Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
a r t i c l e i n f o a b s t r a c t
Article history: Fine recycled aggregates are seen as the last choice in recycling for concrete production. Many references
Received 22 June 2009 quote their detrimental inuence on the most important characteristics of concrete: compressive and
Received in revised form 22 September 2009 tensile strength; modulus of elasticity; water absorption; shrinkage; carbonation and chloride penetra-
Accepted 28 September 2009
tion. These two last characteristics are fundamental in terms of the long-term durability of reinforced
Available online 4 October 2009
or prestressed concrete. In the experimental research carried out at IST, part of which has already been
published, different concrete mixes (with increasing rates of substitution of ne natural aggregates
Keywords:
sand with ne recycled aggregates from crushed concrete) were prepared and tested. The results were
Concrete
Durability
then compared with those for a reference concrete with exactly the same composition and grading curve,
Recycling but with no recycled aggregates. This paper presents the main results of this research for water absorp-
Fine aggregates tion by immersion and capillarity, chloride penetration (by means of the chloride migration coefcient),
Water absorption and carbonation resistance, drawing some conclusions on the feasibility of using this type of aggregate in
Chloride resistance structural concrete, while taking into account any ensuing obvious positive environmental impact.
Carbonation resistance 2009 Elsevier Ltd. All rights reserved.
0958-9465/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cemconcomp.2009.09.005
10 L. Evangelista, J. de Brito / Cement & Concrete Composites 32 (2010) 914
Water absorption by immersion measures the permeation (river sand). After this screening, there were signicant differences
capacity of concrete under a given pressure, i.e. the capacity of a between the natural and recycled aggregates in terms of their
uid to move through the pore structure. This phenomenon de- grading curves. The recycled aggregates grading curve therefore
pends mostly on the concrete microstructure (especially on the had to be converted in order to reproduce that of the natural aggre-
open pore structure), moisture condition and properties of the per- gates (so as to obtain the same neness modulus for all the mixes
meating uid [11]. Although water absorption by immersion gives with different replacement ratios and eliminate the inuence of
an estimate of the volume of a concretes open pores, it is not a reli- this parameter on the results). For this, all the recycled aggregates
able way of determining the concretes chloride and carbonation were sieved to obtain the various grading sizes and the different
resistance, since this depends more on capillary absorption [12]. fractions were kept in sealed containers to prevent humidity ex-
Water absorption through capillarity is a phenomenon that is change with the outside environment. Even though the practicabil-
due to the difference between the uids surface capillary pressure ity of this procedure leaves much to be desired, it was decided to
and its gravity pressure, which induces uid movement until a bal- go ahead with it since it would eliminate unwelcome entropies
ance is established. Capillary pressure increases with decreasing that would otherwise most surely hinder the analysis of the results
capillary diameter and is most relevant at the boundaries of con- obtained.
crete elements. The process is particularly visible in drywet con-
ditions and has the most relevance near the elements surface. 2.2. Concrete mix design
The presence of chloride ions in concrete microstructure is
known to lead to steel corrosion [13]. Carbonation is a chemical The calculation of the composition of the various mixes re-
phenomenon where the carbon dioxide that penetrates the con- quired the characterization of the aggregates used in them, with
crete structure reacts with calcium hydroxide to drastically reduce special emphasis on the properties of the ne aggregates. Table 2
concrete alkalinity, enabling steel corrosion by destroying the pas- presents the main properties of the ne recycled aggregates
sive layer. Carbonation depends on the cements composition, (FRA) and compares them with the ne natural aggregates (FNA).
aggregate type and porosity [14]. Together these two phenomena In brief, the recycled aggregates have lower specic densities than
are the main factors in reducing concrete durability [10]. Both the natural aggregates because of their higher porosity, resulting in
chloride penetration and carbonation tend to occur by diffusion, the greater water absorption observed.
which is the movement of an element due to its concentration gra- The different compositions of concrete in terms of the FNA/FRA
dient within the concrete microstructure. This phenomenon is replacement ratio (30% and 100%) and the reference concrete (RC)
characterized by Ficks second law: produced with natural aggregates alone were studied in accor-
dance with the methodology proposed by Faury [15], with work-
@/ @2/ ability (slump as measured by the Abrams cone) as a common
D 2 1:1
@t @x characteristic, within the interval 80 10 mm.
where u is the ionic concentration, t is the time, D is the diffusion The rst mixes were made with the RC, thus determining the
coefcient and x is the position. composition of the remaining concretes. The W/C ratio (and conse-
quently to calibrate the quantities of the remaining materials) had
to be corrected for the subsequent mixes since the more angular
2. Experimental program shape of the particles of ne recycled aggregates produces higher
inner friction [16].
2.1. Fine recycled concrete aggregate production Furthermore, it was necessary to introduce the water that the
recycled aggregates were going to absorb during the mixing, to
The ne recycled aggregates used in the experimental research prevent it from being removed in the hydration process which
reported here were obtained by crushing the original concrete (OC) would jeopardize the concretes performance. Two different W/C
produced in the laboratory for that purpose in an impact crusher. ratios therefore needed to be determined: the global W/C ratio,
By choosing this concrete as a source of recycled aggregates two i.e. the ratio between the overall quantity of water introduced into
problems were solved: rst, it was guaranteed that all the aggre- the mix and the amount of binder that, even though very simple to
gates used came from the same source, thus reducing the scatter determine, is of no particular interest to the study of the concrete
that they often exhibit; second, it was possible to determine the mixes, and the effective W/C ratio, i.e. the ratio between the
main properties of the original concrete thereby permitting a bet- amount of free water within the mix and the amount of binder
ter interpretation of the results. Table 1 gives the composition of whose determination, though fundamental to understanding the
the original concrete and some of its properties. concretes performance, is difcult because the evolution of the
After 35 days, the concrete was crushed. The resulting material water absorption by the recycled aggregates versus time has to
was composed of aggregates whose size ranged from 0 to 38.1 mm. be known in detail. This particular feature of recycled aggregates
Of these sizes only the fractions up to 1.19 mm were used, since is of fundamental importance to the composition and performance
this was the maximum dimension of the ne natural aggregates of concretes made with them. In a research program at IST and UPC
(Barcelona, Spain) under the joint supervision of the second author,
Table 1
the effect of the method of compensating for the coarse recycled
Original concrete composition. concrete aggregates excessive water absorption in concrete
Materials OC
CEM II 32.5 N (kg/m3) 362 Table 2
River sand (kg/m3) 615 Fine recycled and ne natural aggregates properties.
Coarse aggregate 1 (kg/m3) 717
Properties FRA FNA
Coarse aggregate 2 (kg/m3) 478
Water (l/m3) 188 Dry specic density (kg/m3) 1913 2544
W/C 0.52 Surface dry specic density (kg/m3) 2165 2564
Slump (mm) 70 10 Dry bulk density (kg/m3) 1234 1517
fcm: 7 days (MPa) 21.0 Water absorption (%) 13.1 0.8
fcm: 28 days (MPa) 29.6 Fineness modulus 2.38 2.38
L. Evangelista, J. de Brito / Cement & Concrete Composites 32 (2010) 914 11
composition and mixing has been studied. The results have been neering (LNEC) standard E394 [19]; 0.10 m edge cubic specimens
included in a paper presently submitted for publication and clearly were used. Four test specimens of each composition were kept
show that the method of water compensation during mixing, in or- submerged at 20 3 C until constant mass was conrmed (Ms).
der to achieve a given effective W/C ratio, leads to better perfor- The specimens were then oven dried at 105 5 C until constant
mance results for concretes made with these aggregates, when mass was achieved (Mwd). Water absorption by immersion mea-
compared to those mixed using the traditional method (pre-satu- sures the ratio of water absorbed during the process, by means of:
rating the aggregates).
Ms Mwd
The evolution of the amount of water absorbed by the recycled W% 2:1
M wd
aggregates over time was determined using the method proposed
by Leite [17]. The author devised a test which led to the conclusion Water absorption through capillarity was determined on pris-
that the amount of water absorbed by the ne recycled aggregates matic specimens with 0.10 0.10 m2 base and 0.20 m height
grew during the initial 10 min of immersion, more or less stabi- according to Portuguese standard E393 [20]. The specimens were
lized until the 30 min mark at around 50% of the maximum absorp- placed on a tray in contact with water 5 1 mm deep, and the
tion capacity of the aggregates, and grew again thereafter until apparatus was kept at constant humidity of 60% and temperature
saturation was achieved. of 20 C. Test readings were taken on four test specimens after 3,
The author also quotes Neville [18] in the observation that the 6, 24, and 72 h, following standard procedures, after which time
introduction of cement into the mix blocks the pores of the aggre- the specimens were weighed to determine the amount of water
gates, thus dramatically reducing absorption and even halting it. absorbed.
Because of all this and since the pre-saturation process of the The chloride penetration test was performed according to the
aggregates was going to last between 10 and 20 min before adding Nordtest method [21], with a modied setup: the migration cell
the binder, it was considered necessary to add an amount of water was assembled according to ASTM C1202 [22], as shown in
corresponding to 50% of the maximum water absorption capacity Fig. 1. This test allows the accelerated establishment of resistance
of the aggregates (Table 2: 13.1%), minus the water content already to chloride penetration in concrete by determining the non-steady-
contained by the aggregates (4.3%) in environmental conditions. state migration coefcient of the chlorides, Dnssm, according to Eq.
The composition of the different concrete mixes under analysis (2.2). The results allowed the depth reached by the chloride pene-
is presented in Table 3. Three coarse natural aggregates (CA) were tration front at a pre-determined age to be extrapolated. Four
used, all of them crushed from limestone. The cement used was cylindrical test specimens of 100 mm diameter and 50 mm height
CEM I 42.5R with a content of 380 kg/m3 and this was kept con- were tested for each composition.
stant in all the mixes. A superplasticizer made of modied carbox- p
RT xd a xd
ylates was also used. The content was 1.3% by weight of cement, Dnssm 2:2
zFE t
which was also kept constant in the other mixes.
The mechanical properties of the concrete mixes were also where
tested with very good results [8]. Compressive strength presented U2
little to no variation, even for concrete with 100% replacement ra- E 2:3
L
tio, while the maximum reduction of the modulus of elasticity was
18.5%. The maximum reduction of tensile strength and abrasion E is the electric eld strength (V/m) and U is the absolute value of
resistance was about 30%. Table 4 presents a summary of the ob- applied voltage (V).
tained results. r
RT 1 2cd
a2 erf 1 2:4
zFE c0
2.3. Specimen preparation and testing procedures
Dnssm is the non-steady-state migration coefcient (m2/s), R the gas
constant, (R = 8.314 J/(K mol1)), z the absolute value of ion valence,
Water absorption by immersion was tested using procedures
for chloride (z = 1), F the Faradays constant (F = 9.648 104 J/
established by the Portuguese National Laboratory of Civil Engi-
(V mol)), T the average temperature in the anode between the
Table 3 beginning and the end of the test (K), L the thickness of the speci-
Mixture composition (/m3). men (m), xd the average chloride penetration depth (m), measured
after splitting the specimen in half and applying a 0.1 M silver
RC C30R C100R
% of replacement 0 30 100
CEM I 42.5R (kg) 380
Water (l) 155.8 170.2 180.9
W/C 0.41 0.45 0.48
(W/C) ef 0.41 0.44 0.45
FNA (kg) 668 460 0
FRA (kg) 0 154 509
CA 1 (kg) 409 402 400
CA 2 (kg) 382 376 374
CA 3 (kg) 397 390 388
Superplasticizer (kg) 4.9
Table 4
Mechanical behavior of tested compositions.
RC C30R C100R
Compressive strength (MPa) 59.3 57.3 54.8
Modulus of elasticity (GPa) 35.5 34.2 28.9
Splitting tensile strength (MPa) 3.85 3.65 2.95
Abrasion resistance (mm) 1.96 1.86 1.37
Fig. 1. Migration cells assembled for NT Build 492 test.
12 L. Evangelista, J. de Brito / Cement & Concrete Composites 32 (2010) 914
nitrate solution, t the test duration (s), erf1 the inverse of error
function, cd the chloride concentration at which the colour changes
(cd 0.7 N, for ordinary Portland cement) and c0 is the chloride con-
centration in the cathode (c0 = 2 N).
The test specimens were cylinder-shaped, with a diameter of
100 mm and a height of 50 mm, and were obtained by sawing tal-
ler cylinders of the same diameter. The cylindrical faces of the
specimens were painted with epoxy based paint, in order to seal
the migration cells.
Carbonation resistance was measured in an accelerated state,
with specimens in a carbonation chamber with 5% concentration
of CO2, following the recommendations of RILEM CPC-18 [23].
For each test age, specimens were split in half and immediately
sprayed with a phenolphthalein solution, in order to measure the
carbonation depth. Eight test specimens of each composition were
Fig. 2. Capillary water absorption.
tested.
Table 5
Water absorption by immersion.
V mean; r standard deviation; Cv coefcient of variation. Fig. 4. Capillary absorption versus time for concrete C30R.
L. Evangelista, J. de Brito / Cement & Concrete Composites 32 (2010) 914 13
15
RC
5
Fig. 6. Variation of sorptivity with FRA replacement ratio.
0
materials (SCM), such as silica fume, y ash or blast furnace slag, as
0 20 40 60 80 100
replacement for ordinary Portland cement (OPC), as these are
Time (days)
known to improve sorptivity performance [28].
Fig. 8. Carbonation depth variation versus time.
Table 7
Non-steady-state migration coefcient.
l mean; r standard deviation; Cv coefcient of variation. Fig. 9. Carbonation depth variation with FNA by FRA replacement ratio.
14 L. Evangelista, J. de Brito / Cement & Concrete Composites 32 (2010) 914
p
d Cd t 2:7 References
where d is the carbonation depth (mm), Cd is the carbonation coef- [1] Basheer L, Kropp J, Cleland DJ. Assessment of the durability of concrete from its
cient and t is the time (days). permeation properties: a review. Construct Build Mater 2001;15(23):93103.
[2] Grbl P, Nealen A. Construction of an ofce building using concrete made from
Average carbonation depth results and the best correlation t recycled demolition material. Darmstadt Concr 1998;13:16377.
for each composition are given in Fig. 8. Table 8 gives the correla- [3] Koulouris A, Limbachiya MC, Fried AN, Roberts JJ. Use of recycled aggregate in
tion data for each of the analyzed curves. Carbonation depth in- concrete application: case studies. Sustainable waste management and
recycling: construction demolition waste. In: International Conference,
creased with replacement ratio, which was to be expected since Concrete and Masonry Research Group, Kingston University; September 14
both capillary absorption and chloride penetration exhibited the 15 2004. p. 24557.
same trend. Considering the variation of carbonation depth with [4] Poon C, Chan D. The use of recycled aggregate in concrete in Hong Kong. Res,
Conserv Recycl 2007;50(3):293305.
the replacement ratio of FNA by FRA (Fig. 9), it is inferred that car- [5] Grbl P, Rhl M. German committee for reinforced concrete (DAfStb) code:
bonation depth increases almost linearly with the replacement ra- concrete with recycled aggregates. Sustainable construction: use of recycled
tio, with a maximum increase of about 110% measured at 21 days concrete aggregate, University of Dundee, Concrete Technology Unit;
November 1112 1998. p. 8.
for concrete made only with FRA, and a maximum increase of
[6] Wainwright P, Trevorrow A, Yu Y, Wang Y. Modifying the performance of
about 40%, measured at the same age, for concrete with 30% concrete made with coarse and ne recycled aggregates. In: Third
replacement of FNA by FRA. international RILEM symposium on demolition and reuse of concrete and
masonry; October 1994. p. 31930.
[7] Zaharieva R, Buyle-Bodin F, Skoczylas F, Wirquin E. Assessment of the surface
permeation properties of recycled aggregate concrete. Cement Concr Compos
4. Conclusions
2003;25(2):22332.
[8] Evangelista L, de Brito J. Mechanical behaviour of concrete made with ne
From the results obtained in the tests on concrete mixes where recycled concrete aggregates. Cement Concr Compos 2007;29(5):397401.
ne natural aggregates (FNA) were replaced by ne recycled con- [9] Khatib JM. Properties of concrete incorporating ne recycled aggregate.
Cement Concr Res 2005;35(4):7639.
crete aggregates (FRA), the following conclusions can be drawn [10] Mehta P, Monteiro P. Concrete: microstructure, properties and materials. 3rd
in relation to water permeability, capillary absorption and chloride ed. McGraw Hill; 2005.
diffusion: [11] Olorunsogo FT, Padayachee N. Performance of recycled aggregate concrete
monitored by durability indexes. Cement Concr Res 2002;32(2):17985.
[12] de Schutter G, Audenaert K. Evaluation of water absorption of concrete as a
Water absorption by immersion increases with the replacement measure for resistance against carbonation and chloride migration. Mater
ratio of FNA by FRA, up to a maximum of 46% for concrete made Struct 2004;37(9):5916.
[13] Ann KY, Ahn JH, Ryou JS. The importance of chloride content at the concrete
solely with FRA, compared with a reference concrete using only surface in assessing the time to corrosion of steel in concrete structures.
FNA. Construct Build Mater 2009;23(1):23945.
FRA has a more detrimental effect on resistance to water absorp- [14] Levy S, Helene P. Durability of recycled aggregates concrete: a safe way to
sustainable development. Cement Concr Res 2004;34(11):197580.
tion by capillarity than on water absorption by immersion; the
[15] Faury J. Le Bton. 3rd ed. Paris: Dunod; 1958.
sorptivity coefcient has a relative increase of 70.3% for concrete [16] 555 AC. ACI 555-01: removal and reuse of hardened concrete. American
made with 100% FRA compared with the reference concrete. Concrete Institute; 2001.
[17] Leite M. Evaluation of the mechanical properties of concrete made with
In both cases, it seems that water absorption increases linearly
aggregates from construction and demolition waste, PhD Thesis. Porto Alegre,
with the replacement ratio; this is to be expected since FRA have Brazil; 2001. (in Portuguese).
a more porous structure, which extends to the concrete matrix. [18] Neville A. Properties of concrete. 4th ed. London: Longman; 1995.
The non-steady-state chloride migration coefcient increases [19] LNEC. Concrete: determination of the absorption of water by immersion.
Lisbon: LNEC; 1993.
linearly with the replacement ratio of ne aggregates, reaching, [20] LNEC. Concrete: determination of the absorption of water through capillarity.
in the present case, an increase of 34% for the concrete with total Lisbon: LNEC; 1993.
replacement of FNA compared with the reference concrete with [21] Nordtest. NT Build 492: concrete mortar and cement based repair materials:
chloride migration coefcient from non-steady-state migration experiments.
no FRA. Springer; 1999.
Carbonation resistance is reduced with the addition of FRA to [22] ASTM. C1202-97: standard test method for electrical indication of concretes
the concrete; the CO2 penetration depth increased about 40% ability to resist chloride ion penetration. ASTM; 1997.
[23] RILEM TC. CPC 18 measurement of hardened concrete carbonation depth. In:
for concrete made with 30% of FRA and around 110% for concrete TC R. RILEM recommendations for the testing and use of constructions
made solely with FRA. materials; 1988. p. 568.
For durability reasons the total replacement of the FNA by FRA [24] Gmez-Sobern J. Porosity of recycled concrete with substitution of recycled
concrete aggregate: an experimental study. Cement Concr Res
in a concrete mix may present some serious difculties; how-
2002;32(8):130111.
ever, for smaller replacement ratios (e.g. 30%), the use of FRA [25] Wirquin E, Hadjieva-Zaharieva R, Buyle-Bodin F. Use of water absorption by
for structural concrete production is feasible, contrary to the concrete as a criterion of the durability of concrete application to recycled
aggregate concrete. Mater Struct 2000;33(6):4038.
widespread notion (even in codes) that they are inappropriate
[26] Hall C. Water sorptivity of mortars and concretes. Mag Concr Res
for this purpose. 1989;26(8):5161.
Considering that FRA have a non-negligible effect on concrete [27] Balayssac JP. Relations entre performances mcaniques, microstructure et
performance, it is feasible that the use of binders other than durabilit des btons PhD thesis held at INSA. Tolouse; 1992.
[28] Papadakis VG. Effect of supplementary cementing materials on concrete
ordinary Portland cement (such as SCM) could compensate for resistance against carbonation and chloride ingress. Cement Concr Res
the presence of FRA, maybe leading to higher acceptable 2000;30(2):2919.
replacement ratios. [29] Gomes M, de Brito J. Structural concrete with incorporation of coarse recycled
concrete and ceramic aggregates: durability performance. Mater Struct
Furthermore, considering the acceptable mechanical behavior of 2009;42(5):66375.
these materials [8] and the fact that some structures are not [30] Limbachyia MC, Leelawat T, Dhir RK. Use of recycled concrete aggregate in
affected by durability issues (e.g. concrete elements protected high-strength concrete. Mater Struct 2000;33(9):57480.
[31] Abbas A, Fathifazl G, Isgor OB, Razaqpur AG, Fournier B, Foo S. Durability of
from natural aggressive agents), concrete FRA can be a useful recycled aggregate next term concrete designed with equivalent mortar
source of material for concrete production and could potentially volume method. Cement Concr Compos 2009.
replace FNA in some concrete mixes.