Construction and Building Materials 113 (2016) 246–254

Contents lists available at ScienceDirect

Construction and Building Materials

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

Investigation of using recycled powder from waste of clay bricks and

cement solids in reactive powder concrete
Peng Zhu a, XinQi Mao a,b, Wenjun Qu a,⇑, Zongyang Li a, Zhongguo John Ma c,d
a
Department of Structural Engineering, Tongji University, 1239 Siping Road, Shanghai 200092, China
b
Department of Engineering, Qingdao Agricultural University, 700 Changcheng Road, Qingdao 266109, China
c
Department of Civil and Environmental Engineering, University of Tennessee, Knoxville, TN 37996, USA
d
School of Civil Engineering, Southwest Jiaotong University, Sichuan 610031, China

h i g h l i g h t s

 Use of the recycled powder from construction and demolition wastes in RPC is studied.
 Environmentally-friendly and cost-saving RPC mixtures with high performance are developed.
 Natural sand can be used instead of crushed quartz in the green RPC mixtures.
 Mix proportions of the green RPC are studied through theory and designed experiments.

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

Article history: Recycled powder, produced from the construction and demolition wastes, contains unhydrated cement
Received 4 January 2016 particles. It can reduce environment pollution to use recycled powder as a cementing material. It is inves-
Received in revised form 7 March 2016 tigated to use recycled powder to partially replace silica fume or cement in Reactive Powder Concrete
Accepted 8 March 2016
(RPC) to develop environmentally-friendly and cost-saving RPC mixture with high performance.
Available online 16 March 2016
Properties of the recycled powder are tested firstly. And according to maximum packing theory, the
RPC mix with the recycled powder is designed. Influences of the fine aggregate, quartz and natural sand,
Keywords:
on the RPC are investigated. The water-cementitious materials ratio (w/cm) for RPC mixes with the recy-
Recycled powder
Reactive powder concrete
cled powder are studied and selected. The recycled powder is used to replace the silica fume and cement
Strength in RPC respectively, and influences of the replacement ratio on the flowability, strength and durability are
Flowability investigated. The standard curing is used for all the tests instead of steam curing normally required for
Shrinkage RPC. Considering the flowability, strength, durability, cost and environmental savings, it is suggested that
Chloride penetration the recycled powder can be used to replace the silica fume and cement in RPC partially, and natural sand
can be used instead of quartz. Additionally, influences of GGBFS powder on RPC mix with the recycled
powder are investigated.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction cement dosage is normally 800–1000 kg/m3, SF content is often

over 25% (by the weight of cement), and crushed quartz is 40%
RPC was developed with dense microstructure and reduced (by the weight of cement). This increases the production cost. Also
defects (pores and micro-fractures), and thus ultra-high strength high amounts of cement and silica fume can raise the hydration
and high durability can be ensured. Compressive strengths of heat and may cause shrinkage problems. To overcome these prob-
200–800 MPa and flexural strength of 140 MPa have been achieved lems, some mineral admixtures were used to replace cement or
with RPC. And with the steel fiber, RPC also exhibits remarkable reduce the SF amount [3]. Yazici et al. (2009) [3] found that RPC
ductility, about 250 times higher than that of conventional con- containing high volume mineral admixtures (fly ash (FA) and
crete. Typical compositions of RPC include cement, silica fume ground granulated blast furnace slag (GGBFS)) have satisfactory
(SF), crushed quartz, superplasticizer and steel fiber [1,2]. The mechanical performance (compressive strength, flexural strength,
and toughness). Compressive strength exceeded 200 MPa after
standard water curing, although the cement and silica fume
⇑ Corresponding author.
E-mail address: quwenjun.tj@tongji.edu.cn (W. Qu).

http://dx.doi.org/10.1016/j.conbuildmat.2016.03.040
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254 247

Table 1
Chemical composition of cementitious materials.

Constituent SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3

Cement (%) 23.25 7.79 2.41 46.78 0.459 0.116 0.574 4.10
Silica fume (%) 94.51 0.87 0.10 0.43 1.56 0.082 0.33 –
Slag powder (%) 28.48 13.88 – 36.79 9.06 0.43 0.33 1.32
Recycled powder (%) 53.8 13.2 5.15 13.6 2.58 0.65 2.77 –

Table 2
Physical and mechanical properties of cement.

Fineness (%) Density (g/cm3) Setting time (min) Flexural strength Compressive strength
(MPa) (MPa)
Initial Final 3d 28d 3d 28d
5.7 3.13 150 490 5.5 7.6 24.6 44.3

contents of the mixtures lower than conventional RPC. SF can also of packing properties of pozzolanic minerals were successful ways
be reduced by increasing GGBFS and/or FA content. to obtain super-high performance cementitious materials (Reac-
Kasami et al. (2001) [4] investigated recycled concrete powder, tive Powder Concrete).
by-product in the process of producing recycled aggregate. It was The mix proportion is designed based on Dinger-Funk equation
found that the recycled concrete powder can be used for self- [14] of the maximum packing theory. Influences of recycled pow-
compacting concrete without further processing, although the der on the strength, workability and durability of RPC are investi-
dosage of high-range water reducer for a given slump-flow and gated through designed experiments.
in drying shrinkage may increase. The superplasticizing effect of
high-range water reducer and properties of concrete can be 2. Experiment preparation
improved by the addition of recycled concrete powder together
with ground blast-furnace slag to self-compacting concrete. 2.1. Materials
Malhotra and etc. (1996) [5] reported that clay brick powder
P
O 42.5 ordinary Portland cement conforming to GB175-2007 [15] was used,
can be used as a pozzolanic supplement like fly ash because of and the chemical composition and physical and mechanical properties are listed
the rich pozzolanic ingredients. It can replace part of cement in in Tables 1 and 2 respectively. The chemical composition of silica fume used is listed
concrete, and the expansion induced by the alkali–silica reaction in Table 1. The average grain size is 0.26 lm and the specific surface area is over
could be suppressed [6]. Kartini et al. (2012) [7] investigated 20 m2/g. S95 GGBFS powder conforming to GB/T18046-2008 [16] was used, and
the chemical composition and physical and chemical properties are listed in Tables
10%, 20% and 30% replacement of cement in concrete by recycled
1 and 3 respectively.
clay brick powder. 30% replacement still attained strength of grade Recycled powder in this study was made through smashing, grinding, drying
30 concrete, and water absorption and water permeability can be and grading abandoned clay bricks and cement solids. The chemical compositions
reduced efficiently. Wild et al. (1997) [8] investigated the ground of the recycled powder were summarized in Table 1. The density measured was
brick from different European countries. It was confirmed that all 2.63 g/cm3.
Crushed quartz used has the grain diameter range of 200–650 lm and the aver-
the ground brick types studied exhibit pozzolanic activity by both
age grain size of 280 lm. The natural sand used has the grain size range of 150–
results of the chemical tests for pozzolanicity and also the results 8000 lm and the average grain size of 300 lm. Superplasticizer used was a type
for the strength development of the mortars. Baronio et al. of polycarboxylate agent. The solid content is 41% and the water reduction rate is
(1997) [9] concluded that a good degree of hydraulicity of the mor- 31.8%.
The particle size distribution of recycled powder was measured by laser particle
tar could be developed with finely powdered clay bricks.
size analyzer. The result is shown in Fig. 1. The average grain size was 31.4 lm. The
Large amount of construction and demolition wastes of clay amount of particles with a grain size under 10 lm was high.
bricks and cement solids are produced. It is possible to use the The water demand of recycled powder was tested following GB/T 18736-2002
recycled powder of clay bricks and cement solids as a cement sup- [17] and GB/T17671-1999 [18]. The dosage suggested for the fly ash was used for
plement, considering its pozzolanicity. Use of the recycled powder the recycled powder, considering the fineness (15.8% for fly ash and 6.8% for recy-
cled powder) and the particle size distribution (Fig. 1) of the two materials are sim-
from construction and demolition wastes in RPC is studied, and
ilar. The ratio of water demand is 99.3%, and it can be concluded that the water
environmentally-friendly and cost-saving RPC mixtures with high demand of recycled powder is similar to that of cement.
performance are developed. Mix proportions of the green RPC are Activity index of the recycled powder was tested following GB/T 18736-2002
studied through theory and designed experiments. [17] and GB/T17671-1999 [18]. The standard sand by GB/T17671-1999 [18] was
used. The compressive strength and flexural strength at 28 days were tested. The
Concrete mix design is always desirable to compose the aggre-
compressive strength for the mix with the recycled powder was 71% of the control
gates as densely as possible, i.e. with maximum packing [10]. The mix and the flexural strength was 68% of the control mix. The results indicated that
performance of a concrete mix can be optimised by maximising the recycled powder had pozzolanic activity.
the packing densities of the aggregate particles and the cementi-
tious materials. It is believed that maximisation of the packing 2.2. Specimen preparation
density of the cementitious materials and the packing density of
the aggregate particles could improve the overall performance of Specimens for strength and shrinkage tests were prepared following GB/
T17671-1999 [18] and JC/T 603-2004 [19] respectively. The cementitious materials
the resulting concrete mix and thus should be the general guide-
were mixed for 1 min in the mixer. Then 70% water and 70% superplasticizer were
line for mix optimisation [11]. Lange and etc. [12] and de Larrard
and Sedran [13] demonstrated that the increase of packing density
of the cementitious materials can reduce the water/cementitious Table 3
materials (W/CM) ratio, decrease the porosity, speed up the Physical and chemical properties of GGBFS powder.
hydration process, and improve the microstructure which result Specific surface area Density (g/ Activity index at Cl Loss on
in good mechanical properties and great durability. Investigation (m2/kg) cm3) 28 days (%) (%) ignition (%)
of Richard and Cheyrezy [1,2] showed that the incorporation of 437 2.85 97 0.03 2.8
ultra-fine powders with different grain sizes and the improvement
248 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254

Table 4
Designed mix proportion.

Cement Silica fume Recycled powder Crushed quartz

Control mix 1 0.25 — 1.0
Recycled powder mix 1 0.1 0.15 1.0

3. Results and discussion

3.1. Fine aggregate comparison

Influences of different fine aggregate on properties of RPC were

studied. Based on results of tests on mixture in Table 4, the mix
design was modified as shown in Table 5. The amount of super-
Fig. 1. The particle size distribution of recycled powder and fly ash. plasticizer used was based on the manufacturer’s recommenda-
tion, which is 2% of the mass of cementitious materials. The
same amount of superplasticizer was used for the following tests
added and mixed for 3 min. The sand was added and mixed for 1 min. The remain- in the paper. Test results were shown in Table 5.
ing 30% water and 30% superplasticizer were added and mixed for 5 min. After
Based on Table 5, for RPC with the recycled powder, the flowa-
these, the mixture was poured into moulds of 40 mm  40 mm  160 mm (for
strength tests) or 25 mm  25 mm  280 mm (for shrinkage tests) and consoli- bility of the mixture with natural sand was better than that with
dated using a vibration table. Then the specimens were maintained in the standard crushed quartz, the flexural strengths of two mixtures were simi-
curing case (temperature 20 ± 1 °C and moisture above 90%) for 24 h. After demold- lar, and the compressive strength of the mixture with natural sand
ing, the specimens were kept in the curing room (temperature 20 ± 2 °C and mois- was slightly smaller.
ture above 50%) till the tests.
Specimens for chloride-penetration tests were prepared following ASTM C192
The size distribution of natural sand is more continuous and
[20] and ASTM C1202 [21]. more even than crushed quartz, as shown in Fig. 3, which can
result in the mixture with natural sand more dense than that with
crushed quartz. Based on Dinger-Funk equation of the maximum
2.3. Test methods packing theory, the mix with natural sand and crushed quartz
Flowability was tested following GB/T2419-2005 [22], and the compressive
was designed respectively, and fitting curves were shown in
strength and flexural strength following GB/T17671-1999 [18]. Flexural specimens Fig. 3. The fitting curve for the natural sand mix is closer to the
were tested at a loading rate of 50 N/s. The specimens were loaded from their mid- objective curve than crushed quartz, which means the mixture
span and the clear distance between simple supports was 100 mm. The two broken with natural sand is denser than that with crushed quartz. How-
pieces left from the flexural test were subjected to compressive strength test at a
ever, the crushed quartz has better quality, higher proportion of
loading rate of 2500 N/s. Three specimens were tested for flexural strength. And
thus, the average of six specimens was used to evaluate compressive strength of SiO2, greater hardness, and less impurity than natural sand. And
the mixtures. thus the strengths of RPC mix with natural sand were slightly
The shrinkage tests were performed following JC/T 603-2004 [19].The chloride- smaller than crushed quartz.
penetration resistance was tested based on ASTM C1202 [21].

2.4. Mix design 3.2. Water-to-cementitious materials ratio

Three types of binders, cement, silica fume and recycled powder, were used in Influences of water-to-cementitious materials ratio on proper-
the mix. Based on Dinger-Funk equation [14] of the maximum packing theory, the ties of RPC were studied. Based on Table 4, the mix proportion
mix proportion was designed. The grain size distribution curve was obtained firstly,
as shown in Fig. 2. And then the objective function, adjustment function and bound-
was developed, as shown in Table 6. The w/cm of 0.16, 0.18, 0.20,
ary conditions were determined. From Fig. 2, the maximum packing can be 0.22, 0.24 were studied. The flow, compressive strength and flexural
achieved at the proportions shown in Table 4. strength were tested, and the results are shown in Figs. 4–6.

Fig. 2. Particle size distribution of components and fitting curve.

 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254 249

Table 5
Mix propositions, flow and strength of the RPC with different fine aggregate.

W/CM Cementitious material Fine aggregate type FA/CMa Superplasticizer Compressive Flexural Flow (mm)
strength strength
(MPa) (MPa)
Cement Silica fume Recycled powder 7d 28d 7d 28d
RQ 0.20 1 0.15 0.1 Crushed quartz 1.0 0.025 11.1 13.7 58.9 79.7 176
RN 0.20 1 0.15 0.1 Natural sand 1.0 0.025 10.7 13.2 54.9 74.1 183
a
FA/CM fine aggregate-to-cementitious materials ratio.

Fig. 3. Particle size distribution of mix components and fitting curves for RPC mix with natural sand and crushed quartz.

Table 6
Mix proposition, flow and strengths of RPC.
16
W/CM Cementitious material FA/CMa Superplasticizer
15
28d
Flexural strength /MPa

Cement Silica fume Recycled powder

14
WB-1 0.16 1 0.15 0.1 1.0 0.025
WB-2 0.18 1 0.15 0.1 1.0 0.025 13 7d
WB-3 0.20 1 0.15 0.1 1.0 0.025
WB-4 0.22 1 0.15 0.1 1.0 0.025
12
WB-5 0.24 1 0.15 0.1 1.0 0.025 11
a
FA/CM fine aggregate-to-cementitious materials ratio. 10
9
8
0.16 0.18 0.2 0.22 0.24
240 W/CM
220
Fig. 5. Flexural strength versus water-to-cementitious materials ratio.
Flow/mm

200
180
160
120
140
Compressive strength /MPa

110 28d
120 100
0.16 0.18 0.2 0.22 0.24 7d
90
W/CM
80
Fig. 4. Flow versus water-to-cementitious materials ratio. 70
60
50
According to the test results, shown in Fig. 4, the flow increased 40
with the increase of w/cm. For the w/cm ratio over 0.20, the flow is 30
good. For the w/cm between 0.18 and 0.20, the flow is acceptable. 0.16 0.18 0.2 0.22 0.24
With the increase of w/cm, the flexural strength and compressive W/CM
strength decreased, as shown in Figs. 5 and 6.The w/cm was
suggested between 0.18 and 0.20. Fig. 6. Compressive strength versus water-to-cementitious materials ratio.
250 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254

3.3. Replacement of silica fume in RPC with recycled powder 180

3.3.1. Flow, flexural strength and compressive strength 175

In the RPC mixture, silica fume is replaced with recycled pow-

Flow /mm
der by 20%, 40%, 60%, 80% and 100% by weight. The mix design is 170
shown in Table 7. The flow, flexural strength and compressive
strength were tested, and results are summarized in Figs. 7–9. 165

(1) The flow of the mix decreased slightly as the increase of the 160
recycled powder amount, as shown in Fig. 7. The flow of the
mix with 20% replacement rate was larger than that without 155
recycled powder, the reason for which may be that the water 0 20 40 60 80 100
demand of recycled powder is less than that of silica fume. Recycled powder replacement percentage /%
As the amount of recycled powder increased, the flow
decreased. One reason is that the shape of recycled powder Fig. 7. Flow versus replacement rate of silica fume in RPC with recycled powder.

is irregular and its surface is coarse. Another reason may

be the particle size of silica fume is much smaller than that
of recycled powder (the average particle size of silica fume is 16
2.6 lm and that of recycled powder is 31.4 lm.) Because of 28d
15
small particle size, silica fume can fill among the cement 7d
14
particles so that free water can be released from the gaps 3d

Flexural strength /MPa

of cement particles, and the increase of free water due to 13
the addition of silica fume can improve the flow of mix. 12
The flow of all mix with recycled powder is good due to 11
smaller water demand of recycled powder. 10
(2) The compressive and flexural strengths tended to decrease 9
as the increase of the recycled powder, as shown in Figs. 8 8
and 9. Compared with the recycled powder, silica fume has
7
higher pozzolanic activity, and also smaller particle size,
which can improve the packing density [14]. As the replace- 6
ment rate of recycled powder increased to 20%, the compres- 5
sive strengths at 3d and 28d reduced by 27% and 17% 4
respectively. And as the replacement rate increased to 40% 0 20 40 60 80 100
from 0%, the compressive strengths at 3d and 28d reduced Recycled powder replacement percentage/%
by 35% and 25% respectively. As the silica fume was replaced
Fig. 8. Flexural strength versus replacement rate of silica fume in RPC with recycled
with recycled powder completely, the compressive and flex- powder.
ural strengths at 3d and 28d decreased by 75% and 31%
respectively.
(3) As silica fume is replaced with recycled powder partially or
completely, the flow, compressive and flexural strengths are 110
acceptable. 28d
100
7d
Compressive strength/MPa

90 3d
3.3.2. Shrinkage 80
The shrinkage of RS-0, RS-2, RS-5 (the replacement rate of recy-
70
cled powder: 0%, 40%, 100%) were tested according to JC/T 60-2004
60
[19]. The mixture design is shown in Table 7, and the results are
shown in Table 8 and Fig. 10. Each result is the average of three 50
specimens. 40
Before 28 days, the shrinkage developed significantly. And the 30
development was small after that. As the recycled powder was 20
introduced into the mixture, the shrinkage of all ages reduced 10
0 20 40 60 80 100
Recycled powder replacement percentage/%
Table 7
Mix proportion of RPC with silica fume replaced with recycled powder. Fig. 9. Compressive strength versus replacement rate of silica fume in RPC with
recycled powder.
Mix ID W/ Cementitious material FA/ Superplasticizer
CM CMa
Cement Silica Recycled powder
fume (replacement
percentage %)
Table 8
RS-0 0.20 1 0.25 0 (0) 1.0 0.025 Shrinkage of RPC with silica fume replaced by recycled powder.
RS-1 0.20 1 0.20 0.05 (20) 1.0 0.025 6
RS-2 0.20 1 0.15 0.10 (40) 1.0 0.025 Mix ID Shrinkage of different ages/10
RS-3 0.20 1 0.10 0.15 (60) 1.0 0.025 1d 3d 7d 14d 28d 45d 60d
RS-4 0.20 1 0.05 0.20 (80) 1.0 0.025
RS-0 281 556 724 872 936 944 952
RS-5 0.20 1 0 0.25 (100) 1.0 0.025
RS-2 240 362 480 564 606 618 629
a RS-5 98 282 390 476 563 608 636
FA/CM fine aggregate-to-cementitious materials ratio.
 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254 251

1000 180
900
800 175

Flow/mm
Shrinkage/10-6

700
600 170
500
400 RS-0 165
300
RS-2
200 160
RS-5 9 18 27
100
Recycled powder replacement percentage /%
0
0 10 20 30 40 50 60
Fig. 11. Flow versus replacement rate of cement in RPC by recycled powder.
Time/days

Fig. 10. Shrinkage development of RPC with silica fume replaced by recycled
powder.
15
28d
14
Table 9 7d

Flexural strength/Mpa
13
Chloride-penetration resistance of RPC with silica fume replaced by recycled powder. 3d
12
Mix ID Charge passed/coulombs Chloride ion penetrability 11
RS-0 60.10 Negligible 10
RS-2 125.67 Very low
9
RS-5 167.42 Very low
8
7
significantly, and the development rate decreased also. The reason 6
for that is the recycled powder has lower pozzolanic activity and
5
larger particle size than silica fume. 9 18 27
Recycled powder replacement percentage/%
3.3.3. Chloride-penetration resistance
The chloride-penetration resistance of RS-0, RS-2, RS-5 (the Fig. 12. Flexural strength versus replacement rate of cement in RPC by recycled
powder.
replacement rate of recycled powder: 0%, 40%, 100%) was tested
based on ASTM C1202 [21]. The mixture design is shown in Table 7,
and the results are shown in Table 9. Each result is the average of
three specimens.
(1) The flow of the mix decreased slightly as the increase of the
According to Table 9, as the amount of the recycled powder
recycled powder amount, as shown in Fig. 11. This is because
increased, the chloride-penetration resistance of RPC decreased.
the water demand of recycled powder is similar to that of
Compared with the mixture with no recycled powder replacement,
cement.
charge passed for the mixture with 40% recycled powder replace-
(2) The flexural strengths at 3d and 28d reduced by 27% and 18%
ment increased for 109%, and it increased for 179% for that of
respectively, as the replacement rate increased to 27% from
100% replacement, the chloride-penetration resistance of which
9%, as shown in Fig. 12. This is because the cement has
is still good.
higher pozzolanic activity than the recycled powder. The
Compared with the recycled powder, silica fume has higher
change of the compressive strength as the replacement rate
pozzolanic activity, and also smaller particle size, which can
increased from 9% to 27% was slight, as shown in Fig. 13. The
improve the packing density [14]. Also, the adsorption of chloride
grain size of the recycled powder was similar with that of
by recycled powder is weak, due to its irregular shape and coarse
the cement as shown in Fig. 2. However, a considerable part
surface. Moreover, the recycled powder was made with abandoned
of particles in the recycled powder had the grain size under
materials, which may contain chloride ion.
5 lm. So the replacement can increase the packing density
of the RPC. Also, the pozzolanic activity of silica fume pro-
3.4. Replacement of cement in RPC with recycled powder
moted the hydration reaction as the silica fume/cement ratio
increased with the increase of the recycled powder. Hence,
3.4.1. Flow, flexural strength and compressive strength
partial replacement of cement by the recycled powder had
In the RPC mixture, the cement is replaced with recycled pow-
no significant effect on the compressive strength of RPC.
der by 9%, 18% and 27% by weight. The mix design is shown in
(3) Considering the flow and strengths, the replacement rate of
Table 10. The flow, flexural strength and compressive strength
18% is suggested.
were tested, and results are shown in Figs. 11–13.

Table 10
Mix proportion for RPC with cement replaced by recycled powder.

Mix ID W/B Cementitious material FA/CMa Superplasticizer

Cement Silica fume Recycled powder (replacement percentage %)
RC-1 (RS-2) 0.20 1 0.15 0.10 (9) 1.0 0.025
RC-2 0.20 0.9 0.15 0.20 (18) 1.0 0.025
RC-3 0.20 0.8 0.15 0.30 (27) 1.0 0.025
a
FA/CM fine aggregate-to-cementitious materials ratio.
252 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254

85 700

600
Compressive strength/Mpa

75
28d
500

Shrinkage /10-6
65 7d
3d 400
55
300 RC-1
45 200 RC-2
35 100 RC-3

25 0
9 18 27 0 20 40 60
Recycled powder replacement percentage/% Time/days

Fig. 13. Compressive strength versus replacement rate of cement in RPC with Fig. 14. Shrinkage development of RPC with cement replaced by recycled powder of
recycled powder. different rates.

3.4.2. Shrinkage
Table 12
The shrinkage of RC-1(RS-2), RC-2, RC-3 (the replacement rate Chloride-penetration resistance of RPC with cement replaced by recycled powder of
of recycled powder: 9%, 18%, 27%) were tested according to JC/T different rates.
603-2004 [19]. The mixture design is shown in Table 10, and the
Mix ID Charge passed/coulombs Chloride ion penetrability
results are shown in Table 11 and Fig. 14. Each result is the average
of three specimens. RC-1(RS-2) 125.67 Very low
RC-2 163.67 Very low
As the recycled powder ratio increased, the shrinkage
RC-3 178.60 Very low
decreased. The shrinkage development rate decreased as the age
increased. One reason is that the chemical shrinkage and drying
shrinkage caused by the hydration reaction of cement decreased
as the amount of cement reduced. Another reason is the small par- (1) The flow of the mix increased as the GGBFS rate increased to
ticles of the recycled powder can fill the pore in the concrete. The 20%, and decreased after the GGBFS rate of 20%, as shown in
water evaporation was reduced and thus the drying shrinkage was Fig. 15. The particle size of slag is smaller than that of the
reduced. cement and recycled powder but larger than that of silica
fume. The filling effect of GGBFS can improve the flow of
3.4.3. Chloride-penetration resistance the mix to some extent.
The chloride-penetration resistance of RC-1(RS-2), RC-2, RC-3 (2) The compressive and flexural strengths increased as slag
(the replacement rate of recycled powder: 9%, 18%, 27%) was tested replacement increased from 0 to 10% and then decreased
based on ASTM C1202 [21]. The mixture design is shown in as the increase of the slag, as shown in Figs. 16 and 17. As
Table 10, and the results are shown in Table 12. Each result is the slag replacement rate increased from 0% to 10%, the
the average of three specimens. compressive and flexural strengths at 28 days increased by
According to Table 12, as the amount of the recycled powder 11% and 7%, respectively. As the slag replacement rate
increased, the chloride-penetration resistance of RPC decreased. increased to 30% from 0%, the compressive and flexural
The chloride-penetration resistance of RPC with recycled powder strengths at 28 days decreased by 8%. The particle size of
of 27% replacement is still good. slag is smaller than that of cement, which can improve the
The pozzolanic activity of recycled powder is relatively low, and packing density to some extent.
the production of C-H-S gel was reduced, which may decrease the (3) For RPC with silica fume and recycled powder, as the cement
packing density of RPC. Moreover, the recycled powder was made is partially replaced with slag, the flow, the compressive and
with abandoned materials, which may contain chloride ion. flexural strengths can be improved. The replacement rate of
10% was suggested.

3.5. RPC with recycled powder and GGBFS 4. Conclusion

In the RPC mixture with recycled powder, cement is replaced The use of the recycled powder from construction and demoli-
with GGBFS by 10%, 20% and 30%. The mix design is shown in tion wastes in RPC is proposed, and an initial investigation is per-
Table 13. The flow, flexural strength and compressive strength formed to study the possibility of developing environmentally-
were tested, and results are shown in Figs. 15–17. friendly and cost-saving RPC mixtures with high performance.
Mix proportions of the green RPC are studied through theory and
designed experiments. Influences of recycled powder on the
Table 11
strength, workability and durability of RPC are investigated
Shrinkage of PRC with cement replaced by recycled powder.
through designed experiments. Following conclusions can be made
6
Mix ID Shrinkage of different ages/10 from this study:
1d 3d 7d 14d 28d 45d 60d
RC-1(RS-2) 240 362 480 564 606 618 629 (1) For RPC with recycled powder, the flow, the compressive and
RC-2 80 213 361 487 560 616 639 flexural strengths by standard curing were not significantly
RC-3 57 135 241 318 409 471 513 influenced, as natural sand is used instead of crushed quartz.
 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254 253

Table 13
Mix proportion of RPC with cement replaced by GGBFS.

Mix ID W/CM Cementitious material FA/CMa Superplasticizer

Cement Silica fume Recycled powder GGBFS (replacement percentage %)
SC-0 (RS-2) 0.20 1 0.15 0.1 0 (0) 1.0 0.025
SC-1 0.20 0.9 0.15 0.1 0.1 (10) 1.0 0.025
SC-2 0.20 0.8 0.15 0.1 0.2 (20) 1.0 0.025
SC-3 0.20 0.7 0.15 0.1 0.3 (30) 1.0 0.025
a
FA/CM in aggregate-to-cementitious materials ratio.

190 the shrinkage reduced, and the chloride-penetration resis-

tance of RPC decreased. The chloride-penetration resistance
of the mixture with recycled powder replacement of 100%
Flow/mm

185
was still good.
(4) As the replacement rate of cement in RPC by recycled pow-
180 der increased, the flow of the mix decreased slightly, the
flexural strength decreased, the compressive strength chan-
175 ged slightly, the shrinkage reduced, and the chloride-
0 10 20 30 penetration resistance of RPC decreased. The chloride-
Slag replacement percentage/ % penetration resistance of the mixture with recycled powder
replacement of 27% was still good.
Fig. 15. Flow versus replacement rate of cement in RPC with GGBFS.
(5) As the replacement rate of cement in RPC by GGBFS
increased, the flow, compressive and flexural strengths was
14 first increased and then decreased. The replacement rate of
10% was suggested.
13
Flexural strength/MPa

12
28d
11 Acknowledgements
7d
10 3d
The research reported in this paper has been performed under
9 the National Science Foundation for Distinguished Young Scholars
8 of China 51208373, Shanghai Pujiang Program 12PJ1409000, and
Specialized Research Fund for the Doctoral Program of Higher Edu-
7
cation (SRFDP) 20120072120008.
6
0 10 20 30
Slag replacement percentage/% References

[1] P. Richard, M. Cheyrezy, Composition of reactive powder concretes, Cem.

Fig. 16. Flexural strength versus replacement rate of cement in RPC with GGBFS.
Concr. Res. 25 (7) (1995) 1501–1511.
[2] P. Richard, M. Cheyrezy, Reactive powder concretes with high ductility and
200–800 MPa compressive strength, ACI Spec. Publ. 144 (1994) 507–518.
110 [3] H. Yazıcı, M.Y. Yardımcı, S. Aydın, A.S ß . Karabulut, Mechanical properties of
28d reactive powder concrete containing mineral admixtures under different
100 curing regimes, Constr. Build. Mater. 23 (3) (2009) 1223–1231.
Compressive strength/MPa

7d [4] H. Kasami, M. Hosino, T. Arasima, H. Tateyasiki, Use of recycled concrete

90 powder in self-compacting concrete, in: Fifth CANMET/ACI International
3d
80 Conference on Recent Advances in Concrete Technology, Singapore, 2001, pp.
381–397.
70 [5] V.M. Malhotra, P.K. Mehta, Pozzolanic and Cementitious Materials, Gordon and
Breach Publishers, 1996.
60
[6] L. Turanli, F. Bektas, P.J.M. Monteiro, Use of ground clay brick as a pozzolanic
50 material to reduce the alkali–silica reaction, Cem. Concr. Res. 33 (10) (2003)
1539–1542.
40 [7] K. Kartini, M.N. Rohaidah, Z.A. Zuranini, Performance of ground clay bricks as
partial cement replacement in grade 30 concrete, World Acad. Sci. Eng.
30
Technol. 68 (2012) 362–365.
20 [8] S. Wild, A. Gailius, H. Hansen, L. Pederson, J. Szwabowski, Pozzolanic properties
0 10 20 30 of a variety of European clay bricks, Build. Res. Inf. 25 (3) (1997) 170–175.
[9] G. Baronio, L. Binda, Study of the pozzolanicity of some bricks and clays,
Slag replacement percentage/% Constr. Build. Mater. 11 (1) (1997) 41–46.
[10] M. Glavind, E.J. Pedersen, Packing calculations applied for concrete mix design:
Fig. 17. Compressive strength versus replacement rate of cement in RPC with proceedings of the utilizing ready mixed concrete and mortar, in: Proceedings
GGBFS. of the International Conference Held at the University of Dundee, Scotland, UK
on, 1999, pp. 121–130.
[11] A.K.H. Kwan, H.C. Wong, Packing density: a key concept for mix design of high
(2) With the increase of w/cm ratio, the flow increased, and the performance concrete, in: Proceedings of the Materials Science and
Technology in Engineering Conference, HKIE Materials Division, Hong Kong,
flexural strength and compressive strength decreased. The 2005, pp. 1–15.
w/cm ratio was suggested between 0.18 and 0.20. [12] F. Lange, H. Mörtel, V. Rudert, Dense packing of cement pastes and resulting
(3) As the replacement rate of silica fume in RPC by recycled consequences on mortar properties, Cem. Concr. Res. 27 (10) (1997) 1481–
1488.
powder increased, the flow of the mix decreased slightly, [13] F. de Larrard, T. Sedran, Optimization of ultra-high-performance concrete by
the compressive and flexural strengths tended to decrease, the use of a packing model, Cem. Concr. Res. 24 (6) (1994) 997–1009.
254 P. Zhu et al. / Construction and Building Materials 113 (2016) 246–254

[14] D. Dinger, J. Funk, Particle-size analysis routines available on Cerabull, Am. [19] JC/T 603-2004, Chinese Standards, Standard Test Method for Dying Shrinkage
Ceram. Soc. Bull. 68 (8) (1989) 1406–1408. of Mortar, Chinese Standards Association, China, 2004.
[15] GB175-2007, Chinese Standards, Common Portland Cement, Chinese Standards [20] ASTM Standard C192/C192M-13, Standard Practice for Making and Curing
Association, China, 2007. Concrete Test Specimens in the Laboratory, ASTM International, West
[16] GB/T18046-2008, Chinese Standards, Ground Granulated Blast Furnace Slag Conshohocken, PA, 2013.
Used for Cement and Concrete, Chinese Standards Association, China, 2008. [21] ASTM C1202-12, Standard Test Method for Electrical Indication of Concrete’s
[17] GB/T 18736-2002, Chinese Standards, Mineral Admixture of High Strength and Ability to Resist Chloride Ion Penetration, ASTM International, West Conshohocken,
High Performance Concrete, Chinese Standards Association, China, 2002. PA, 2012.
[18] GB/T 17671-1999, Chinese Standards, Method of Testing Cements-Determination [22] GB/T2419-2005, Chinese Standards, Test Method for Fluidity of Cement Mortar,
of Strength (ISO), Chinese Standards Association, China, 1999. Chinese Standards Association, China, 2005.