Construction and Building Materials 23 (2009) 1292–1297

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Mechanical properties of brick aggregate concrete

Paulo B. Cachim *
University of Aveiro & LABEST, DECivil – University of Aveiro, 3810-193 Aveiro, Portugal

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents the results of a research program carried out at University of Aveiro, Portugal to eval-
Received 10 October 2007 uate the properties of concrete made with crushed bricks replacing natural aggregates. Two types of brick
Received in revised form 21 July 2008 were investigated. The bricks were crushed in order to obtain a usable aggregate. The properties inves-
Accepted 24 July 2008
tigated were the workability and the density of fresh concrete, and the compressive strength, tensile
Available online 7 September 2008
splitting strength, modulus of elasticity and stress–strain behaviour of hardened concrete. Replacement
ratios of natural aggregates by 15% and 30% were investigated as well as water/cement ratios of 0.45 and
Keywords:
0.5. Strength indexes were used to assess the effectiveness of aggregate replacement. The results of con-
Recycling
Aggregates
crete produced with recycled aggregates were compared with a reference concrete produced with natural
Concrete limestone aggregates currently used in Portugal. Observed results indicate that ceramic residuals could
Brick be used as partial replacement of natural aggregates in concrete without reduction of concrete properties
for 15% replacement and with reductions up to 20% for 30% replacement. The type and the manufacturing
process of bricks seem to influence the properties of the resulting concrete. The properties and aesthetics
of concrete with bricks indicate the possibility of using this type of concrete in precast applications.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction construction industry. Most of the wasted of the manufacturers

is already incorporated as raw material for new ceramic materials.
The sustainable construction concept was introduced due to the Nevertheless, part of these wastes and those produced by the con-
growing concern about the future of our planet because construc- struction industry are placed in landfills. The studies in this field
tion industry is a huge consumer of natural resources and, simulta- showed that some problems arise when using bricks from CDW be-
neously, waste producer. Concrete industry, in particular, is one of cause of its contamination with mortar, timber and concrete [8].
the biggest natural resource consumer as a consequence of being Akhtaruzzaman and Hasnat [9] carried out some research using
one the most used construction materials. Concrete is a material well-burned brick as coarse aggregate in concrete where they
that is often seen as a potential place for wastes, because of its found that it was possible to achieve concrete of high strength
composite nature (a binder, water and aggregates) and because it using crushed brick as the coarse aggregate. Brito et al. [10] re-
is widely used, which means that if a waste could be used in con- ported strength ranging from 22% to 45%, respectively for 33%
crete, then certainly large quantities of it can be recycled. Since and 100% of natural aggregate replacement. Debieb and Kenai
aggregates in concrete comprise about 60% to 75% of the total vol- [11], using both coarse and fine crushed bricks reported a decrease
ume of concrete any reduction in natural aggregates consumption in strength ranging from 20% to 30% depending of the degree of
will have significant impacts in the environment. Environmental substitution. Khatib [12] and Poon and Chan [13] also reported a
constrains of stone pits, such as noise, dust, vibrations, consider- decrease of strength when using crushed bricks as fine aggregate
able impact on the countryside, besides the consumption of a substitute. Khaloo [14] and Hansen [15] found that there is an in-
non-renewable material tend to considerably limit their exploita- crease in tensile strength if crushed brick aggregates are used in
tion. Consequently, alternative materials such as construction concrete compared to concrete made with natural aggregates.
and demolition waste (CDW) as well as other industries by-prod- Regardless of all limitations that may exist, the utilization of
ucts are increasingly being tested and used as environmental sus- recycled aggregates in concrete will certainly be mandatory in a
tainable natural aggregates substitutes, see for instance [1–7]. near future.
Ceramic materials are largely used in Portugal, both as bricks The aim of this paper is the assessment of the properties of con-
and tiles. Consequently, big quantities of wastes are produced crete which use crushed bricks as natural coarse aggregate partial
simultaneously by the brick and tile manufacturers and by the substitute. To achieve this, fresh and hardened state properties of
concrete are measured to ensure that fundamental parameters
* Tel.: +351 234370049. needed in design are evaluated (such as slump, compressive and
E-mail address: [email protected] tensile strength, stress–strain relation and modulus of elasticity).

0950-0618/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2008.07.023
 P.B. Cachim / Construction and Building Materials 23 (2009) 1292–1297 1293

2. Materials 100

All the materials used in this study were commonly available in 80

the central region of Portugal. The cement used was ordinary Type

Absortion [%]
II 32.5 Portland cement complying with NP EN 197-1:2001. Natural
60
sand and calcareous coarse aggregates were used. Coarse aggre-
gates were divided into two groups; one in the range 5–10 mm
(NA-1) and the other in the range 10–20 mm (NA-2). Sieve analysis 40
of all types of aggregates is shown in Table 1.
Recycled aggregates were ceramic bricks obtained as construc- Brick A
20
tion waste, from two local industries (namely types A and B), that Brick B
were crushed to obtain suitable sizes for using in concrete. The 0
final grading of both types of crushed bricks was very similar 0 100 200 300 400 500 600
(Table 1). A photo of the crushed brick of type A is presented in Time [s]
Fig. 1.
Prior to crushing the bricks were tested in compression accord- Fig. 2. Water absorption during first minutes.
ing to NP 80:1960 and the results were 1.70 MPa for brick A and
2.55 MPa for brick B. The oven temperature at which bricks were
aggregates in a cylindrical shape to which a pre-determined stress
made was, according to manufacture data, 850–870 °C and 800–
was applied. The results measure the amount of material trans-
850 °C for bricks A and B, respectively.
formed into powder. The results, presented in Table 2 showed that
One of the most important aspects of ceramic aggregates is their
the natural aggregate is better than recycled bricks and that brick B
high water absorption capacity. This is of extreme importance dur-
is better than brick A, which is in agreement with those obtained
ing the mixing procedure of concrete. Then, in addition to the total
for compression of bricks.
water absorption capacity of these aggregates, a curve of the evo-
Density tests were performed using NP EN 1097-6:2003. The re-
lution of the water absorption with time was determined for cera-
sults, presented in Table 2 showed that bricks had smaller density
mic aggregates (Fig. 2). This allows the quantification of water
than natural aggregates which leads to lighter concretes. Another
absorbed by ceramic aggregates during the mixing process that is
property evaluated was the shape index that was measured using
very important for the workability of concrete. It can be observed
NP EN 933-4:2002. This test calculates the amount of particles that
that during the first 2 min at least 75% of total absorption as oc-
do not have a cubic shape, defined as the ratio between the maxi-
curred while after 5 min this value increases for at least 91% of
mum and minimum dimension being less than 3. Consequently, as
the total absorption.
the index increases, the amount of cubic particles decreases. The
Mechanical strength of aggregates was measured using NP
results, presented in Table 2 indicate that natural aggregate and
1039:1974. This test was performed by placing an amount of

Table 1
Sieve analysis of aggregates (NP EN 932-1:2000)
Table 2
Sieve (mm) Sand NA-1 NA-2 Brick A Brick B Physical properties of aggregates

31.5 100.0 100.0 100.0 100.0 100.0 Property Standard Units Natural Recycled
16.0 100.0 100.0 84.0 100.0 100.0
Sand NA-1 NA-2 Br-A Br-B
8.0 100.0 78.0 2.1 99.0 99.0
4.0 100.0 1.9 0.2 39.5 46.2 Crushing strength NP 1039 % 21.7 30.8 27.3
2.0 84.4 0.29 0.18 6.53 11.5 Water absorption NP EN 1097-6 % 0.79 1.33 1.07 15.81 18.91
1.0 53.5 0.29 0.17 2.55 3.05 Shape index NP EN 933-4 % – 13 7 30 16
0.50 22.7 0.17 0.16 1.67 1.04 Saturated 2813 2670 2652 2146 2233
0.250 5.7 0.13 0.15 1.33 0.67 Density NP EN 1097-6 kg/m3
0.125 1.8 0.12 0.12 1.13 0.55 Dry 2791 2641 2617 1805 1928

Fig. 1. Crushed bricks of type A.

1294 P.B. Cachim / Construction and Building Materials 23 (2009) 1292–1297

brick B had similar shape index (13 and 16, respectively), while
brick A had a substantially bigger shape index. Since the crushing 16
process of the two bricks was exactly the same, the differences in

Slump (cm)
results were caused by the nature of the brick itself. Again, this 12
seems to indicate the lower quality of brick A. Average slump w/c = 0.50
8 w/c = 0.45

3. Methodology
4
3.1. Concrete mix proportions
0
The main objective of the experimental program was the NN NA NB AA BB
Concrete series
assessment of concrete properties made with waste clay brick
aggregates produced from crushed bricks. Two series of tests were Fig. 3. Slump of concrete.
carried out, one with a water/cement ratio (W/C) of 0.5 and other
with W/C of 0.45. The water absorbed by aggregates was not in-
cluded for the calculation of W/C. This methodology ensures that 4. Results and discussion
the workability of fresh concrete could be kept constant for each
W/C. Water saturation of aggregates was achieved by placing them 4.1. Concrete workability
in the mixer during 2 min. The water absorbed by the aggregates
during this period of time should be added to the concrete mix. Workability of concrete is measured with the slump test,
Aggregate replacement was made only with NA-1 that is the finer according to NP EN 12350-2:2002. The average slump for
portion of coarse aggregate, due to final grading of the crushing W/C = 0.45 was 5 cm and for W/C = 0.5 was 15 cm, as shown in
process. This option was determined by the similarity of the grad- Fig. 3. Apart from the amount of natural aggregate replaced, the
ing curve of these aggregates. Partial (50%) and total (100%) workability was kept approximately constant for each W/C,
replacement of NA-1 has been made, leading to 15% and 30% of confirming the efficiency of the adopted mixing procedure in
natural coarse aggregate substitution. The substitution of natural maintaining the workability of concrete.
by recycled aggregate was made in volume.
The final mix proportions for all types of studied concrete are 4.2. Density
shown in Table 3. Five different mixes were studied for each W/C
ratio: a reference mix with natural aggregates (NN), 2 mixes with Density of fresh concrete was determined using NP EN 12350-
50% of NA-1 aggregates replaced by bricks (namely NA and NB, 6:2002. Results exhibit a clear and obvious reduction of density
for brick A and B respectively) and 2 mixes with 100% of the NA- when recycled aggregates were used. Nevertheless, the reduction
1 natural aggregate replaced by crushed bricks (namely AA and of density was only 5% and 6% for concrete with W/C = 0.45 and
BB). 0.5, respectively (see Fig. 4).

3.2. Mixing procedure

2600
Concrete manufacturing process was conducted as follows.
Fresh density (kg/m3)

First, the water content of aggregates was calculated. Afterwards,

2500
coarse aggregates were placed in the mixer along with the water
for 5 min. The water placed in the mixer was that indicated in
Table 3 plus 80% of the total absorption water of aggregates. This 2400
procedure was used to ensure that aggregates absorbed part of
the water and that the pores were at least partly saturated.
2300 w/c = 0.50
Additionally, this guaranteed that aggregate surface was wet
when cement was added. Cement and fine aggregate were then w/c = 0.45
added to the mixer. The total mixing time was about 15 min. This 2200
procedure, due to the presence of water at the aggregate surface, NN NA NB AA BB
ensures that a large quantity of cement particles will adhere to Concrete series
aggregate, thus causing a good connection between cement paste
and aggregates. Fig. 4. Fresh density of concrete.

Table 3
Summary of concrete mixes used in this study

Materials Concrete series name

NN45 NA45 NB45 AA45 BB45 NN50 NA50 NB50 AA50 BB50
Cement (kg/m3) 400 400 400 400 400 400 400 400 400 400
Water (kg/m3) 180 180 180 180 180 200 200 200 200 200
Sand (kg/m3) 713 713 713 713 713 693 693 693 693 693
NA-1 (kg/m3) 398 199 199 – – 386 193 193 – –
NA-2 (kg/m3) 748 748 748 748 748 727 727 727 727 727
Brick A (kg/m3) – 137 – 274 – – 133 – 267 –
Brick B (kg/m3) – – 147 – 293 – – 142 – 293
Saturation water (kg/m3) – 20 18 42 33 – 19 17 41 32
W/C (by wt.) 0.45 0.45 0.45 0.45 0.45 0.50 0.50 0.50 0.50 0.50
 P.B. Cachim / Construction and Building Materials 23 (2009) 1292–1297 1295

4.3. Compressive and tensile strength

Splitting tensile strength [MPa]

3.5

3.0
Concrete curing was carried out with specimens immersed in
water. Compressive strength of concrete was measured in 15 cm 2.5
cubes according to NP EN 12390-3:2003 and the results are pre- 2.0
sented in Table 4 for 7, 14, 28 and 90 days. Results show that brick
1.5
A always gave smaller values than brick B. Taking into consider- w/c = 0.50
ation that compressive strength of brick B is higher, that it has bet- 1.0
w/c = 0.45
ter crushing strength and that its shape index is closer to that of 0.5 EC2
natural aggregates these are predictable results. For substitution
of 15% of natural aggregates by bricks no reduction of strength is 0.0
20 25 30 35 40
observed. For series NB45 a 10% increase in strength at 90 days
is observed. When all NA-1 aggregate was replaced by crushed Compressive strength [MPa]
bricks, there was always a decrease of the compressive strength.
Fig. 5. Relation between compressive and tensile splitting strength.
Tensile splitting strength of concrete was measured in 30 cm
cylinders using NP EN 12390-6:2003 and the results are showed
in Table 4. Concrete was tested at 28 days. The conclusions are sim-
ilar to those mentioned in compressive tests. The relation between 40

Modulus of elasticity [GPa]

compressive and tensile strength is plotted in Fig. 5 simultaneously 35
with Eurocode 2, EN 1992-1-1:2004 curve. All the results agree
30
relatively well with the proposed curve, although concretes with
W/C = 0.45 are closer to the curve. 25
20
4.4. Modulus of elasticity 15
W/C = 0.50
10
Modulus of elasticity of concrete was measured on 30 cm cylin- W/C = 0.45
ders with strains being measured in the central 10 cm of the cylin- 5
EC2
der. Load was applied cyclically approximately between 10% and 0
30% of ultimate load. The results are shown in Table 4 where it 20 25 30 35 40
can be seen that, basically, results are similar to those obtained Compressive strength [MPa]
for strength. This can be shown in Fig. 6 that illustrates the relation
Fig. 6. Relation between compressive strength and modulus of elasticity.
between compressive strength and the modulus of elasticity. The
curve proposed by Eurocode 2 is also plotted. It can be observed
that the results agree relatively well with Eurocode 2 curve and are plotted in Fig. 7. In order to compare the post-peak behaviour,
are independent of the type of aggregate used. Again the results the curves are normalised by its maximum stress. It can be ob-
for W/C = 0.45 have better agreement with proposed curve than served that the results are very similar independently of the type
concrete with W/C = 0.5. of aggregate used: natural, brick A or brick B. The results for con-
cretes with W/C = 0.5 are similar.
4.5. Stress–strain curves
4.6. Strength indexes
The stress–strain behaviour of concrete was assessed by com-
pressive tests on cylinders. Displacement was measured between In order to quantify the strength behaviour with the amount of
the load platens, and strain was calculated dividing the displace- natural aggregate replacement, strength indexes, following an ap-
ment by the cylinder height. This important feature of concrete proach similar to that in [16], were introduced. The specific
behaviour is often not assessed. However it is extremely important strength ratio, R, is defined by
in order to correctly model concrete, since the softening branch of
the curve might have some effect on the failure behaviour of ele- R ¼ f =q; ð1Þ
ments and structures. Since bricks are weaker than natural aggre- where f is concrete compressive strength and q is the amount of
gates, differences in the stress–strain curves could be expected. natural coarse aggregate. Rn represents the contribution of natural
The results obtained for both types of bricks and for a W/C = 0.45 coarse aggregate to strength and is consequently defined as the spe-
cific strength ratio for concrete with natural aggregates. Rb ex-
presses the contribution of unit natural aggregate to strength
Table 4
Concrete hardened properties (compressive, fc, and tensile, fct, strength and modulus Rn = Rb (q = 100%), while Rr is the contribution of the brick aggre-
of elasticity, Ec) gates to strength:
Age => fc (MPa) fct (MPa) Ec (GPa) Rr ¼ Rb  Rn : ð2Þ
7 14 28 90 28 28
The specific strength ratio, K, can be defined as
NN50 20.8 26.6 30.5 33.5 2.34 32.9
AA50 16.8 22.5 24.5 28.3 2.44 28.5 K ¼ Rb =Rn : ð3Þ
NA50 22.6 26.4 29.4 34.0 2.25 35.6
BB50 21.6 24.8 29.0 31.6 2.38 35.9 and is bigger than 1 if the strength reduction is smaller than the
NB50 24.3 26.7 32.3 34.1 2.72 32.4 percentage of natural aggregate replaced (or if strength increases).
NN45 29.0 32.2 36.2 40.1 2.97 35.2 Additionally the contribution of aggregate replacement to strength,
AA45 20.0 26.1 27.6 32.3 2.38 31.4
P, represents the percentage of the contribution of brick aggregates
NA45 26.9 29.0 32.1 40.0 2.72 33.7
BB45 25.4 29.0 32.3 36.0 2.67 32.6
to concrete strength and is defined as
NB45 26.1 31.5 38.5 44.8 2.97 34.7 P ¼ 100ðRr =Rb Þ: ð4Þ
1296 P.B. Cachim / Construction and Building Materials 23 (2009) 1292–1297

1.2 Table 5
Strength indexes
Relative compressive strength

1 AA45 Series q Rb Rr K P, 28 P, 90
NA45 NN50 100 0.31 0.00 1.000 0.0 0.0
0.8 NN45 NA50 85 0.35 0.04 1.134 11.8 16.3
AA50 70 0.35 0.05 1.148 12.9 17.1
NB50 85 0.38 0.08 1.246 19.7 16.5
0.6 BB50 70 0.41 0.11 1.358 26.4 25.8
NN45 100 0.36 0.00 1.000 0.0 0.0
0.4 NA45 85 0.38 0.02 1.043 4.1 14.8
AA45 70 0.39 0.03 1.089 8.2 13.1
NB45 85 0.45 0.09 1.251 20.1 23.9
0.2 BB45 70 0.46 0.10 1.275 21.5 22.0

0
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 quently, concrete produced with crushed bricks have additional
Strain water that could be used for the hydration of cement at later stages
and, simultaneously, does not seem to contribute for the porosity
1.2
of concrete. Thus, the moderate use of saturated bricks in concrete
Relative compressive strength

act as self-curing agent for concrete. Additionally, if the water that

1 is inside bricks is not used for cement hydration and remained in-
NB45
BB45
side the bricks, then it fills brick pores creating a denser aggregate.
0.8 The correct balance between the reduced strength of crushed brick
NN45
aggregates and its effectiveness as self curing agent is the key point
0.6 to use them as aggregates in concrete.

0.4
5. Conclusions
0.2
The experimental results clearly show that crushed bricks can
be used as natural aggregates substitutes in percentages up to
0
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007
15% without strength reduction. For 30% of natural aggregate sub-
Strain stitution, there is a reduction of concrete properties (up to 20%,
depending on the type of brick). The use of strength indexes allows
Fig. 7. Stress–strain curves for concrete with W/C = 0.45. a better understanding of the effect of aggregate replacement. The
uses of strength indexes allowed a better comparison between
results.
The definition of P implies that if it is bigger than the amount of The stress–strain relations are very similar for both types of
replaced natural aggregate there is a positive contribution of the concrete, even in the softening branch, corroborating the use of
replaced aggregate to strength. The results of the defined indexes brick aggregate concrete in low demanding structural applications.
for concrete tested in compression at 28 days are shown in Table Crushed bricks, when moderately used as natural aggregate
5. together with the value of P at 90 days. With the above men- substitutes, may act as self curing agents for concrete when they
tioned definitions it is clear that 15% replacement with brick B are pre-saturated, because they can provide water for cement
has a positive contribution to strength (P = 19.7 and 20.1 for hydration that does not affect the initial water/cement ratio. When
W/C = 0.5 and 0.45, respectively) while for 30% replacement the the quantities of crushed bricks increase, this advantageous effect
values of P are 26.4 and 21.5 indicate that the contribution to is overridden by the augment in porosity of the bricks, leading to
strength is somewhat smaller than the amount of natural aggre- poor results.
gate replaced. For brick A the results are slightly different. While The type of brick seemed to affect the results, at least to some
at 28 days all replacements had a negative influence on strength, extent, meaning that some caution must be taken when crushed
at 90 days only concrete with 30% of bricks had a negative effect bricks are used as natural aggregate substitution, because in many
on strength. cases the type, quality and origin of recycled bricks is unknown.
An interesting facet that might also be considered is the aes-
4.7. Discussion thetic possibilities that concrete produced using crushed bricks
hold and that could be explored for instance in the case of urban
The strength reduction observed for concrete with 30% of aggre- equipment.
gate replacement is in agreement with the studies of Brito et al.
[10] and Akhtaruzzaman and Hasnat [9]. However, the attained
strength showed that crushed bricks can be used to substitute nat- References
ural aggregate in concrete (24.5–38.5 MPa at 28 days) for percent-
[1] Chen HJ et al. Use of building rubbles as recycled aggregates. Cement Concrete
ages about 15% to 20%. Res 2003;33(1):125–32.
The results indicate that the analysed properties are not very af- [2] Katz A. Properties of concrete made with recycled aggregate from partially
fected by the substitution of natural aggregate by crushed bricks. hydrated old concrete. Cement Concrete Res 2003;33(5):703–11.
[3] Khalaf FM, Devenny AS. Recycling of demolished masonry rubble as coarse
The reason for this may be due related to an internal curing effect
aggregate in concrete: review. J Mater Civil Eng 2004;16(4):331–40.
of the crushed brick. As bricks are saturated when they are used, [4] Olorunsogo FT, Padayachee N. Performance of recycled aggregate concrete
and taking in consideration the absorption of bricks, inside con- monitored by durability indexes. Cement Concrete Res 2002;32(2):179–85.
crete there is a further percentage of 8–23% excess water than that [5] Rao A, Jha KN, et al. Use of aggregates from recycled construction and
demolition waste in concrete. Res Conserv Recyl 2007;50(1):71–81.
specified by the water/cement ratio. However, this water is inside [6] Robinson GR, Menzie WD, et al. Recycling of construction debris as aggregate
bricks and is not available during the mixing process. Conse- in the Mid-Atlantic Region, USA. Res Conserv Recyl 2004;42(3):275–94.
 P.B. Cachim / Construction and Building Materials 23 (2009) 1292–1297 1297

[7] Tu TY et al. Properties of HPC with recycled aggregates. Cement Concrete Res [12] Khatib JM. Properties of concrete incorporating fine recycled aggregate.
2006;36:943–50. Cement Concrete Res 2005;35:763–9.
[8] Tam VWY, Tam CM. A review on the viable technology for construction waste [13] Poon C-S, Chan D. The use of recycled aggregate in concrete in Hong Kong. Res
recycling. Res Conserv Recyl 2006;47(3):209–21. Conserv Recyl 2007;50(3):293–305.
[9] Akhtaruzzaman AA, Hasnat A. Properties of concrete using crushed brick as [14] Khaloo AR. Properties of concrete using crushed clinker brick as coarse
aggregate. ACI Con Int Design Constr 1983;5(2):58–63. aggregate. ACI Mater J 1994;91(2):401–7.
[10] Brito J et al. Mechanical behaviour of non-structural concrete made with [15] Hansen TC. Recycling of demolished concrete and masonry. RILEM Report 6.
recycled ceramic aggregates. Cement Concrete Compos 2005;27(4):429–33. E&FN Spon, London; 1992.
[11] Debieb F, Kenai S. The use of coarse and fine crushed bricks as aggregate in [16] Yu LH et al. Investigation on pozzolanic effect of perlite powder in concrete.
concrete. Constr Build Mater 2008;22(5):886–93. Cement Concrete Res 2003;33(1):73–6.