INFLUENCE OF THE INCORPORATION OF RECYCLED CONCRETE COARSE AGGREGATES ON THE PUNCHING STRENGTH OF CONCRETE SLABS Nuno Carvalho Martins dos Reis Extended Abstract Master’s Degree in Civil Engineering Supervisors Prof. Dr. Jorge Manuel Caliço Lopes de Brito Prof. Dr. João Pedro Ramôa Ribeiro Correia Jury President: Prof. Dr. José Manuel Matos Noronha da Câmara Supervisor: Prof. Dr. Jorge Manuel Caliço Lopes de Brito Member: Prof. Dr. António Manuel Pinho Ramos September 2014
18
Embed
INFLUENCE OF THE INCORPORATION OF RECYCLED CONCRETE … · RC 129 0.54 0.54 C20 124 0.55 0.54 C50 131 0.57 0.54 C100 121 0.60 0.54 Figure 1 - Density of fresh concrete 2.3. Test specimens
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INFLUENCE OF THE INCORPORATION OF RECYCLED CONCRETE
COARSE AGGREGATES ON THE PUNCHING STRENGTH OF
CONCRETE SLABS
Nuno Carvalho Martins dos Reis
Extended Abstract
Master’s Degree in Civil Engineering
Supervisors
Prof. Dr. Jorge Manuel Caliço Lopes de Brito
Prof. Dr. João Pedro Ramôa Ribeiro Correia
Jury
President: Prof. Dr. José Manuel Matos Noronha da Câmara
Supervisor: Prof. Dr. Jorge Manuel Caliço Lopes de Brito
Member: Prof. Dr. António Manuel Pinho Ramos
September 2014
1
1. Introduction
The construction sector is responsible for 40% of the consumption of extracted natural re-
sources and the major waste producer worldwide. Therefore, proper management of waste and
resources consumption is essential to achieve sustainable construction. With this purpose,
based on a study carried by Ortiz et al. (2010), 80% of the total construction and demolition
waste (CDW) is capable of being recycled and reused. However, according to a Bio Intelligence
Service report, only about 46% of all waste produced in the European Union (EU) are reused /
recycled. The use of recycled aggregates (particularly originating from crushing concrete) in
concrete is a recent approach in the reuse and recycling of CDW.
In order to contribute to the knowledge about the incorporation of recycled concrete aggregates
(RCA) in concrete, a research line has been developed in Instituto Superior Técnico (IST). Sev-
eral studies on this subject (investigating the processing of demolished products, mix proportion
design, mechanical properties and durability aspects have provided a better understanding
about the performance of concrete made with RCA (CRCA) and indicated positive prospects
concerning its use. Recently, the structural performance and economic aspects of using recy-
cled concrete coarse aggregate concrete (RCCAC) have also been analyzed and they pointed
out the potential of applying RCCAC in civil engineering structures.
The mechanical and durability properties of RCCAC have been studied in depth in the last
years. However the structural performance of this concrete type is still underdeveloped. Thus,
this work aims to contribute to the knowledge of the structural behavior of RCCAC. Particularly,
it evaluates the effect of the incorporation of recycled concrete coarse aggregates (RCCA) on
the punching strength of concrete slabs. According to the best of the author’s knowledge, this
topic has been addressed only by Sudarsana Rao et al. (2012).
The first step of this research consisted of a literature review, in order to collect information on
RCCA properties and the effect of their incorporation in terms of concrete mechanical properties
(such as compressive strength, tensile strength and modulus of elasticity) and structural perfor-
mance. The comparison between RCCAC and natural aggregate concrete (NAC) demonstrates
a generalized worsening of the mechanical properties with the increase of RCCA content (Brito,
2005). The punching tests on concrete slabs performed by Sudarsana Rao et al. (2012) indicat-
ed a decreasing trend in the punching strength of concrete slabs when the RCCA content in-
creases, as well as lower stiffness and cracking load.
The next stage was to prepare and execute the experimental campaign. First, the RCCA and
natural aggregates (NA) properties were analyzed, aiming at designing the four concrete mixes
produced. Their workability was determined to guarantee compliance with the target slump (125
± 10 mm). Concrete cubes and cylinder test specimens (to characterize the mechanical proper-
ties) and eight slabs (two for each type of concrete, with 1100 x 1100 x 90 mm, all submitted to
the punching test) were prepared. Finally, the hardened concrete and punching tests were per-
2
formed and their results were analyzed and discussed in detail.
Alongside the experimental campaign, a three-dimensional non-linear finite element model
(FEM) was also developed using ATENA 3D program to simulate the punching test of the slabs.
Numerical results were compared with experimental data, allowing for a better understanding
about the structural response of the tested slabs and also to validate the experiments.
2. Experimental programme
2.1. Materials
In this research, two main materials were used: concrete and steel.
Two types of river sand and three different grades of crushed limestone were used as NA. The
RCCA were obtained by crushing concrete beams (cast in situ at the Construction Laboratory of
IST) with 28 days of age, using a concrete jaw crusher. The other constituents used were tap
water and CEM II 42.5 Portland cement.
To design the concrete mixes, the measurement of some aggregate properties was essential.
Thus, the following properties were determined: particle density and water absorption at 24
hours (according to NP EN 1097-6 (2003)); bulk density (NP EN 1097-3 (2002)); Los Angeles
coefficient (that measures the abrasion resistance, according to NP EN 1097-2 (2010)); shape
index (NP EN 933-4 (2008)); and water content of RCCA (NP EN 1097-5 (2011)). The experi-
mental results of these tests are presented in Table 1.
Table 1 - Experimental results of the aggregates’ tests
Property Fine sand
Coarse sand
Fine gravel
Intermediate gravel
Coarse gravel RCCA
Particle density (kg/dm3) 2.55 2.51 2.53 2.54 2.56 2.28
Water absorption (%) 0.3 0.6 1.4 1.3 1.2 6.2
Bulk density (kg/dm3) 1.55 1.59 1.43 1.44 1.45 1.23
Los Angeles coefficient (%) - - 23.2 26.5 29.2 41.9
Shape index (%) - - 17.2 17.6 14.9 23.1
Water content (%) - - - - - 3.05
As expected, the cement paste in the structure of the RCCA decreases their quality compared to
the NA, mainly due to their high porosity and lightness. Thus, these aggregates present lower
particle and bulk density and higher water absorption and content, shape index and Los Angeles
coefficient than NA. As a consequence of the higher water absorption of RCCA, the mixing water
must be compensated to maintain the workability of the RCCAC.
As top and bottom reinforcement of the slabs, 6 and 8 mm diameter hot-rolled steel (S500) bars
were used. These present yielding and ultimate stresses of 600 MPa and 640 MPa, respective-
ly, and elasticity modulus of 200 GPa.
3
2.2. Concrete mixes design
Four different concrete compositions (Table 2) were designed according to the Faury’s method
with various replacement rates of NCA by RCCA: 0% - reference concrete (RC); 20% (C20);
50% (C50); and 100% (C100). The RC mix was designed to have a strength class C30/37,
equivalent to the concrete that was crushed (the RCCA source), with C20, C50 and C100 hav-
ing the same target class, and an Abrams cone slump of 125 ± 10 mm (given by an effective
water/cement (w/c) ratio of 0.54). In order to maintain constant the workability in all the mixes
and due to the higher RCCA’s water absorption, additional water was added to the mix (equal to
the absorbed water in 15 minutes by the RCCA minus their initial water content times the vol-
ume of RCCA). This explains the difference between the apparent and effective w/c ratio in
RCCAC mixes. As mentioned, Portland cement (CEM II 42.5 R) was used and the maximum
particle size was 22.4 mm.
Table 2 - Composition of the mixes
Component Volume (m
3/m
3)
RC C20 C50 C100
Natural coarse
aggregates
22.4 - 16 mm 0.121 0.097 0.061 0.000
16 - 11.2 mm 0.120 0.096 0.060 0.000
11.2 - 8 mm 0.047 0.038 0.024 0.000
8 - 5.6 mm 0.047 0.037 0.023 0.000
5.6 - 4 mm 0.041 0.033 0.020 0.000
Recycled concrete
coarse ag-gregates
22.4 - 16 mm 0.000 0.024 0.061 0.121
16 - 11.2 mm 0.000 0.024 0.060 0.120
11.2 - 8 mm 0.000 0.009 0.024 0.047
8 - 5.6 mm 0.000 0.009 0.023 0.047
5.6 - 4 mm 0.000 0.008 0.020 0.041
Coarse sand 0.250 0.250 0.250 0.250
Fine sand 0.054 0.054 0.054 0.054
Cement 0.115 0.115 0.115 0.115
Water 0.188 0.193 0.199 0.210
Apparent w/c ratio 0.54 0.55 0.57 0.60
Effective w/c ratio 0.54 0.54 0.54 0.54
The slump of the mixes was measured based on the Abrams slump cone test (in accordance
with NP EN 12350-2 (2009)). The bulk density of the fresh concrete was calculated according to
NP EN 12350-6 (2009). The results of both tests are, respectively, presented in Table 3 and
Figure 1. As observed, the slump results are in the defined range and the bulk density decreas-
es as the content of RCCA increases, which is explained by the lower density of the RCCA.
4
Table 3 - Abrams cone slump test results and w/c ratios
Concrete Slump (mm)
Apparent w/c
Effective w/c
RC 129 0.54 0.54
C20 124 0.55 0.54
C50 131 0.57 0.54
C100 121 0.60 0.54
Figure 1 - Density of fresh concrete
2.3. Test specimens
The test specimens included eight reinforced concrete slabs (two per type of concrete analyzed)
and their size was established in order to simulate the area around a column, in which the radial
moments at the contour are approximately null. As seen in Figure 2, the total size of the slab
corresponds to 44% of the span length (Lv) that it intends to simulate, considering that the mo-
ment is zero at a distance of 0.22Lv from the column axis. Thus, the tests were performed in
square slabs with 1100 x 1100 x 90 mm. The distance between (radial) supports was 1000 mm,
in order to simulate a 2.30 m spam between columns. The ratio Lv/h of the slabs is 25, which is
within the range of values recommended by EC2 for this type of structural elements.
In all specimens, as mentioned hot-rolled bars of steel (S500) were used as reinforcement. The
amount of the top flexural reinforcement is 0.93% (in a 8 mm//75 mm grid). The bottom rein-
forcement consists of 6 mm diameter bars spaced by 150 mm. A cover of 10 mm was used.
Thus, the mean depth (d) is about 72 mm.
Plywood moulds were used to cast the specimens. After 3 days, the specimens were de-moulded
and subjected to curing (watering): once a day for the first 7 days and thereafter every 2 days up
to the test date (28 days).
Figure 2 – Approximate radial moments distribution in slabs
In order to evaluate the mechanical properties of each type of concrete used to cast the slabs,
the following tests were performed: compressive strength at 28 and 56 days (according to NP
EN 12390-3 (2009)); splitting tensile strength at 28 days (NP EN 12390-6 (2009)); and modulus
of elasticity at 28 days (LNEC E-397 (1993)). The specimens used in these tests (150 mm cu-
2385 2357
2325
2267
2250
2300
2350
2400
0 10 20 30 40 50 60 70 80 90 100
De
nsi
ty (
kg/m
3 )
Replacement rate of NCA by RCCA (%)
0.22L 0.44L
L
5
bes for compression strength and cylinders with 300x150 (d) mm for the other properties) were
de-moulded 24 hours after casting and placed in a wet curing chamber during 28 days.
2.4. Punching tests setup, instrumentation and procedure
The setup used in the punching tests is presented in Figure 3. The slabs were simply supported
on 8 supports uniformly distributed in a circular pattern, placed on top of four concrete blocks.
Each of the metal supports comprised three thick steel plates, the upper two enclosing 20 mm
steel rods that allowed free rotation in the radial direction. To avoid restricting the horizontal
movement of the slabs, 1–mm thick sheets of PTFE were placed between the bottom two steel
plates of the supports. Figure 4 illustrates the supports described.
Figure 3 - Punching test setup Figure 4 - Support devices
A vertical (descending) point load was applied with an Enerpac hydraulic jack (load capacity of
600 kN, mounted on a steel reaction frame) placed on the center of the slab and contacting with
it through a 150 x 150 x 30 mm steel plate that aimed at simulating a column. A spherical hinge
was placed between this plate and the hydraulic jack to correct possible geometrical imperfec-
tions. The slab specimens were tested upside down relative to the normal orientation of a slab-
column connection. As a result, the tensile flexural reinforcement was located on the bottom
side and the column (in this case, the steel plate) was located above the slab. The overall ge-
ometry and dimensions of the slabs, as well as the arrangement of the longitudinal reinforce-
ment and position of the supports are presented in Figure 5.
During the preparation of the test system, fine layers of plaster were used to level the contact
surfaces between: the laboratory strong floor and the concrete blocks; the concrete blocks and
the supports; the supports and the slab; and the slab and the column (steel plate). In these op-
erations, a level was used to guarantee that the upper face of the elements of the test system
was horizontal.
To monitor the structural behavior of the slab specimens, various measurements were continu-
ously recorded. The applied load was measured with a Novatech load cell (capacity of 400 kN)
6
placed between the hydraulic jack and the steel plate. Two displacement transducers (TML
brand) were used on the top and bottom surfaces of the slab to measure the deflection of the
center point. The top one has a range of 500 mm and the bottom one is more accurate and has
a range of 25 mm. The rotation of the slab was measured by two TML clinometers located
above diametrically opposite supports. Therefore, the correct evolution of the test (i.e. perfectly
symmetric punching) was guaranteed by checking similar slopes in the two clinometers.
Figure 5 - Overall geometry of the slab specimens: plan view (top) and section view (bottom)
After placing the slabs and the elements described in the planned position, the following proce-
dure was carried out:
checking the instrumentation;
application of a load up to 20 kN (for the settlement of the test system);
complete unloading of the slab;
reset the instruments;
loading up to 20 kN;
register of the crack pattern (by unaided visual observation);
loading up to 120 kN (in 20 kN steps in order to register changes in the crack pattern);
loading up to the failure;
recording the failure mode and the crack pattern.
7
For safety reasons, it was decided not to register the crack pattern beyond 120 kN, since this
procedure was conducted under the test specimen.
The readings obtained from the different sensors were gathered in a PC using an HBM data
logger. The slabs were tested under load control. An average load speed of 0.5 kN/s was used.
3. Experimental results
3.1. Concrete properties
Table 4 shows the compressive strength (fcm,28), the splitting tensile strength (fctm,28) and the
modulus of elasticity (Ecm,28), all at 28 days of age. The relative performance reduction between
RCCAC and RC mechanical properties (RC,28) is also presented.
As expected, concrete’s mechanical properties were not significantly affected by the incorpora-
tion of RCCA. All the mixes present a high compressive strength (of the strength class expected
- C30/37) and similar values between the various mixes, demonstrating the RCCA’s good quali-
ty. The splitting tensile strength experienced a slight reduction (regardless of the RCCA incorpo-
ration rate) with the incorporation of RCCA. Finally, the modulus of elasticity was reduced. Such
reduction was lower than that reported in most studies concerning RCCAC, possibly due to the
increasing of effective w/c ratio that reduces the effect of a weak aggregates’ quality.
Table 4 - Mechanical properties for each type of concrete