACKGROUND
1.99 - a(1 - a)(2.15 - 3.93a + 2.7a 2)
With these two parameters it is possible to predict the
F(c0 = X/rd/Z(1 + 2c0(1 - a) 3/2
maximum load of a specimen or a structure of any
where a = aJd, Pm~• -----the measured maximum load arbitrary geometry [1-3]. A crack of given length (a0,
[N], W = WoS/L [N], and Wo = self-weight of the beam which can be equal to zero) will critically propagate
when it reaches a critical effective crack length ac. To
[N].
determine ac two fracture parameters are needed: K~c
and CTOD c. The sub-committee on Mode I testing of
5.4 Critical crack tip opening displacement, CTODc R I L E M 89-FMT has examined test results for a large
The critical crack tip opening displacement is calculated number of notched beams of varying dimensions, and
concluded that these two parameters can be considered
using the equation
as valid material parameters [4].
6Pmax Sac V1(a) This testing method is unique in that all material
CTODc = EdZb properties (i.e. K~c, CTODc and E) can be determined
from a single test performed on a notched beam
[(1 - / 3 ) 2 + (1.081 - 1.149a)(/3 -/32)1~/2 [m] (5) specimen. As a result, the application of these material
parameters to the associated effective crack model (i.e.
in which a = aJd and fl = ao/a.
the two-parameter fracture model [1]) is self-consistent.
For those laboratories which cannot perform a stable
Using these two parameters and the Young's modulus, it
three-point bend test, Cu values can be approximately
is also possible to predict the critical load under mixed-
calculated by assuming that the unloading path will
mode loading conditions [2] as well as for com-
return to the origin. The values of K~c and CTODc
pact-tension and wedge-loaded cubical specimens [5].
determined based on this assumption are about 10 to
The critical crack extension obtained from these two
25% higher than the values calculated using the actual
parameters has been shown to decrease with an increase
unloading compliance.
in concrete strength and with an increase in rate of
loading [6, 7]. Thus, these two parameters well repre-
6. REPORT sent the fracture response of quasi-brittle materials like
concrete. A geometry-dependent R-curve approach can
The report shall include the following:
also be predicted by using these two parameters [8].
1. Specimen dimensions, weight of specimen, It should be noted that Llorca et al. [9] have
mix-proportion, number of specimens tested, and date shown that the size effects predicted by this proposed
of test. two-parameter fracture model are quite comparable with
460 RILEM Draft Recommendations

the results predicted by the accompanying proposed testing methods', in Fracture Toughness and Fracture
size-effect law. Energy - Test Methods for Concrete and Rock (edited
by H. Mihashi, H. Takahashi and F. H. Wittmann) (A.
A. Balkema Publishers, Rotterdam, Netherlands, 1989,
pp. 443-463).
REFERENCES 6. John, R. and Shah, S. P., 'Fracture mechanics analysis of
high strength concrete', J. Mater. Civil Engng, ASCE 1
1. Jenq, Y. S. and Shah, S. P., 'Two-parameter fracture (4) (1989) 185-198.
model for concrete', J. Engng Mech. Div., ASCE 111 7. John, R., Shah, S. P. and Jenq, Y. S., 'A fracture
(10) (1985) 122%1241. mechanics model to predict the rate sensitivity of Mode
2. Idem, 'Mixed-mode fracture of concrete', Int. J. Fract. 38 I fracture of concrete', Cement Concr. Res. 17 (2) (1987)
(1988) 123-142. 249-262.
3. Idem, 'On the fundamental issues of mixed-mode crack 8. Ouyang, C., Mobasher, B. and Shah, S. P., 'An R-curve
propagation in concrete', in Fracture of Concrete and approach for fracture of quasi-brittle materials', Engng
Rock - Recent Developments (edited by S. P. Shah, Fract. Mech. (in press).
S. E. Swartz and B. Barr) (Elsevier Applied Science, 9. Llorca, J., Planas, J. and Elices, M., 'On the use of
1989, pp. 2%38). maximum load to validate or disprove models of
4. Karihaloo, B. L. and Nallathambi, P., 'A notched beam concrete fracture behaviour', in Fracture of Concrete
test: Mode I fracture toughness', Final Report to and R o c k - Recent Developments (edited by S. P. Shah,
Sub-committee of RILEM TC89-FMT (April 1988). S. E. Swartz and B. Barr) (Elsevier Applied Science,
5. Jenq, Y. S. and Shah, S. P., 'On the concrete fracture 1989, pp. 35%368).