ACI STRUCTURAL JOURNAL TECHNICAL PAPER

Title no. 105-S59

Characteristics of Compressive Stress Distribution

in High-Strength Concrete
by Halit Cenan Mertol, Sami f!izkalla, Paul Zia, and Amir Mirmiran

This paper describes fundamental characteristics of the compressive unreinforced HSC members as well as a significant
stress distribution in the compression zone of flexural members amount of data found in the literature. Stress-strain curves
with concrete compressive strengths up to 18 ksi (124 MPa). The and stress block parameters for HSC were obtained, evaluated.
proposed model is based on testing of 21 plain concrete specimens and compared to test results available in the literature. Test
subjected to combined flexure and axial compression up to failure. results of this study as well as those from previous
The main variable considered was the strength of concrete that
research served as the basis for recommended revisions
ranged from 10.4 to 16 ksi (71.7 to 110.3 MPa). Each specimen
was subjected to two independent loads with a specific configuration
to code provisions on stress block parameters for concrete
to induce maximum compressive strain at one face and zero strain strength up to 18 ksi (124 MPa).
at the opposite. The measured stress-strain curves and stress block
parameters were compiled with the data found in the literature. EXPERIMENTAL PROGRAM
The results were used to develop recommended revisions for the The experimental program consisted of 21 concrete
LRFD specifications to extend their current limitation of 10 ksi specimens with a cross section of 9 x 9 in. (229 x 229 mm)
(69 MPa)for concrete compressive strength up to 18 ksi (124 MPa). and 40 in. (1 m) long. A general view of the concrete specimen
is presented in Fig. 1. The main parameter considered in this
Keywords: compression; flexure; high-strength concrete; loading; rectangular
study was the concrete strength. Three different target
stress block; strain; stress.
concrete compressive strengths of 10, 14, and 18 ksi (69, 97.
INTRODUCTION
Flexural failure of a reinforced concrete member occurs
when its extreme fiber reaches the ultimate compressive
strain uf concrete. Concrete in the compression zone is
subjected to a stress distribution, referred to as the stress
block, that follows the stress-strain relationship of a concrete
cylinder tested in axial compression. This paper focuses on
the evaluation of the stress block in the compression zone of
high-strength concrete (HSC) flexural members. The
strength considered in this investigation ranged from 10 to
18 ksi (69 to 124 MPa). The authors used a similar test setup
developed by Hognestad eta!. 1 in which an eccentric bracket
specimen was subjected to an axial compression and a
moment to simulate the stress profile in the compression (a) General view
zone of a rectangular flexural member.
The load and resistance factor design (LRFD) specifications. 2
,. 2.5 in. 63.5' mm
2.5i~mm
first published in 1994, include an article (5.4.2.1) limiting , 2.5 in. 63.5 rom
its applicability to a maximum concrete strength of 10 ksi
(69 MPa), unless physical tests are made to establish the
relationship between concrete strength and its other properties.
This limitation was imposed due to the lack of sufficient
research data on HSC at the time when the specifications
were developed. Many design provisions stipulated in the
LRFD specifications 2 are still based on test results obtained
from specimens. with compressive strengths up to 6 ksi
(41 MPa). Although such a strength limit is not explicitly
Side View Neutral Face View
imposed by ACI 318-05,3 except in its provisions for shear
and development length, its applicability to HSC is not fully (b) Steel reinforcement configuration
and explicitly addressed either. Details on other design codes
are given in Mertol. 4 Fig. }-General view and steel reinforcement configuration.

RESEARCH SIGNIFICANCE ACI Structural Journal, V. 105, No.5, Sept.-Oct. 2008.

This study focuses on general characteristics of the stress
profile in the compression zone of HSC flexural members.
The proposed model is based on the test results of 21
MS No. S-2007-129 received April9, 2007, and reviewed under Institute publication
policies. Copyright© 2008, American Concrete Institute. All rights reserved, including the
making of copies unless permission is obtained from lhe copyright proprietors. Pertinent
discussion including author's closure, if any, will be published in the July-August
2009 ACI Structural Journal if the discussion is received by March l, 2009.

626 ACI Structural Journal/September-October 2008


AC! member Hal it Cenan Mertol is a Design Engineer at Yuksel Proje International,
Ankara, TUrkey. He received his ESc aruf MS from Middle East Teclmical University,
Ankara, Turkey, in 1999 and 2002, respectively, and his PhD from North Carolina
State University, Raleigh, NC. He is a member of ACI Committee 363, High-
Strength Concrete, and Joint ACI-ASCE Committee 441, Reinforced Concrete
Columns. His research interests include characteristics of high-strength concrete
under combined flexure and axial compression.
Sami Rizkalla, FAC/, is Distinguished Professor of Civil Engineering and Construction in
the Department of Civil, Construction, and Environmental Engineering, North
Carolina State University. He is the Director of the Constructed Facilities Laboratory
and NSF l!UCRC on Repair of Structures and Bridges at North Carolinn Sta1e University
He is a member of ACI Committees 118, Use of Computers; 440, Fiber Reinforced
Polymer Reinforcement; £803, Faculty Network Coordinating Committee; and Joint
ACI-ASCE Committees 423, Prestressed Concrete; and 550, Precast Concrete Structures.
ACI Honorary Memher Paul Zia is a Distinguished University Professor Emeritus at
North Carolina State University. He served as ACI President in 1989, and is a member of
ACI Committees 363, High-Strength Concrete; 440, Fiber Reinforced Polymer
Reinforcement; Joint ACI-ASCE Committee 423, Prestressed Concrete: AC/'s
Concrete Research Council; and the TAG Technology Transfer Committee.
Amir Minniran, FACI, is Professor and Interim Dean of the Department of Civil and
Environmental Engineering at Florida International University, Miami, FL He is a
member of ACI Committee 440, Fiber Reiriforced Polymer Reinforcement; £803, Faculty
Network Coordinating Committee; and Joint ACI-ASCE Committee 343, Concrete Bridge
Design. His research interests include high~petfornumce concrete and composite mate':'aLv.
section, the extreme fiber was subjected to a monotonically
increasing compressive strain. In each loading step, the main
axial load from the test machine, P 1> was applied first to a
predetermined level to produce a uniform axial strain in the
section. Then, the secondary load P 2 was applied by a jack to
develop the strain gradient, maintaining zero strain at the exterior
face, and a maximum compressive strain at the opposite face,
Two steel moment anns were connected to the specimen
using six threaded rods through the holes in the rectangular
steel tube at each end. Each steel arm consisted of two 24 in.
(610 mm) long C8 x 11.5 channel sections welded to two
9 x I x 24 in. (229 x 25 x 610 mm) steel plates at the top and
the bottom. Half-inch (13 mm) stiffeners were used for the
steel anns. Two specially designed roller connections were
used to eliminate the end restrictions due to the applied axial load
from the machine. Each roller connection consisted of six 1 in.
(25 mm) diameter rollers and two curved plates, tapering
through inside and outside, respectively. The roller connection
assembly was fixed to side plates that were released at the time

and 124 MPa) were used. Five specimens were tested for the
target concrete compressive strength of 10 ksi (69 MPa)
whereas six and 10 specimens were tested for the target
concrete compressive strengths of 14 and 18 ksi (97 and
124 MPa), respectively. Three 4 x 8 in. ( 100 x 200 nun) 0 <>
@ ©
cylinders were also tested for each specimen to evaluate the @ @

concrete strength at the time of testing. Additional cylinders

were cast for each batch of concrete to establish the 28-day
compressive strength of the batch.
To prevent premature localized failure, both ends of the
specimens were heavily reinforced with three No. 4 U-shaped
longitudinal and three No. 3 transverse reinforcements, as
shown in Fig. 1. In addition, the ends of the specimens were
confined with 112 in. (13 nun) thick, 10 in. (254 mm) long
'i
rectangular steel tubes with holes on two opposite faces to
ensure proper transfer of the axial load and moment to the Neurra/face Compression
Typical side view face view
middle 16 in. (406 nun) plain concrete test section. view

The specimens and !he cylinders were demolded 24 hours Fig. 2-Test setup.
after casting and were then covered with wet burlap and
plastic sheets for a week. The specimens were then stored in
the laboratory where the temperature was maintained at Table 1-Three mixture designs for target concrete
approximately 72 °F (22 °C) with 50% relative humidity compressive strengths
until the time of testing. The ends of the cylinders were Target concrete compressive strengths
ground before testing. IOksi 14ksi 18 ksi
Material (69 MPa) (97 MPa) (124 MPa)
Materials Cement, lb/yd 3 (kg/m3) 703 (417) 703 (417) 935 (555)
Concrete mixture desims for the three different target strengths Silica fume, lb/yd3 (kgtm3) 75 (44) 75 (44) 75 (44)
were developed by Logan5 and are summarized in Table I.
Fly ash, lb/yd3 (kgtm3) 192 (114) 192 (114) 50 (30)
The coarse aggregate was crushed stone with a nominal
maximum size of 3/8 in. (10 mm). Two types of fme aggregate Sand, lb/yd3 (kgtm 3) 1055 (625)' 1315 (780)1 1240 (736) t
were used depending on the target compressive strength. The Rock, lb/yd3 (kg/m3) 1830 (1085) 1830 (I 085) 1830 (1085)
first was natural s~d and the second was manufactured Water, lb/yd3 (kg/m3) 292 (173) 250 (148) 267 (158)
sand, generically known as 2MS Concrete Sand. The High-range water-reducing
cement was Type I!II. Fly ash, silica fume, high-range 17 (1110) 24 (1565) 36 (2345)
admixture, fl ozllOO lb (mVlOO kg):l:
water-reducing admixture (HRWRA), and retarding admix- Retarding agent,
ture were used to obtain the appropriate strength. 3 (195) 3 (195) 3 (195)
I fl oz/100 1b (rnU100 kg)'
:1
wlcm 0.30 0.26 0.25
Test method and test setup 28-day compressive strength,
A schematic view of the test setup is shown in Fig. 2. The ksi (MPa) 11.5 (78.9) 14.4 (99.1) 17.1 (117.8)
two axial loads of P 1 and P2 were adjusted during the test to "'Natural sand.
maintain the location of the neutral axis, that is, zero strain at the tManufactured sand.
exterior face of the specimen. On the opposite side of the cross +ounces per 100 lb of cementitious material.

ACI Structural Journal/September-October 2008 627

of testing. Both the steel arms and the roller connections were
designed with a factor of safety of at least two against yielding to
ensure failure in the test region.
Instrumentation
The primary axial load P 1, applied by the 2000 kip (8900 kN)
load-controlled hydraulic compression machine, was
measured by an internal load cell. The secondary load P 2
was applied using a 120 kip (530 kN) manual hydraulic jack
and was measured using a I 00 kip (440 kN) flat load cell.
Each specimen was instrumented with 2.4 in. (60 mm)
electrical resistance strain gauges. A total of nine strain
gauges were mounted on each test specimen. Two gauges
were applied on the zero strain face. Four gauges were
mounted on the two sides of the specimen. The remaining
three were placed on the maximum compression side of the
specimen, one of which was used to measure the transverse
strain of concrete. Three I in. (25 mm) linear variable
displacement transducers (LVDTs) were placed at the top,
bottom, and midsection of the specimen to measure its
deflected shape and to incorporate the secondary moment
effect. The instrumentation layout is illustrated in Fig. 2.
Test procedure
The bottom roller connection was placed first into the
compression machine. The specimen was then positioned,
aligned, and leveled on the roller connection. The bonom and top
steel anus were then connected to the specimen using threaded
rods. The load cell and the jack, used to apply the secondary load,
were placed on the top arm. The bottom arm and the top arm
assembly were connected to each other using a threaded rod.
The top roller connection was then positioned and leveled.
The specimen was leveled using a thin layer of hydrostone.
As the primary axial load was increased incrementally, the
secondary load was applied to maintain the neutral axis at the
exterior face. The loading rate was kept at 2 micros trains per
second on the compression face of the specimen. Each test
lasted for about 25 minutes, until concrete was crushed.
Three companion cylinders were tested on the same day in
accordance with ASTM C39 6

TEST RESULTS AND DISCUSSIONS

The test-day average cylinder strengths for the three target :I,,
strengths were 11.1, 14.9, and 15.4 ksi (76.4, 102.5, and II
106 MPa), respectively. The highest test-day average cylinder
strength achieved in this research was 16.0 ksi (110 MPa). All
test specimens had a similar explosive failure mode with no
visible crack up to failure. Typical failure mode for the eccenttic
bracket tests is shown in Fig. 3. The cylinder strength, the
age at testing, the loading rate, and the ultimate compressive
strain achieved by the specimens are summarized in Table 2.

(a) Before testing (b) After testing Stress block parameters

The approach presented by Hognestad et al 1 was used to
Fig. 3-Typical failure mode for eccentric bracket specimens determine the stress-strain relationship for each specimen.
(18EB6). This approach was used to calculate the concrete stress fc as

Table 2-Tabulated test results

; ..
f: at testing, f: at testing, Age at testing, Loading rate, illtimate
Specimen ID ksi MPa days ~J,&Isecond strain, }!E k, k, k, "1 ~I v
lOEB! 11.0 76.2 63 12.2 3738 0.65 0.38 1.03 0.90 0.75 -

10EB2 11.4 78.7 109 2.0 3138 0.62 0.36 1.12 0.95 0.72 0.25
10EB3 11.7 80.7 Ill 2.4 3407 0.65 0.36 1.14 1.02 0.73 0.21
10EB4 10.4 71.4 63 2.1 3102 0.64 0.36 1.20 1.06 0.73 0.19
IOEB5 10.9 75.2 62 2.2 3023 0.62 0.36 1.16 1.01 0.72 0.20
14EB1 14.6 100.9 49 2.3 3316 0.63 0.37 1.00 0.85 0.74 0.22
14EB2 14.3 98.7 51 1.8 3162 0.60 0.36 1.08 0.85 0.72 0.20
14EB3 14.7 101.2 52 2.2 3177 0.61 0.36 1.09 0.93 0.71 0.23
14EB4 15.0 103.7 57 2.3 3032 0.58 0.35 1.10 0.92 0.70 0.23
14EB5 15.4 105.9 100 5.3 2868 0.57 0.34 1.10 0.92 0.68 0.24
J4EB6 15.2 104.5 101 4.1 2954 0.60 0.35 1.06 0.91 0.69 0.23
18EBJ 15.8 109.1 76 2.2 3684 0.69 0.38 0.82 0.74 0.77 0.23
18EB2 16.0 110.2 77 2.3 3364 0.67 0.37 0.85 0.77 0.74 0.22
18EB3 15.6 107.5 81 2.4 2914 0.63 0.37 0.81 0.69 0.73 0.23
18EB4 15.8 108.7 82 2.6 3306 0.65 0.36 0.88 0.78 0.73 0.20
18EB5 16.0 110.6 83 2.1 3144 0.65 0.36 0.85 0.76 0.72 0.24
18EB6 15.5 106.8 84 2.1 3404 0.66 0.37 0.88 0.78 0.74 0.22
18EB7 15.0 103.7 96 2.5 3585 0.64 0.37 1.05 0.90 0.75 0.24
18EB8 14.5 99.9 97 2.7 3507 0.65 0.37 1.03 0.91 0.74 0.22
18EB9 14.9 102.8 99 2.2 3494 0.62 0.36 1.06 0.91 0.72 0.23 s
18EBIO 14.6 100.7 102 2.0 3532 0.64 0.38 0.97 0.82 0.77 0.24 [

Note.- means could not be obtamed. <

628 ACI Structural Journal/September-October 2008

 a function of measured strain at the most compressed fiber £c
and the applied stresses fo and m 0 . The following equations
were obtained from equilibrium of external and internal
loads and moments. Note that the eccentricities due to
deflection of the member were also considered in the
calculation for total applied moment M.
(I)
~ ;ji
(2)
'lllf
I where Cis the total applied load, a 1 and a2 are the eccentricities
II.:.
~
,
with respect to the neutral surface, b is the width of the
section •. c is the depth of neutral axis, and
:tfi
' respectively. When converted to a rectangular distribution,
(3)
(4)
are the applied stresses. Some of these definitions are illustrated
in Fig. 2. Differentiating the last terms of the equations for C
and M with respect to Ec yields the following equations
(5)
previously. These parameters are shown in Fig. 5. The stress
block parameters for each specimen are also given in Table 2.
The ultimate strain of concrete member subjected to
flexure is generally higher than that of concrete cylinder
subjected to pure compression. The linear strain gradient in
the compression zone of flexural members helps in
achieving higher strain value at failure. Other reasons for
higher strain are the shape and size effects of the concrete
cylinder compared with the actual reinforced concrete structural
member. Furthermore, the rate of loading of a structrual
member is usually much slower than that of a concrete
cylinder. The stress distribution of concrete in flexure,
however, may still be represented adequately by the stress-strain
relationship of the concrete cylinder using an empirical constant
k 3 to account for all of these differences. This constant is
determined by comparing the beams tested in flexnre to the
companion cylinders tested under compression. 1
The stress distribution of normal-strength concrete (NSC)
is represented by the curved shape, as shown in Fig. 6. For
this stress distribution, k 1 and k 2 are equal to 0.85 and 0.425,
a 1 and 13 1 correspond to k 3 and 0.85, respectively. If the
stress distribution ofHSC is assumed to be triangular, k 1 and
k 2 would be equal to 0.50 and 0.333, respectively. Then, the
rectangular stress block parameters, a 1 and 13 1, would be
0.75k3 and 0.667, respectively. These parameters are also
shown in Fig. 6.
The test results of this research and other research reported
in the literature indicate that the majority of the collected
data for the generalized stress block parameter k 1 for HSC is
higher than 0.58 for concrete with compressive strengths
between 10 and 18 ksi (69 and 124 MPa), as shown in Fig. 7.
Therefore, the lower-bound value of 0.58 is suggested for k 1
parameter for concrete with compressive strengths higher

(6) " '"

g14" -.
1110 ~

Using these equations, twq similar stress-strain relationships !. 12

" M
were obtained for each eccentric bracket specimen and the 1: 10
-~
average of these two was used as the stress-strain relationship of i• " ~
the specimen. A typical obtained stress-strain distribution for ~ ~
HSC is shown in Fig. 4. The numerical values of the simplified '
~
40
•
stress-strain relationships for all specimens are given in Mertol.4 8' j
In general, the stress block in the compression zone of a "
flexure member can be defined by three parameters: k 1, k2,
and k 3 . The parameter k 1 is defined as the ratio of the average 500 1000 1500 2000 2500 3000
Concrete CompressiVe Strali. (11£)
3500 4000

compressive stress to the maximum compressive stress in the

compression zone k:J;. The parameter k2 is the ratio of the Fig. 4-Typical stress-strain distribution for eccentric
i!' depth of the resultant compressive force C to the depth of the bracket specimens (18EB9).
1:,
':1
compression zone c. The parameter k 3 is the ratio of the
~ maximum compressive stress in the compression zone to the
compressive strength measured by concrete cylinder/;. The
design values of the stresS block parameters are determined 1lE
'}•----=-'- --· -~~~-k:~~~~
when the strains at the extreme fibers reach the ultimate
strain of the concrete Scu· The three generalized parameters
Generalized Rectangular
of a stress block can be reduced into two parameters to Stress Block Stress Block
establish an equivalent rectangular stress block using a 1 Parameters Parameters
and p1, which ensures the compressive stress resultant being
at the same location. The stress block parameters in this
llj =klk3
study were calculated using the compressive strengths 2k,
measured by concrete cylinders, the stress-strain relationships
obtained in this study, and the methodology described Fig. 5-Stress block parameters for rectangular sections.

ACI Structural Journal/September-October 2008 629

than 15 ksi (103 MPa). The collected data for stress block In light of previous discussions, the following relationships
parameters from other researchers include test results are proposed for the rectangular stress block parameters a 1 and s
obtained by Hognestad eta!} Nedderman,? Kaar et al., 8·9 p1, for concrete strengths up to 18 ksi (124 MPa)
Swartz et a!., liT Pastor, 11 Schade,'2 Ibrahim,U Tan and
Nguyen, 14 and Sargin et al. 15 The tabulated values of the
research data are presented in Mertol. 4 a,= { 0.85 forf;,; 10.ksi (9 )
The k 2 parameter implied in ACI 318-05 3 and LRFD 0.85- O.OZ (]; - 10);, 0.75 for f; > 10 ksi
specifications 2 is already set to 0.33 for concrete with
compressive strengths greater than 8 ksi (55 MPa), because where fc' is in ksi.
the assumed p1 parameter used in design is equal to 0.65.
This provision is also confirmed by the coll~cted data for
HSC between 8 and 18 ksi (55 and 124 MPa), as shown in 0.85 for fc' ,; 69 ksi(lO)
{ 0.85- 0.0029(]; - 69);, 0.75
Fig. 8. Therefore, p1 can be assumed to be 0.33. a, = for > 69 ksi f:
The collected data also indicate that the stress block
parameter k 3 for HSC is similar to NSC, as shown in Fig. 9. where fc' is in MPa.
Hence, using the same value of k3 = 0.85 for concrete with
compressive strengths up to 18 ksi (124 MPa) is appropriate
for design purposes.
,,-
A _ { 0.85 for 4 ksi f: : :; (II)
Using the aforementioned values proposed for the generalized 0.85 - 0.05 (]; - 4) ;, 0.65 forf; >4 ksi
stress block parameters, the lower-bound relationships for
rectangular stress block parameters a 1 and p1 can be
obtained as follows where f::' is in ksi.

0.58 X 0.85 (7)

J:l, = {0.85 for f/ ,; 28 ksi(12)
= 0.75
2 X 0.33 0.85 - 0.007252(]; - 28) ;, 0.65 for f; > 28 ksi

1\ = 2k2 = 2 x 0.33 = 0.65 (8) where fc' is in MPa.

The comparisons of the proposed relationships to all
collected test results are shown in Fig. 10 and 11.

Concrete Compressive Strength (MPa)

i:
0 20 40 60 80 100 120
I
1.2

1.0
Others include Hogneslad el a\. 2, Nedderman7,
Kaar et al.8. 9, Swartz et al. 10, Paslor11 ,
13 14
!I!
~ '
1.
:
.,
'

0.8
Schade' 2
, lbrahim , and Tan and Nguyen .
1
I'
k, = 0.85
k, = 0.425
a,= kJ
p, =0.85
a,= 0.75kl
p, = 0.667 ..:."' 0.6
I
"'"'€"'"'~x>lo:",.,""b,.,"' x x
~X"fYy,%~~"'~
Nonnai-Strength Concrete Stress Triangular Stress Distribution 0.4
Distribution X XyltJ<

0.2
, r
Proposed Value
k~ = 0.33 for f'c :> 8 ksi (55 MPa)
k1 = -=-'';:;'";:dc:•dc;A:;;":::"=,---
Area of Dotted Rectangle 0.0
'----~T 0 2 4 6 8 10 12 14 16 18 20
Concrete Compressive Strength (ksi)
Fig. 6--Stress block parameters for different stress distributions.
Fig. 8-Proposed relationship for the stress block parameter k 2.
Concrete Compressive Strength (MPa)
0 20 40 60 80 100 120 Concrete CompressiVe Strength (MPa)
1.2 0 40 60 80 100
20 120

1.0
Others include Kaar et aLM, Swartz et al. 10,
Pastor11 , Schade' 2 , and lbrahim1 s.
1.2 r--~--,,-~--.-.....,~--~-~~-,
.,.. "'
"
....
)( ••~

lP
X -:.< /"-
1.0
x xxx"x"'x ~ r:li<""*,(xx "'x• x;x"
0.8 0.8 1-----.-1---''-~--"---"---'-'--":,.r--',

.:E 0.6 Proposed Value

:2 0.6 kJ = 0.85

0.4 Proposed Value 0.4

0.2
I+ This Research
1 x Others
k, = 0.58

0.2
i ~ This Research
1 x Others
I 10
!Others include Kaar el aL"- 9, Swartz et al. ,
fPastor", Schade 12 , lbrahim 13, and Sargin 15•
I
o.o 0.0 '-~~~-~--'--'--'--'--'-~--'
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Concrete Compressive Strength (ksl) Concrete Compressive Strength (ksi)

Fig. 7-Proposed relationship for the stress block parameter k1. Fig. 9-Proposed relationship for the stress block parameter k3.

630 ACI Structural Journal/September-October 2008

 An analysis was performed by Mertol 4 to evaluate the strengths up to 20 ksi (138 MPa) under eccentric loading. an
sensitivity of the ultimate moment capacity of a reinforced ultimate compressive strain of 0.003 is considered applicable
concrete member to the rectangular stress block parameters for design purposes for concrete with compressive strengths
a 1 and p1, and the results of the sensitivity analysis are up to 18 ksi (124 MPa). A comparison of the proposed ultimate
shown in Fig. 12 and 13. The analysis indicates that. for compressive strain of concrete with test results of this and
underreinforced concrete members, a reduction in the other research7- 16 reported in the literature is shown in Fig. 15.
rectangular stress block parameter a 1 by 11.8% leads to a When only the test results for concrete compressive strengths
reduction of the ultimate moment capacity by only 1.9%. For over 10 ksi (69 MPa) are considered, the 90 percentile line for
an over-reinforced concrete member, however, a reduction Ecu is very close to the proposed value of 0.003.
in a 1 by 11.8% may lead to a reduction of the ultimate
moment capacity by as much as 10.3%. Note tJ;lat a reduction
1.05 · - - - - - - . , . . - , . - , - - - - - - - - - - - - - - ,
of j3 1 by 23.5% has no effect on the ultimate moment [3, "'0.65 Under-Reinforced
Section (cld=0.375)

i
capacity of under-reinforced concrete members, but it will Eou = 0.003
(1.9% Reduction)
cause a reduction of slightly over 12% of the ultimate
100
moment capacity of over-reinforced concrete members.

Concrete strain measurements

"~
0
~ 0.95
The surface strain measurements at different load levels
for Specimen 18EB#2 are shown in Fig. 14. Similar behavior ~
5
Over-Reinforced
Section (cld=0.75)
was observed for other specimens. The graph validates the ~ 0.90
(10.3% Reduction)
assumption that plane sections remain plane after deformation
~

I
for HSC members.
i 1.8% Reduction in o:,
The ultimate concrete compressive strains measured at 0.85 '---~--~--~--~--~--.J
failure on the compression face of concrete are shown in 0.725 0.75 0.775 0.8 0.825 0.85 0.875
Table 2. Based on a regression analysis by Mertol4 on 188 Stress Block Parameter a,
! test results available in the literature with concrete compressive
I Fig. 12-Ratio of ultimate moment capacity versus change
in a1.

I 1.0
0
1.2 .--~--~--~--~-~--~----,
Concrete Compressive Strength (MPa)
40 60 80 100 120

...u
1·05 r-,-,e-0.-85-----,.U,-ad7,-,.-::R-,-,1,-oc-c-,d,-S",-c-tio-,--------,
Eou"' 0.003 (c!d=0.375) (No Reduction)
---------L----~~

' "' !j 0.95

'
€ 0.6
:i 0.9 AlP"'---- Balanced Section
Proposed Relationship (cld=0.6)
0.4
Others include Hognestad et al.~, Nedderman , 7
fora, ~ (12.7% Reduction)
§ 0.85
Kaar et al. 8 ·9 , Swartz et al. 10, Pastor11 , Schade 12 , '. This Research 0
0.2 1

lbrahim 13, and Tan and Nguyen 14 . 1

Others i ~

0.0 L:::::::;:::=:;~:::;:=~~=::::;:~..-====::J &! o.8

2 4 6 8 10 12 14 16 18 20 23.5% Reduction in (3 1
0
Concrete Compressive Strength (ksi) 0.75 '---~--~---~--~---'---_.1
0.6 0.65 0.7 0.75 0.8 0.85 0.9
Stress Block Parameter j).,
Fig. 10-Proposed relationship for the rectangular stress
block parameter a 1. Fig. 13-Ratio of ultimate moment capacity versus change
in fJJ.
Concrete Compressive Strength (MPa)
0 20 40 60 80 100 120
1.2 3000

1.0 ~ Concrete
~""*""" X: ~ )( )( )(
.
- _l'xei_ )( Strain
..... .....·~~'\'~
0.8 (<")
X X; ~ t:JO<h~>K~
.E.. 0.6
X /X'"- X
LRFD Specifications 1'
* XX

\
.'""
forf3 1
• Proposed Relationship
for 131
0.4
Others include Hognestad et al?, Nedderman 7 ,
Kaar et al. 8·9, Swartz et al. 10 , Pastor1 \ Schade 12,
0.2
lbrahim 13 , and Tan and Nguyen 14 . 1: This Research 1
Others 1 Compression
Face
0.0
0 2 4 6 8 10 12 14 16 18 20
Neutral
Concrete Compressive Strength (ksi) Face Center line oft he
'------J''----_j test region

Fig. ]]-Proposed relationship for the rectangular stress

block parameter j]1. Fig. 14-Strain distribution on side face of Specimen 18EB#2.

ACI Structural Journal/September-October 2008 631

 Concrete Compressive Strength (MPa) ranged from 10.4 to 16 ksi (71. 7 to 110.3 MPa). The data
0 20 40 60 80 100 120 obtained herein were used to determine the fundamental
0.006 r--:.;--~--~--~---.----;:::;:;~;;::;1 characteristics of the stress distribution in the compression
• This Research I zone of a flexural member. The test results obtained in this
1 0.005 Ix Others
research combined with the available data in the literature
I" 0..004 were used to develop recommended changes for code
provisions. The following conclusions can be drawn with

1*: :
E
~ 0.001
respect to flexural design with HSC up to 18 ksi (124 MPa).
1. The assumption that plane sections remain plane after
deformation is valid for concrete compressive strengths up to
18 ksi (124 MPa);
s
}

]
A
2. The ultimate concrete compressive strain value of 0.003
for design by the current code provision is acceptable for
0 2 4 6 8 10 12 14 16 18 20 concrete compressive strengths up to 18 ksi (124 MPa);
Concrete Compressive Strength (ksi) 3. A Poisson's ratio of 0.2 as used in the current code
provision is also acceptable for concrete compressive
Fig. IS-Proposed relationship for ultimate concrete strengths up to 18 ksi (124 MPa);
compressive strain &ell' 4. The test results, confirmed by other data in the literature,
Concrete Compressive Strength (MPa)
indicate that the stress block parameter a 1 of 0.85 should be
0 20 40 60 80 100 120
reduced when concrete compressive strength exceeds 10 ksi
(69 MPa). A new relationship is proposed for tbe parameter a 1
rfni~;:;;;~;;;:~:;;;:~;;;~~~~~:;:;;;;=f::;:::;:;::;-;;:;;;;;J
10
~~Others include Logan • Swartz et al. , Komendant et

'
0.40 4
0.35 al. 17 , Perenchio and Klieger' 8 , Carrasquillo et aL 19 ,
Jerath and Yamane 20 , Radain et al. 21 , and travani~.
0.30 I is deemed appropriate for concrete compressive strengths up
0
:a; 0.25
"'•
~~ ::::
~
0.10
0.05
0.00 . _ _ _ . _ _ _ . _ _ . _ _ . _ _ . _ _ . _ _ . _ _ . _ _ . _ _ _ _ ,
0 2 4 6 8 10 12
Concrete Compressive Strength (ksi)
Fig. 16~Proposed relationship for Poisson's ratio.
Poisson's ratio
The measurements of a horizontal strain gauge on the
compression face were us~d to calculate the Poisson's ratio
for HSC. The calculated values of Poisson's ratio v for all
target concrete strength specimens are shown in Table 2. The
comparison of the test results of this research and other
research in the literature5·10•17 -22 is shown in Fig. 16. There
is no apparent trend for Poisson's ratio as concrete compressive
strength increases.
A total of 246 test results with concrete compressive
strengths up to 20 ksi (138 MPa) were analyzed by Mertol4
using regression technique to develop the relationship
between Poisson's ratio v and concrete compressive strength
f~. Based on all collected test results, a Poisson's ratio of 0.2
was found to be suitable for concrete strengths up to 18 ksi
(124 MPa). If only the test results of specimens with concrete
compressive strengths over 10 ksi (69 MPa) are considered,
there is a slight increase in the Poisson's ratio as concrete
compressive strength increases. The proposed value of 0.2 for
Poisson's ratio represents the 44 percentile of the test data
SUMMARY AND CONCLUSIONS
A total of 21 plain HSC specimens were tested under
eccentric compression to simulate the compression zone of a
flexural member by varying the applied axial load and
moment. The dimensions of the specimens were 9 x 9 x 40 in.
(229 x 229 x 1016 mm) and the concrete cylinder strength
for concrete compressive strengths up to 18 ksi (124 Mpa); and
I• This Research j
5. The current value of Pr = 0.65 for f~ > 8 ksi (55 MPa)
I " Others
to 18 ksi (124 MPa).
ACKNOWLEDGMENTS
The authors would like to acknowledge the support of the NCHRP
through Project 12-64 and the Senior Program Officer, D. Beal. They are
also grateful for the contributions of H. Russell of Henry Russell, Inc., and
R. Mast of Berger/ABAM Engineers, Inc., who served as consultants for the
project. The cooperation of Ready Mixed Concrete Company of Raleigh,
NC, and the help of the personnel of the Constructed Facilities Laboratory
14 16 18 20 are greatly appreciated. The assistance provided by A. Logan, S. Kim,
Z. Wu, and W. Choi during all aspects of the re~ea.rch program was crucial
to the success of the project.
REFERENCES
1. Hognestad, E.; Hanson, N. W.; and McHenry, D., "Concrete Stress
Distribution in Ultimate Strength Design," ACI JouRNAL, Proceedings
V. 52, No.4, Dec. 1955, pp. 455-479.
2. ''AASHTO LRFD Bridge Design Specifications," third edition, American
Association of State Highway and Transportation Officials, Washington, DC,
2004, 1450 pp.
3. ACI Committee 318, "Building Code Requirements for Structural
Concrete (ACI 318-05) and Commentary (318R-05)," American Concrete
Institute, Fannington Hills, MI, 2005, 430 pp.
4. Merta], H. C., "Behavior of High-Strength Concrete Members Subjected
to Combined Flexure and Axial Compression Loadings," PhD dissertation,
Department of Civil, Construction and Environmental Engineering,
North Carolina State University, Raleigh, NC, Dec. 2006, 320 pp.
5. Logan, A. T., "Short-Tenn Material Properties of High-Strength
Concrete." MS thesis, Department of Civil, Construction and Environmental
Engineering, North Carolina State University, Raleigh, NC, June 2005, 116 pp.
6. ASTM C39, "Test Method for Compressive Strength of Cylindrical
Concrete Specimens," ASTM International, West Conshohocken, PA,
2005,7 pp.
7. Nedderman, H., "Flexural Stress Distribution in Very-High Strength
Concrete,'' MS thesis, Civil Engineering Department, University of Texas
at Arlington, TX, Dec. 1973, 182 pp.
8. Kaar, P. H.; Hanson, N. W.; and CapeU, H. T., "Stress-Strain Characteristics
of High Strength Concrete," Douglas McHenry International Symposium
on Concrete and Concrete Structures, SP-55, B. Bresler, ed., American
Concrete Institute, Fannington Hills, MI. Aug. 1978, pp. 161-185.
9. Kaar, P. H.; Fiorato, A. E.; Carpenter, J. E.; and Corely, W. G.,
"Limiting Strains of Concrete Confined by Rectangular Hoops," Research
and Development Bulletin RD053.0JD, Portland Cement Association,
Skokie, IL, 1978, 12 pp.
10. Swartz, S. E.; Nikaeen, A.; Narayan Babu. H. D.; Periyakaruppan, N.;
and Refai, T. M. E., "Structural Bending Properties of Higher Strength
Concrete," High-Strength Concrete, SP-88, H. G. Russell, ed., American
Concrete Institute, Fannington Hills, MI. 1985, pp. 147-178,

632 ACI Structural Journal/September-October 2008 ACI

 11. Pastor, J. A., "High-Strength Concrete Beams," PhD thesis, Department
of Civil Engineering, Cornell University, Ithaca, NY, Jan. 1986, 277 pp.
12. Schade, J. E., "Flexural Concrete Stress in High Strength Concrete
Columns," MS thesis, Civil Engineering Department, University of
Calgary, Calgary, AB, Canada, Sept. 1992, 156 pp.
13. Ibrahim, H. H. H., "Flexural Behavior of High-Strength Concrete
Columns," PhD thesis, Department of Civil and Environmental Engineering,
University of Alberta, Edmonton, AB, Canada, 1994,221 pp.
14. Tan, T. H., and Nguyen, N. B., "Flexural Behavior of Confined High-
Strength Concrete Columns," AC/ Structural Journal, V. 102, No. 2, Mar.-
Apr. 2005, pp. 198-205.
15. Sargin, M.; Ghosh, S. K.; and Handa, V K., "Effects of Lateral
Reinforcement upon the Strength and Deformation Properties of Concrete,"
Magazine of Concrete Research, V 23, No. 75-76, June-Sept 1971, pp. 99-110.
16. Hognestad, E., "A Study of Combined Bending and Axial Load in
Reinforced Concrete Members," University of Illinois Bulletin Series
No. 399, V 49, No. 22, Nov. 1951, 128 pp.
17. Komendant, J.; Nicolayeff, V.; Polivka, M.; and Pirtz, D., "Effect of
ACI Structural Journal/September-October 2008
Temperature, Stress Level, and Age at Loading on Creep of Sealed
Concrete," Douglas McHenry International Symposium on Concrete and
Concrete Structures, SP-55, B. Bresler, ed., American Concrete Institute,
Fannington Hills, MI, Aug. 1978, pp. 55-82.
18. Perenchlo, W. F., and Klieger, P., "Some Physical Properties of High-
Strength Concrete," Research and Development Bulletin RD056.0JT,
Portland Cement Association, Skokie, IL, 1978, 6 pp.
19. Carrasquillo, R. L.; Nilson, A. H.; and Slate, F. 0., "Properties of
High-Strength Concrete Subject to Short-Term Loads," ACI JouRNAL
Proceedings V 78, No. 3, May-June 1981, pp. 171-178.
20. Jerath, S., and Yamane, L. C., ''Mechanical Properties and Workability of
Superplasticized Concrete," Cement Concrete and Aggregates, V. 9, No. 1,
1987, pp. 12-19.
21. Radain T. A.; Samman, T. A; and Wafa, F. F., "Mechanical Properties of
High Strength Concrete," Proceedings of Utilization of High-Strength Concrete
Symposiwn, Lillehammer, Norway, 1993, pp. 1209-1216.
22. lravani, S., "Mechanical Properties of High-Strength Concrete," ACI
Materials Journal, V. 93, No.5, Sept.-Oct. 1996, pp. 416-426.
633