Beneficial Use of Recycled Materials in Concrete Mixtures

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Construction and Building Materials 29 (2012) 428–437

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials


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

Beneficial use of recycled materials in concrete mixtures


Patrick L. Maier a, Stephan A. Durham b,⇑
a
United States Bureau of Reclamation, P.O. Box 25007, Denver, CO 80225, USA
b
Department of Civil Engineering, University of Colorado Denver, Denver, CO 80217, USA

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

Article history: The need to incorporate recycled materials in building products is becoming more important than ever
Received 11 July 2011 before. The use of recycled materials in concrete mixtures creates landfill avoidance and decreases the
Received in revised form 19 September 2011 depletion of virgin raw materials. The basis for this research was to investigate the effects of using recy-
Accepted 2 October 2011
cled materials, in varying amounts, on the fresh and hardened concrete properties. The recycled materials
used in this study consisted of ground granulated blast furnace slag (GGBFS), recycled concrete aggregate
(RCA) and crushed waste glass. The GGBFS, or slag cement, was used as a replacement for the cement. The
Keywords:
RCA and waste glass were used to replace the coarse and fine aggregates, respectively. The concrete mix-
Recycled materials
Concrete
tures designed ranged from a 25% replacement to one 100% replacement with recycled materials. In addi-
Blast furnace slag (GGBFS) tion, a standard concrete mixture using cement and virgin aggregates was designed for comparison
Waste glass purposes. Fresh and hardened concrete properties were examined including slump, air content, unit
Recycled concrete aggregate (RCA) weight, compressive strength, rate of strength gain, freeze–thaw durability, permeability, and alkali-silica
Alkali–silica reactivity (ASR) reactivity (ASR) potential. The 100% recycled materials concrete had very low permeability and a com-
Strength pressive strength of 4200 psi (29.0 MPa) with 6.5% air content. Concrete mixtures composed of 50%
Durability and 75% recycled materials achieved strengths of nearly 7000 psi (48 MPa) and 6350 psi (43.8 MPa)
respectively. Beneficial and negative effects of using recycled materials in concrete mixtures were inves-
tigated, including the potential alkali-silica reactivity (ASR) of using waste glass as aggregate. The slag
cement, when used at replacement levels of 50%, was found to eliminate these concerns. The use of recy-
cled materials was beneficial with regards to strength and durability up to 50% when compared with a
normal concrete made from virgin materials.
Published by Elsevier Ltd.

1. Introduction The production of Portland cement (PC) requires significant


amounts of energy. Production of PC produces significant amounts
Concrete is the most commonly used building material in con- of carbon dioxide (CO2). Although recent advancements in cement
struction today. In 2009 an estimated 70 billion tons of cement production have reduced the amount of CO2 produced to below 3%
was produced in the United States alone [1]. Everything related of the United States total industrial CO2 emissions, cement produc-
to construction is open for scrutiny in today’s eco-conscience soci- tion in 2009 produced an estimated 29 Tg of CO2 (1 terra
ety. People today are more in tune and informed about the nega- gram = 1 million metric tons) [2,3]. Carbon dioxide is a green house
tive effects humankind leave behind for future generations. gas, and is believed to be a main contributor to global warming.
Green is the new buzz word, and every facet of today’s industry Most of the CO2 produced comes from the high temperature kilns
is attempting to reduce their carbon footprint. Builders today are used in PC production plants. Of all the raw materials used in con-
under constant pressure to become more ‘‘earth friendly’’ and are crete today, PC is the largest contributor to green house gases.
constantly looking for more ways to incorporate recycled materials The use of recycled materials to replace cement is common
into their products. The potential use of recycled materials in con- practice and has been for many years. It has been demonstrated
crete is a growing interest. Although the use of recycled materials that concrete’s strength, durability and workability can be in-
in concrete is not a new advancement, typical replacement values creased from the use of certain recycled materials. Supplementary
have commonly been on a small order as well as the combined use cementitious materials (SCM) have been used to replace cement in
of recycled materials into a single concrete mixture. concrete for thousands of years. Common cement replacements
used today are fly ash, silica fume and ground granulated blast fur-
nace slag (GGBFS). GGBFS, also referred to as slag cement, is a
⇑ Corresponding author. Address: Campus Box 113, P.O. Box 173364, Denver, CO byproduct of the iron manufacturing industry. Slag cement is a
80217, USA. Tel.: +1 303 352 3894; fax: +1 303 556 2368. hydraulic cementitious material that has pozzolanic characteris-
E-mail address: stephan.durham@ucdenver.edu (S.A. Durham). tics. The term pozzolan is used to describe any reactive aluminosil-

0950-0618/$ - see front matter Published by Elsevier Ltd.


doi:10.1016/j.conbuildmat.2011.10.024
P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437 429

icate material. Pozzolans react with by-products of the cement aggregates and 100% cement. The fresh concrete properties tested
hydration process in order to develop strength characteristics in included slump, unit weight and air content. The hardened con-
concrete. Pozzolans will not typically produce strength alone when crete properties examined were compressive strength, rate of
mixed with water and therefore require cement within the mix- strength gain, permeability, freeze–thaw resistance and alkali-sil-
ture. For this reason the common thought in the concrete industry ica reactivity (ASR). The RCA and waste glasses were fully tested
is that cement is needed, in at least a moderate amount, to produce prior to batching. Multiple gradations for both aggregates were
strength. Slag cement possesses cementitious properties much like completed as well as fineness modulus, specific gravity, absorption
cement, and when mixed with water will hydrate and produce capacities and unit weights. All testing conformed to ASTM testing
strength alone. standards and when deviated from, notation was made. All data re-
Aside from cement, concrete is composed of several other ingre- sults, details and conclusion of findings of this research have been
dients, mainly aggregates. Although important in varying degrees summarized for this article.
to a concrete mixture, aggregates by and large are a filler material.
In fact, it is the bond between the aggregate and cement that is the
weakest link in the concrete matrix. Typically, it is only in high 2. Background
strength concretes where aggregate strength becomes a contribut-
ing factor. Quality aggregate sources are becoming more difficult to A literature review was performed to investigate any past re-
find. Many aggregate sources used in the past have been depleted search completed regarding slag cement, RCA, and waste glass on
and concrete batch plants are forced to use lesser quality aggre- the effects each has on concretes properties. No documented re-
gates. To acquire aggregates from the earth, considerable energy search could be found on the combined effects that all three of
must be used to quarry and refine the rock before being suitable these components together have on concrete. This research
for use in concrete. Mining operations are always at the forefront involves testing concretes made with varying amounts of com-
of environmental debate not only from the destructive aspect, bined recycled materials. These replacements will be done to-
but from an aesthetic standpoint. For these reasons, aggregates gether on all three ingredients equally. The combined effects that
are of primary interest with regards to potential replacement with slag cement, RCA and waste glass will have on concrete is un-
recycled materials. Many forms of aggregate replacement have known and can only be speculated based on the observed and re-
been used in the past, from recycled automotive tires and waste ported effects that each alone have had on concrete. Since
metal to pure trash. One coarse aggregate replacement that is gain- aggregate volume is much higher than paste volume it could be as-
ing interest and use is recycled concrete aggregate (RCA). Recycled sumed that the aggregate qualities would have the most pro-
concrete comes from the demolition of buildings, sidewalks, nounced effect. However, the cementitious paste is the heart of
streets, etc. Increasing concern over the potential harmful effects the mixture and the slag cement effects may govern.
that crushed concrete can have on the environment have caused The workability of a concrete was shown to increase with
growing concerns on how to dispose of the materials (due to leach- increasing slag cement [6–10] and decrease with increasing RCA
ing of chemicals into the watershed), not to mention the increased [11–13] and waste glass [14,15]. The increase in workability from
costs associated with concrete disposal. The diminishing landfill slag cement was attributed to the glassy surface of the particles
space is a growing concern throughout the world. Reusing crushed and lower absorption rate (as compared to cement particles), thus
concrete is of considerable interest for these reasons. creating a slippery lubricating surface. The decrease in workability
Another less common aggregate replacement that is gaining from the RCA was mainly attributed to the very high absorption
more attention is the use of recycled glass as a fine aggregate capacity (AC) and also the presence of fines and rough angular
replacement. Recycling of glass containers is difficult and costly, shape. The decrease in workability from the waste glass was attrib-
especially if separating the glass into different colors is required. uted mainly with the ‘‘harshness’’ of the glass particles (sharp and
Removal of contaminants is difficult and much of the glass pro- angular, high fineness modulus). The AEA dosage of slag cement
duced today ends up in landfills. In 2007 approximately 13.6 mil- mixtures tended to increase with increasing content [9,10,16]
lion tons (12.3 metric tons) of waste glass were generated in the and was typically shown to increase this demand more as the
United States, and 76% of this glass was disposed of in landfills Blaine fineness of the slag cement increased. The AEA requirements
[4]. Whether glass containers are crushed and recycled or re-used, of RCA also tended to increase but not as significantly and was re-
it is estimated that for every ton of glass recycled, 1000 lb (454 kg) lated to fines content [17,18]. The air contents of waste glass mix-
of CO2 gas is saved from being emitted into the atmosphere [5]. tures did not appear to be effected [14,15].
This research investigates the effects that these recycled mate- The strength of concretes made with slag cement tended to in-
rials will have on the fresh and hardened properties of concrete. crease up to a certain percentage replacement and then decreased
Concrete mixtures containing recycled materials were designed thereafter [19,9]. Concretes made with slag cement also tend to ex-
and batched for this research. The amounts of recycled materials hibit slower strength gains (low early age strength) and higher la-
used in each concrete mixture were varied, and their fresh and ter age strengths, [19,9,8,20,21,8]. Slag cement hydration is
hardened properties compared to a control mixture composed of believed to be more C2S controlled rather than C3S, which would
natural (virgin) aggregates and PC. To develop these recycled con- cause these slower strength gains [10,22]. The use of RCA tends
crete mixtures, the natural coarse aggregates were replaced with to decrease a concrete mixtures strength with increasing content
RCA, the natural fine aggregates with waste glass and the cement [21,18,23]. Certain studies have found that source concrete
was replaced with slag cement. To fully investigate the effects strength may be related to new concrete strength (i.e. where the
these recycled materials have on the concrete, six mixtures with RCA originated from) [18]. The use of waste glass also tended to de-
varying amounts of recycled material replacements were devel- crease a concrete’s strength with increasing content [14,15]. This
oped, batched, and tested for structural and durability perfor- was mostly attributed to the strength of the glass, being more brit-
mance. A concrete mixture containing 100% recycled materials tle in nature than natural sand. There is also concern on the lack of
(RCA, waste glass and slag cement) was designed, batched and strength produced at the interfacial transition zone (ITZ) with
tested. Three additional mixtures with recycled material contents waste glass, due to the smooth, relatively impermeable surface.
of 25%, 50% and 75% were batched and tested. One mixture was The permeability of concretes made with slag cement signifi-
batched that contained 100% recycled aggregates and 100% cement cantly decreased with increasing slag cement content [10,24,9]. It
as well as a control mixture composed of 100% natural (virgin) is believed that this decrease in permeability is caused by the
430 P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437

pozzolanic reaction between the calcium hydroxide (CH) creating a gates and 100% Type-I-II PC as the cementitious material content.
denser micro structure. However, there are studies that argue Mixture #2 designated as 100-RA-C was designed to investigate
whether or not slag cement and other SCM’s chemically alter the the effects of recycled aggregates on a concrete made with 100%
composition of the cementitious paste, and therefore render the PC as the cementitious material. This mixture contained 100%
most common test methods for permeability useless (i.e. ASTM C RCA and 100% waste glass as the coarse and fine aggregates respec-
1202) [24]. The permeability of concretes made with RCA as coarse tively. Mixture #3 designated as 100-RA-BF was designed based on
aggregate tended to increase with increasing content [11,13, a 100% replacement of content with recycled materials. This mix-
17,18,23]. This permeability increase was shown to be somewhat ture had 100% slag cement as cementitious and 100% RCA and
connected with source concrete permeability [18]. The use of waste glass as the coarse and fine aggregates respectively. Mixture
waste glass was shown to increase the permeability with increas- #4 designated as 25-RA-BF was designed based on a 25% replace-
ing content [15]. The freeze thaw durability of concretes made ment of content with recycled material. The cementitious content
with slag cement may decrease as replacement levels increase was divided by weight into 25% slag cement and 75% PC. The
[25], however most studies have found little to no effect with coarse aggregate was spilt into 25% and 75% RCA and waste glass
replacement levels up to 50% [19,10]. The freeze thaw durability for virgin rock and sand by weight respectively. Mixture #5 desig-
of concrete was shown to change little with the use of RCA nated as 50-RA-BF was designed based on a 50% replacement of
[17,11,23]. As glass content increased the freeze–thaw durability content with recycled material. The cementitious content was split
was shown to generally decrease [14], when used as both coarse up by weight into 50% slag cement and PC. The coarse aggregate
and fine aggregate replacements. The alkali-silica reactivity of con- was spilt into 50% RCA and virgin rock by weight respectively
cretes made with glass aggregate has been shown to increase as and the fine aggregate was split into 48% waste glass and 52% vir-
content increases [26,15]. Literature indicates that the color of gin sand by weight respectively. The small deviation from the 50%
the glass significantly impacts whether or not deleterious ASR for the fine aggregate is due to the mixture being designed based
expansions would take place [26]. The size of the glass particles upon the absolute volume method and the slight difference in
was shown to play a major role [26], and pessimum sizes for each the specific gravity of the two aggregates. Mixture #6 designated
color were found to exist (i.e. the size at which maximum expan- as 75-RA-BF was designed based on a 75% replacement of content
sions occur). The use of slag cement decreases the ASR expansions with recycled material. The cementitious content was split up by
in concretes made with reactive aggregates [22,27,28]. weight into 75% slag cement and 25% PC which totaled. The coarse
aggregate was then spilt into 75% RCA and 25% virgin rock by
weight and the fine aggregate was split into 73.7 and 26.3% waste
3. Concrete mixture designs and methodology
glass and virgin sand by weight respectively.

3.1. Research concrete mixtures

A total of six mixtures were designed for this study to investigate 3.2. Alkali-silica reactivity (ASR) concrete mixtures
the effects that recycled materials have on concrete and to ulti-
mately design and test a concrete composed of entirely recycled Due to the concerns related to the use of mixed colored waste
materials. As a general guide for selecting quantities and other glass as aggregate, several ASR tests were performed. Based on past
properties, these mixtures were designed based on a Colorado research found during the literature review, the use of waste glass
Department of Transportation (CDOT) Class-D structural concrete in concrete is almost certain to cause ASR expansion if mitigating
mixture. The lowest acceptable cementitious material content techniques are not employed. The extent of this expansion how-
and w/c of 615 lb/ft3 (365 kg/m3) and 0.40 respectively were used ever depends on the aggregate size and gradation, as well as the
for all mixtures. All mixtures incorporated an air entraining admix- color. These mixtures were designed and batched separately from
ture (AEA) to achieve the designed 6.5% air content. One brand of the previous six design mixtures according to ASTM C 1567 proce-
AEA was used for all mixtures and the dosage was held constant dures [30]. All mixtures contained 100% waste glass as aggregate.
at 1.5 fl oz/cwt (98 ml/kg). Although a target slump was not speci- One mixture contained 100% PC as cementitious materials. The
fied for these mixtures, because of the low w/cm a high range water next mixture consisted of 50% slag cement and PC as the cementi-
reducing admixture (HRWRA) was used on all mixtures to achieve tious materials and the third mixture was composed of 100% slag
workability. The dosage and type of HRWRA used varied depending cement as cementitious materials. These mixtures were designed
on mixture. A mixture design matrix is shown in Table 1. to investigate whether the slag cement would mitigate the ASR
The cementitious material content for these mixtures was expansion caused by the glass aggregate and if so, help capture
divided between slag cement and PC, depending on the mixture the replacement amount needed. The waste glass was fully pre-
design. Mixture #1 designated as CC (Control) was the control mix- pared based on ASTM C 1567 procedures. Table 2 shows the three
ture for this research. This mixture contained 100% virgin aggre- mixture proportions investigated.

Table 1
Concrete mixture design matrix.

Mixture # & identification W/CM Watera Cementa Slag cementa Virgin rocka Virgin sanda RCAa Waste glassa AEAb HRWRAc
1 CC (Control) 0.4 246 (146) 615 (365) 0 1646 (977) 1324 (785) 0 0 1.5 (98) 4.6 (0.3)
2 100-RA-C 0.4 246 (146) 615 (365) 0 0 0 1646 (977) 1139 (676) 1.5 (98) 56.4 (3.7)
3 100-RA-BF 0.4 246 (146) 0 615 (365) 0 0 1646 (977) 1103 (654) 1.5 (98) 56.4 (3.7)
4 25-RA-BF 0.4 246 (146) 461 (274) 154 (91) 1200 (712) 1001 (594) 400 (237) 313 (186) 1.5 (98) 26.7 (1.7)
5 50-RA-BF 0.4 246 (146) 308 (183) 308 (183) 800 (475) 649 (385) 800 (475) 608 (361) 1.5 (98) 4.6 (0.3)
6 75-RA-BF 0.4 246 (146) 154 (91) 461 (274) 400 (237) 316 (187) 1200 (712) 887 (526) 1.5 (98) 26.7 (1.7)
a
b/yd3 (kg/m3).
b
fl oz/cwt (ml/kg).
c
fl oz/cwt (L/kg).
P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437 431

Table 2 Table 4
ASR mixture design matrix. Chemical and physical analysis of ASTM type-I-II cement.

Material Mixture identification Chemical and physical properties Test results ASTM C 150 specifications
1 2 3 SiO2 (%) 19.6 –
Al2O3 (%) 4.7 6.0 max
Cement (Type I-II) (g) 440 220 0
Fe2O3 (%) 3.2 6.0 max
Slag cement (GGBFS) (Grade 120) (g) 0 220 440
CaO (%) 63.4 –
Waste glass (retained on No. 8) (g) 99 99 99
MgO (%) 1.5 6.0 max
Waste glass (retained on No. 16) (g) 247.5 247.5 247.5
SO3 (%) 3.4 3.0 max
Waste glass (retained on No. 30) (g) 247.5 247.5 247.5
CO2 (%) 1.4 –
Waste glass (retained on No. 50) (g) 247.5 247.5 247.5
Limestone (%) 3.7 5.0 max
Waste glass (retained on No. 100) (g) 148.5 148.5 148.5
CaCO3 in limestone (%) 84.0 70 min
Water (g) 207 207 207
C3S (%) 59.0 –
W/CM 0.47 0.47 0.47
C2S (%) 11.0 –
C3A (%) 7.0 8 max
C4AF (%) 10.0 –
C3S + 4.75 C3A (%) 92.0 100 max
Table 3 Loss of ignition (%) 2.6 3.0 max
Chemical and physical analysis of ASTM grade 120 slag cement. Blaine fineness (cm2/g) 414 260–430
Air content of PC mortar (%) 6.3 12 max
Chemical and physical properties Test results ASTM C 989 specifications
Specific gravity 3.15 –
Sulfur trioxide (SO3) (%) 2.192 4.0 max
Sulfur sulfide (S) (%) 0.560 2.5 max
Blaine fineness (cm2/g) 5240 –
Percent retained on 325 mesh (%) 0.2 20 max Table 5
Air content of slag mortar (%) 6.3 12 max Fine aggregate properties of waste glass and virgin sand.
Specific gravity 2.93 –
Aggregate property ASTM procedure Waste glass Virgin sand
Slag activity index
7-Day individual (%) 134 90 min Absorption capacity (%) ASTM C 128 0.18 0.70
28-Day individual (%) 138 110 min Specific gravity ASTM C 128 2.47 2.63
Fineness modulus ASTM C 136 4.08 2.74
Compressive strength Gradation analysis ASTM C 136 – –
7-Day reference cement (lb/in2) 4890 –
28-Day reference cement (lb/in2) 6260 5000 min
7-Day slag + Ref. cement (lb/in2) 6560 –
28-Day Slag + Ref. cement (lb/in2) 8660 –
Table 6
Coarse aggregate properties of RCA and virgin rock.

Aggregate property ASTM procedure RCA Virgin rock


4. Materials
Absorption capacity (%) ASTM C 127 4.58 0.80
4.1. Cementitious Specific gravity ASTM C 127 2.45 2.61
Dry rodded unit weight (lb/ft3) ASTM C 29 85.02 103
Because this research involved developing a concrete mixture that had a 100% NMAS (inch) ASTM C 136 3/4 3/4
replacement of cement with slag cement, a Grade 120 slag was chosen because of Gradation analysis ASTM C 136 – –
its higher reactivity compared to Grade 100 or 80. The slag cement was tested in
accordance with ASTM C 989 and the results of this testing are shown in Table 3.
The cement used for this research was a Type-I-II PC. The PC was tested in accor-
50 sieves [33]. The fineness modulus (FM) for the waste glass is much higher than
dance to ASTM C 150 and the results are shown in Table 4.
the virgin sand. For workability reasons, the FM for a fine aggregate should typically
be between 2.3 and 3.1 when used in concrete. Typically the higher the FM value
4.2. Virgin aggregates the coarser the fine aggregate particle size. Based on the test results, the waste glass
is a coarser aggregate than the virgin sand.
Both the virgin coarse and fine aggregates were obtained by regional sources in The recycled concrete aggregate (RCA) was obtained from a concrete recycling
Colorado. Material properties were determined and gradations of these aggregates plant located in Denver, Colorado. The aggregate was taken from a large pile desig-
were performed. The coarse aggregate was tested in accordance to ASTM C 33 No. nated as 3/4 inch (19 mm) nominal. The actual history behind the aggregate taken
57 and 67 Coarse Aggregate and meets these requirements [31]. The sand was for this research is unknown (i.e. the RCA was taken from a general stockpile). The
tested in accordance to ASTM C 33 Fine Aggregate, and meets these requirements. RCA is made from demolished roads, buildings, and other various concrete struc-
These aggregates will be referred to as ‘‘virgin sand’’ and ‘‘virgin rock’’ for the tures that have been removed from service (mostly from roads). Testing was per-
remainder of this article. formed on the RCA to determine the physical and chemical properties. All
samples used for testing properties followed ASTM C 702, Standard Practice for
4.3. Recycled aggregates Reducing Samples of Aggregate to Testing Size. Table 6 shows the properties of
the RCA as well as the virgin rock coarse aggregate for comparison.
The waste glass was obtained from a local glass bottling recycling plant located The RCA has a lower SG which correlates with the lower unit weight as well. The
in Denver, Colorado. The waste glass was used as a replacement for the virgin sand. lower unit weight and SG of the RCA may be a result of entrapped and entrained air
The waste glass was produced mainly from beer bottles and consisted of various within the mortar coatings and mortar chunks. The AC of the RCA is considerably
colors (clear, amber and green). The glass was used ‘‘as received’’ from the recycling higher than the virgin rock aggregate and is caused by the mortar fragments and
plant and no washing took place. Only larger metallic objects (i.e. batteries, bottle mortar coated aggregate which absorb water. These results are not uncommon as
caps, etc.) were hand removed from the waste glass prior to use in the concrete most reports on RCA indicate these same properties. Three separate gradations
mixtures. Testing was performed on the waste glass to determine the physical were performed on the recycled concrete. The gradations for all three specimens
and chemical properties. ASTM C 702, Standard Practice for Reducing Samples of did not meet the requirements of ASTM C 136 due to excess materials retained
Aggregate to Testing Size was followed for obtaining samples for all testing [32]. on the number 4 (all three samples) and 8 (one sample only) sieves.
Table 5 shows the properties of the waste glass as well as the virgin sand fine aggre-
gate for comparison.
The specific gravity (SG) of both fine aggregates is similar. Although both 4.4. Chemical admixtures
aggregates have low absorption capacities (AC), the waste glass is much lower. In
actuality, the waste glass itself has relatively zero AC; however the presence of la- Chemical admixtures were used in all mixtures batched for this research to im-
bels and other foreign materials results in a small AC. Three separate gradations prove workability and to achieve a target air content of 6.5%. One air entraining
were performed on the waste glass and all three specimens did not meet the admixture (AEA) was used for all mixtures. Two different high range water reducing
requirements of ASTM C 136 due to excess materials retained on the 16, 30, and admixtures (HRWRA) were used. The dosage rate for the AEA was constant for all
432 P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437

mixtures at 1.5 fl oz/cwt (1.0 L/kg). The dosage rate of HRWRA used varied consid- Table 8
erably throughout this research due to the effects that recycled materials had on the Fresh Concrete Properties.
workability (see Table 1).
Mixture identification Slump Air Measured Measured
& number content unit unit
4.5. Concrete batching and testing methods (inch) (cm)
(%) weight (lb/ weight
ft3) (kg/m3)
All concrete mixtures batched in this study followed the guidelines stipulated
by ASTM C 192 Standard Practice for Making and Curing Concrete Test Specimens 1 CC (control) 1.75 4.45 10.0 134.8 2160
in the Laboratory. The batching procedure was similar for all mixtures except for 2 100-RA-C 0.50 1.27 5.2 139.2 2230
the addition of admixtures, which varied considerably. The AEA was always added 3 100-RA-BF 0.50 1.27 4.6 140.8 2255
to the water prior to batching. Two batching procedures were followed in regards to 4 25-RA-BF 8.25 21.0 6.4 138.6 2220
HRWRA dosage procedure during this research. Procedure #1 involved adding the 5 50-RA-BF 0.50 1.27 7.0 140.8 2255
HRWRA after all materials were in the mixer. Procedure #2 involved dividing the 6 75-RA-BF 3.00 7.62 7.5 134.1 2148
water into two equal buckets. The first half portion of the mixture water contained
the AEA and was allowed to mix with the aggregates for several minutes alone. The
second half portion of the mixture water contained the HRWRA and was dispersed Mixtures #2 and #3 had the lowest slumps of only 0.5 inches
after the cementitious contents were placed into the mixture. For all batching, the
(12.7 mm). Both of these mixtures had 100% recycled aggregates
coarse aggregate and fine aggregate were placed into the mixer with half of the
water (regardless of procedure this water contained the AEA) and allowed to mix and differed only in the type of cementitious material. Both of
for several minutes. Next, approximately 75% of the cementitious was added and these mixtures contained up to 740 ml. (56.4 fl oz/cwt) of HR-2.
allowed to mix for a short time. Finally, the remaining water and cementitious The first batching used procedure #1 and HR-1 for HRWRA addi-
was added and if procedure #1 was followed, the HRWRA, and the mixture was al- tion and negligible slump was achieved (0 inch). Mixture #5 (50%
lowed to mix for several more minutes. If procedure #2 was followed the remaining
half of water was added with the HRWRA incorporated.
recycled mixture) had a slump of 0.5 inch (12.7 mm). Procedure
A total of twenty-one 4-in  8-in (100-mm  200-mm) cylinders were cast for #1 and HR-1 with a dosage of 60 ml (4.6 fl oz/cwt) was used. Mix-
each mixture as well as two freeze–thaw prisms measuring 3-in  4-in  16-in (75- ture #1 (control mixture) achieved 1.75 inch (44.5 mm) of slump
mm  100-mm  400-mm). All specimens were placed in the humidity controlled with the same HR-1 dosage as Mixture #5. Procedure #1 was fol-
room overnight after batching. The humidity and temperature level of this room
lowed for adding the HRWRA. The F.M. of the virgin sand was
was held constant at 40% and 70 °F (21 °C) respectively. On the following day, all
specimens were removed from the molds and submerged in lime saturated water 2.74 and combined with the rounded stone coarse aggregate and
until time of testing. The temperature of the water bath was held constant at high air content, the higher slump was not surprising. Mixture
73 ± 3 °F (23 ± 2 °C). Fresh concrete properties were examined immediately after #6 (75% recycled mixture) achieved a slump of 3.0 inch (76 mm).
batching while hardened concrete properties were tested over the course of 90- HR-2 with a dosage of 350 ml (26.7 fl oz/cwt) was used. Procedure
days. Table 7 summarizes the testing performed, method and time frames.
#2 was followed for adding the HRWRA. Mixture #4 (25% recycled
mixture) had the highest slump value of 8.25 inch (210 mm). Pro-
4.6. ASR concrete batching and testing methods
cedure #2 and HR-2 with a dosage of 350 ml (26.7 fl oz/cwt) was
The batching procedure for ASR potential followed ASTM C 205 guidelines. The used. Fig. 1 shows the effects of the different HRWRA and proce-
specimens produced were 1 inch (25 mm) rectangular prisms measuring 10 inch dure (HR-1 vs. HR-2) on the slumps of the concrete mixtures bat-
(255 mm) in length. Because mixture #3 was composed of 100% slag cement as ched. What can be deduced from these results is the significant
cementitious, it was decided to allow these specimens to cure for an additional
effect recycled aggregates have on the workability and water de-
24 h prior to testing. Therefore, mixture #3 cured for a total of 48 h (as opposed
to the ASTM recommended 24 h). After initial curing was complete, the specimens
mand of a concrete mixture. As recycled aggregate content in-
were then placed in a water bath of 80.0 °C (176 °F) for 24 h. A new zero reading creased, the slump decreased, even when six times the
was made after 24 h (total of 48 h) and the specimens were then transported to a recommended dosage was used. The RCA had an absorption capac-
NaOH solution for the remainder of testing. This NaOH bath was held at a constant ity of 4.58% and combined with a F.M. of 4.08 for the waste glass
80 °C (176 °F) temperature.
and a w/c of 0.40, these results are not surprising. In addition, it
was notable that the HRWRA effects were not attributed to the
5. Experimental results type of cementitious used because mixture #1 and #2 had similar
slumps, regardless of procedure used. Another interesting point to
5.1. Fresh concrete properties make was the difference it makes to have even a small amount
(25%) of virgin aggregates within the mixture (i.e. mixture #6,
The results of these fresh concrete properties are included in Ta- 75-RA-BF).
ble 8. As previously mentioned, there was no specific target slump The use of AEA was incorporated into all mixtures. The air con-
required for these mixtures. However, in general a slump of 3 tents for mixtures #2 through #6 were somewhat close to each
inches (76.2 mm) was believed to be a reasonable value to achieve other and the target air content. However, Mixture #1 had an air
consolidation on the test specimens. Two types of HRWRA were content average of 10%. It was believed the dosage may have been
used during this research. The first type used was designated as inadvertently doubled during batching, which would explain the
HR-1, and the second as HR-2. It should be noted that when relat- high air content. This was caused by accidently thinking that the
ing amount of HRWRA and AEA, all mixtures batched were approx- AEA was not included in the batch water, when in fact it already
imately 1.95 ft3 (0.055 m3). was, therefore twice the dosage was incorporated. Mixtures #3
and #2 did not achieve the target air content having the lowest
percentages of 4.6% and 5.2% respectively. This may have been
Table 7
caused by additional mixing time required between HRWRA addi-
Fresh and hardened concrete properties tested.
tions to unsuccessfully achieve workability. This added mixing
Fresh concrete tests Standard followed Time of testing time may have essentially deflated the concrete mix as this is a
Slump ASTM C 143 When batched common result of over mixing. The air contents do not appear to
Unit weight ASTM C 138 When batched be related to recycled material content. Mixture #1 which was
Air content ASTM C 231 When batched
the control mixture had the highest air content followed by Mix-
Hardened concrete tests Standard followed Time of testing
Compressive strength ASTM C 39 1, 7, 28, 56, 90 days ture #6 which was composed of 75% recycled materials. This was
RCPT ASTM C 1202 28, 56, 90 days not consistent with mixtures #2 and #3 which had the lowest air
Freeze–thaw durability ASTM C 666 Begin at 28 days contents and highest recycled materials contents. However, if the
Alkali-silica reactivity (ASR) ASTM C 1567 Begin at 2 days
excessive mixing times did not occur for these two mixtures, the
P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437 433

Fig. 1. Mixture slump vs. HRWRA type and procedure.

results may show that air content does increase with increasing data. Since all mixtures were designed based on 6.5% air content
recycled materials content if the control mixture is ignored assum- this value was used as a datum. Concrete strengths were then ad-
ing an incorrect dosage of AEA. justed based on actual measured air content. The equation below
Air content plays a major role in measured unit weights. In the shows how the data was normalized. The average compressive
design process the target air content of 6.5% by volume was spec- strengths and average ‘‘normalized’’ compressive strengths are
ified and accounted for when the theoretical unit weights are cal- tabulated in Tables 9 and 10. The average ‘‘normalized’’ compres-
culated. Four out of the six mixtures followed a trend of increasing sive strength development is illustrated in Fig. 2.
or decreasing unit weight based on air content. To help account for
these differences, air-adjusted theoretical was calculated and the fc0 xð1  0:05 6:5%Þ=ð1  0:05 Air Measured%Þ ¼ fc;normalized
0

percentage difference shown in parenthesis. Mixture #1 had a sig-


nificantly lower unit weight than the theoretical but also had a When comparing the results presented in Tables 9 and 10, it be-
much larger air content of 10%, which was 3.5% above target. The comes clearly evident that air content played a large role in the
difference in unit weight was approximately 5.2% (1.0% from ad- compressive strengths for these concretes. The Mixture #1 (CC
justed). Mixture #2 had a higher unit weight which corresponds control) specimens did not achieve nearly the strengths that would
to the decreased air content of 5.2%, which is 1.3% below target. be expected from a concrete mixture such as this one. The neces-
The difference in unit weights is approximately 3.0% (1.6% from ad- sity to normalize the data based on air content should be obvious,
justed). Mixture #3 had a significantly higher unit weight than the- otherwise any comparisons made between mixtures would be
oretical and this corresponds to the lower air content of 4.6%. The biased. Therefore, any further reference to concrete mixture
difference in unit weights was approximately 5.3% (3.0% from ad- strengths will be based on the ‘‘normalized’’ data.
justed). Mixture #4 had a slightly lower unit weight but also had The 50% recycled mixture (Mixture #5) outperformed the con-
a slightly lower air content as well. This does not correlate with trol mixture by nearly 1000 psi at 28-days and 500 psi by 90-days.
expectations but the difference in unit weights is small, only What is interesting is that 50% recycled material content appears
0.85% (1.0% from adjusted). Mixture #5 had a higher unit weight to be the pessimum replacement amount. Another interesting
than theoretical and had a higher air content as well. This does observation was the dramatic increase in strength for this mixture
not correlate with expectations and the difference in unit weights between 1 and 7 days of age. The 25% and 75% recycled mixtures
was 2.3% (2.8% from adjusted). Mixture #6 had a slightly lower unit (Mixture #4 and #6) achieved very similar strengths and strength
weight which correlated with the increased air content of 7.5%; gains after the first day. This was interesting because they are com-
however, the difference in unit weights was small, only 1.2% (0% pletely different mixtures with respect to recycled material con-
from adjusted). tents. In addition, both of these mixtures achieved an identical
spike of strength gain between 56 and 90-days of age.
The ultimate strength of 6300 psi (43.4 MPa) for a concrete
5.2. Hardened concrete results made from 75% recycled materials was quite impressive. Mixture
#3 gained the least amount of strength but at the same time was
Hardened concrete testing was performed on all mixtures at the most impressive. This strength came entirely from the slag ce-
various ages after batching. These tests consisted of compressive ment and shows that it was possible to achieve strengths without
strength at 1, 7, 28, 56 and 90-days of age, permeability at 28, 56 the use of PC. Although the gain was much slower, an ultimate
and 90-days of age, freeze–thaw durability beginning at 28-days strength of 4200 psi (29.0 MPa) is considered a normal strength
of age and ASR expansion tests. concrete. If the strength gain was more rapid this mixture may
actually be a viable alternative. However, in applications were ra-
5.2.1. Compressive strength pid strength gain is not required (i.e. mass placements) this mix-
The strength of concrete and air content are inversely propor- ture may perform well. Although 120-day strengths were not
tional. The compressive strength of concrete decreases 5.0% for tested for these mixtures the results would probably indicate a
every 1.0% increase in air content [29]. In order to accurately eval- continued strength gain for this mixture, as well as the 75% recy-
uate the results of this research it was necessary to normalize the cled mixture (Mixture #6). The 100% recycled aggregate mixture
434 P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437

Table 9
Average compressive strengths.

Mixture identification & number Compressive strength, lb/in2 (MPa)


1-Day 7-Day 28-Day 56-Day 90-Day
1 Control 1974 (13.6) 3657 (25.2) 4332 (29.9) 4567 (31.5) 4824 (33.2)
2 100-RA-C 2410 (16.6) 3649 (25.2) 4375 (30.2) 4958 (34.2) 5540 (38.2)
3 100-RA-BF 773 (5.3) 2659 (18.3) 3836 (26.4) 4371 (30.1) 4820 (33.2)
4 25-RA-BF 2350 (16.2) 4248 (29.3) 5604 (38.6) 5966 (41.1) 6730 (46.4)
5 50-RA-BF 1509 (10.4) 5061 (34.9) 6674 (46.0) 6677 (46.0) 6739 (46.5)
6 75-RA-BF 879 (6.1) 4219 (29.1) 5148 (35.5) 5239 (36.1) 5870 (40.5)

Table 10
Average ‘‘normalized’’ compressive strengths.

Mixture identification & number Compressive strength, lb/in2 (MPa)


1-Day 7-Day 28-Day 56-Day 90-Day
1 Control 2665 (18.4) 4937 (34.0) 5848 (40.3) 6166 (42.5) 6513 (44.9)
2 100-RA-C 2113 (14.6) 3205 (22.1) 3851 (26.6) 4345 (30.0) 4869 (33.6)
3 100-RA-BF 677 (46.7) 2331 (16.1) 3363 (23.2) 3832 (26.4) 4226 (29.1)
4 25-RA-BF 2332 (16.1) 4216 (29.1) 5563 (38.4) 5923 (40.8) 6681 (46.1)
5 50-RA-BF 1567 (10.8) 5256 (36.2) 6931 (47.8) 6934 (47.8) 6998 (48.2)
6 75-RA-BF 949 (6.5) 4557 (31.4) 5560 (38.3) 5659 (39.0) 6339 (43.7)

Fig. 2. Average ‘‘normalized’’ compressive strength development.

(Mixture #2) did not perform as well as expected. This mixture was SCM’s, due to the chemically altered paste [24], it has been shown
the second lowest strength producer second only to the 100% recy- to be a reliable test method and results correlate well with the
cled mixture (Mixture #3). This indicates that there may be some- standard test methods of measuring permeability. Table 11 shows
thing happening between the PC and the aggregate which would the results of RCPT tests performed at 28, 56 and 90-days of age.
cause this decrease in strength. If something detrimental was The decrease in permeability for Mixtures #4, #5 and #6 can be
occurring between the PC and recycled aggregates, this effect attributed to the pozzolanic reaction that takes place between the
was obviously not occurring with a 25% slag cement replacement slag cement and the calcium hydroxide (CH) released during the
(i.e. Mixture #4). It should be noted that Mixture #2 was batched hydration process. This drop in permeability occurs when any poz-
twice and the results of both batched specimens were very similar. zolanic material is used (i.e. fly ash, silica fume, slag cement). How-
The results presented herein are the average of these two batches. ever, the 100% slag cement mixture (Mixture #3) did not have any
cement hydration products to allow a pozzolanic reaction to occur.
5.2.2. Permeability This shows that a very impermeable concrete is possible without
The permeability of concrete is a very important property. Sus- the use of PC. Slag cement not only hydrates on its own but pro-
ceptibility to chemical attack and corrosion of reinforcement are duces a tightly dense micro structure.
both directly related to a concrete’s permeability. The Rapid Chlo-
ride Ion Penetrability Test (RCPT) following procedures set forth by 5.2.3. Freeze–thaw durability
ASTM C 1202 was the test method used for this research program The freeze thaw durability of a concrete is directly related to
[34]. Although some researchers have argued this test method does concrete’s strength, permeability, and air content. Concrete is por-
not accurately measure the permeability of concrete’s containing ous by nature and will absorb water. How much water it absorbs
P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437 435

Table 11 and thawing exist. This was an arbitrary opinion though as a DF


TM C 1202 RCPT results (classification shown in parenthesis). above 60 is no guarantee that a concrete will behave well in actual
Mixture identification & 28-Day 56-Day 90-Day freezing and thawing conditions [29]. Regardless, this method is
number (Coulombs) (Coulombs) (Coulombs) proven and is a good means of comparing similar concretes.
1 CC (Control) 2106 (Mod.) 1972 (Low) 1969 (Low) According to ACI 318-08, the required air content for these mix-
2 100-RA-C 2907 (Mod.) 2212 (Mod.) 1773 (Low) tures with NMAS of 3/4 inch (19 mm) should be 6% and 5% for se-
3 100-RA-BF 578 (V. Low) 406 (V. Low) 321 (V. Low) vere and moderate exposure climates respectively. The conditions
4 25-RA-BF 1602 (Low) 1326 (Low) 1327 (Low)
5 50-RA-BF 1041 (Low) 1003 (Low) 874 (V. Low)
within the freeze–thaw chamber are no doubt extreme. Secondly,
6 75-RA-BF 516 (V. Low) 424 (V. Low) 403 (V. Low) ACI-318-08 recommends that a concrete strength be 4500 psi
(31 MPa) for extreme exposure climates. Only three out of the six
mixtures achieved these strengths by the time testing began. Most
mixtures showed minor scaling on the surface of the specimens,
without any significant mass loss. However, mixture #2 (100-RA-
will depend on the permeability as well as the other factors. When
C) exhibited major fracturing along the length of the specimen
temperatures within the concrete drops to levels that allow the en-
after approximately 60 cycles. Mixture #3 (100-RA-BF) exhibited
trapped water to freeze, expansive tensile forces are created due to
the most rapid deterioration of all specimens with regards to mass
the ice crystal formation. Much like water penetration, ice crystal
loss. After only 28 cycles the dynamic modulus began to quickly
formation follows a similar principle of least resistance. Air voids
drop combined with an alarming mass loss.
within the concrete will allow ice crystal formation and thus de-
A final interesting observation can be made when comparing
crease these expansive forces. Although air content decreases a
the 25%, 50% and 75% recycled mixtures (mixtures 4, 5 and 6).
concrete’s strength, the substantial decrease in internal forces
These mixtures all had strengths greater than the control mixture
caused by air entrainment far exceeds any concern over the reduc-
at 28-days but had between 3.5% and 2.5% lower air contents.
tion in strength. However, this does not imply the strength is not
Although the air contents of these mixtures were sure to have
important. Although air helps alleviate the internal stresses, it will
caused the decent results, the strength differences between these
not eliminate them completely relying on strength to be another
mixtures and control, combined with the lower DF results, raise
important factor in freeze–thaw durability. To test the ability for
questions as to the quality of strength produced by high content
concrete mixtures in this study to undergo freezing and thawing cy-
slag cement mixtures. For instance, mixture #6 had a high air con-
cles ASTM C 666 Procedure A was chosen as the test method [35].
tent (7.5%) and strength in excess of the control, but the DF values
For this research, two freeze–thaw beam specimens were fabri-
were 13–18 points lower. Although the difference in results may
cated from each concrete mixture. Although ASTM C 666 states
have been caused by the difference in air contents, a secondary
that specimens shall be tested at 14 days of age, this is not a real-
cause may be present. It has been postulated that the strength of
istic approach when dealing with concretes containing large
slag cement is more C2S controlled rather than C3S which corre-
amounts of supplementary cementitious materials due to the
lates with later age, rather than early age strength gains. Whether
slower strength gains. Therefore, it was decided to wait until
the strength or structure of the cementitious paste formed by
28 days of age before testing. During the course of testing, dynamic
hydration of slag cement is as durable as pure cement paste may
and static frequencies were recorded each week, as well as mass
be argued based on these results.
loss. How much these frequencies drop over the course of the test-
ing established the durability factor (DF) for the mixtures. The
durability factors (DF), air contents, final dynamic modulus and 5.2.4. Alkali-silica reactivity testing
the 28-day strengths are shown in Table 12. These strengths are The three mixtures tested for ASR potential were tested for 14-
the ‘‘actual’’ tested strengths (i.e. not the ‘‘normalized’’ strengths). days according to ASTM C 1567. Fig. 3 shows the results of testing
Dynamic modulus, number of cycles and durability factors are the these three mixtures. As previously stated, the three mixtures were
average from two specimens tested. designed as follows; mixture #1 had 100% PC as cementitious, mix-
The results show that the recycled materials content may have ture #2 had 50% slag cement and 50% PC as cementitious and mix-
an effect on the durability of a concrete mixture. When observing ture #3 had 100% slag cement as cementitious. All mixtures had
Table 12 it was shown that the recycled materials do not have an 100% waste glass as aggregate. The curves shown in Fig. 3 display
effect when used at replacement levels up to 50%. However, at the average (3 specimens) percentage of longitudinal expansion.
replacement levels in excess of 50% the durability factors begin According to ASTM C 1567, if a specimen expands less than
to drop. What can be seen was the difference in durability with re- 0.10%, it is believed to be an acceptable mixture and will have a
gards to air content. Mixtures #1, 4, 5 and 6 all had air content at or low risk of deleterious expansion. Expansion more than 0.10%,
above the target air content of 6.5%. All of these mixtures also had the risk of deleterious expansion in the mixture exists. Expansion
durability factors above 80. A concrete with a DF of 60 or more is more than 0.20%, there is a high risk of deleterious expansion in
considered to be able to perform well in conditions where freezing the mixture.

Table 12
Freeze–thaw testing results for all mixtures.

Mixture identification & number Air content (%) 28-Day strengtha Dynamic modulus, (P) Number of cycles (N)b Durability factor (DF)
lb/in2, (MPa) (Hz)
1 CC (Control) 10.0 4300 (29.6) 96.5 320 103
2 100-RA-C 5.2 4375 (30.2) 68.1 272 64
3 100-RA-BF 4.6 3836 (26.4) 56.5 259 49
4 25-RA-BF 6.4 5604 (38.6) 94.5 310 98
5 50-RA-BF 7.0 6674 (46.0) 95.0 320 102
6 75-RA-BF 7.5 5148 (35.5) 84.5 310 88
a
Actual 28-day strength.
b
Number of cycles at which P reaches 60%, or test length, whichever was less.
436 P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437

Fig. 3. ASR testing results for mixture #1, #2 and #3.

These results indicate that the waste glass aggregate used in the 6. Conclusions
research concretes does have potential to react with the alkalis in
the cementitious paste. Mixture #1 shows that the potential for This research study was designed to test the effects of recycled
ASR expansion is high. This mixture had an average expansion of materials, in varying amounts, on the fresh and hardened proper-
0.239% (high of 0.399%, low of 0.238%). The use of 100% PC with ties of concrete. The purpose of this research was to determine
100% waste glass as an aggregate would not be a reasonable choice whether these recycled materials would have negative or benefi-
based on these results. Mixture #2 with 50% slag cement shows lit- cial effects on concrete. If these effects were detrimental, at what
tle risk of deleterious expansion. This mixture had an average point they become detrimental was of importance. The results of
expansion of only 0.022% (high of 0.028%, low of 0.018%). The this research showed that recycled materials can be incorporated
use of 50% slag cement as cementitious material is suitable based into a concrete without detrimental effects. The authors believe
on these results. It may be possible that replacement less than that the term ‘‘beneficial’’ should not be confused with ‘‘increase’’.
50% could have achieved similar results. However, based on previ- When incorporating recycled materials into a concrete mixture the
ous studies, replacement amounts in excess of 35% were required benefit is immediately achieved, provided no significant detrimen-
to decrease expansions below 0.10% when reactive aggregates tal effects are realized from this incorporation. When replacing
were used in similar tests [27]. Mixture #3 with 100% slag cement natural virgin aggregates and cement with materials that would
as cementitious proved to be an acceptable combination with the otherwise be disposed of in a landfill, the benefit is obvious. The
waste glass. This mixture had an average expansion of 0.022% (high use of these recycled aggregates was never believed or anticipated
of 0.030%, low of 0.018%). The use of 100% slag cement as cemen- to ‘‘increase’ a concrete’s strength or durability, rather an investi-
titious material in combination with waste glass is suitable based gation was made into whether these materials would significantly
on these results. Because the results for mixture #2 and #3 are decrease these properties when compared to a normal concrete
essentially identical, the use of slag cement at replacements be- mixture. The same principle applies to the replacement of PC with
yond 50% would therefore appear to be useless, from an ASR mit- slag cement. Ultimately the ‘‘beneficial’’ use of recycled materials is
igation standpoint. Since slag cement is alkali activated, there realized when these recycled concrete mixtures achieve similar
actually may be two mechanisms occurring in mixture #2 and properties when compared with a normal concrete. The primary
#3. There are studies on the use of alkalis (added to accelerate findings of this research are summarized in the following bullets:
the hydration) in a slag cement concrete, but how much of the con-
centration is used during hydration is unknown. According to  A replacement level up to 50% with recycled materials were
ASTM C 989 and other studies, the use of slag cement to mitigate determined to be non-detrimental to a concrete mixture with
ASR potential has been attributed to the dilution of the alkalis regards to hardened properties, and was determined to be the
within the cementitious paste. This is similar to the use of other optimum replacement level. Replacement levels up to 50% recy-
SCM’s such as fly ash. However, the slag cement may also be using cled materials actually enhanced concretes properties. A reduc-
the alkalis as well which would also reduce the concentration. tion in quality began to manifest at 75% and was fully visible at
The results of this test should be used with caution. For in- 100% replacement.
stance, comparison of mixture #2 which was made from 50% slag  Replacement of natural virgin aggregates with RCA and crushed
cement with the 50% recycled concrete cannot be made. The 50% waste glass decreases the workability of a concrete mixture. The
recycled concrete had only 50% waste glass (not 100% waste glass) extremely high absorption capacity of the RCA coupled with the
and RCA as coarse aggregate. The 50% recycled concrete, based on harshness of the waste glass decreases concretes workability. A
these results, would probably perform well because it only has 50% small amount of virgin natural aggregates (25%) will greatly
waste glass as fine aggregate and has 50% slag cement as cementi- improve workability and the effects of HRWRA.
tious. A valid comparison can be made with regards to the 100%  The use of waste glass aggregates without the inclusion of slag
recycled concrete and mixture #3. Although the 100% recycled cement has detrimental effects on the hardened properties of a
concrete had RCA as coarse aggregate, it also had 100% waste glass concrete. A mixture with 100% PC and waste glass as aggregate
as fine aggregate. Based on these results, the 100% recycled con- was shown to have very high expansions when tested for ASR
crete would not experience deleterious expansions caused by potential. The use of 50% slag cement as cementitious was
ASR. These results may also confirm previous beliefs that the per- found to mitigate these expansions to a negligible level.
formance of the 100% recycled aggregate concrete (mixture #2 and  A replacement amount of 50% recycled materials was deter-
mixture #2 re-batch) may have been attributed to ASR expansion mined to be beneficial to a concrete mixture. A concrete com-
within the concrete. posed of 50% recycled materials achieved significantly greater
P.L. Maier, S.A. Durham / Construction and Building Materials 29 (2012) 428–437 437

strengths than a control mixture. An ultimate strength of nearly [6] Wang Ling, Tian Pei, Yao Yan. Application of ground granulated blast furnace
slag in high performance concrete in China. s.l. International workshop on
7000 psi (48.3 MPa) with 6.5% air content at 28-days of age was
sustainable development and concrete technology; 2005.
achieved. The permeability of this mixture was also much lower [7] Osborne GJ. Carbonation and permeability of blast furnace slag cement on
than the control concrete. The freeze–thaw durability of this concretes from field structures. Detroit: American Concrete Institute; 1989.
concrete was substantial and not affected by the recycled [8] Hale Micah. The effect of ground granulated blast furnace slag on the fresh and
hardened properties of concrete. Norman: University of Oklahoma; 2001.
materials. [9] Sivasundaram V, Malhotra VM. Properties of concrete incorporating low
 A replacement amount of 25% recycled materials was deter- quantity of cement and high volumes of ground granulated slag. ACI Mater
mined to be beneficial to a concrete mixture. A concrete com- J 1992;89 [6, sl: ACI].
[10] ACI. ACI committee report 233R-95 (reapproved 2000), ground granulated
posed of 25% recycled materials achieved slightly lower early blast-furnace slag as a cementitious constituent in concrete. Detroit: American
age strengths and greater later age strengths than a control Concrete Institute; 2000.
mixture. An ultimate strength of nearly 6700 psi (46.2 MPa) [11] Gholamreza Fathifazl AG, Razaqpur O, Burkan Isgor, Abdelgadir Abbas, Benoit
Fournier, Simon Foo. A novel method for proportioning structural concrete
with 6.5% air content at 90-days of age was achieved. The per- mixes made with recycled concrete aggregate. Concr Int Mag of ACI 2009 [July
meability and freeze–thaw durability of this concrete was equal 26].
to a control mixture. [12] Transportation Applications of Recycled Concrete Aggregate. US department of
transportation, federal highway administration. FHWA state of practice
 A replacement amount of 75% recycled materials was deter- national review; 2004.
mined to be non-detrimental to a concrete mixture. A concrete [13] Mirjana Malešev, Vlastimir Radonjanin, Snežana Marinković. Recycled
composed of 75% recycled materials achieved slightly lower concrete as aggregate for structural concrete production. Sustainability 2010.
[14] Polley, Craig. The effects of waste glass aggregate on the strength and
early age and later age strengths than a control mixture. An ulti-
durability of Portland cement concrete. Madison: Department of Civil and
mate strength of nearly 6350 psi (43.8 MPa) with 6.5% air con- Environmental Engineering, University of Wisconsin Madison; 1996.
tent was achieved at 90-days of age. The permeability of this [15] Ahmad Shayan, Aimin Xu. Performance of glass powder as a pozzolanic
concrete was significantly lower than the control mixture. The material in concrete: a field trial on concrete slabs. Cement Concr Res
2006;36:457–68.
freeze–thaw durability of this mixture was only slightly lower [16] PCA. Design and control of concrete mixtures. 14. Skokie: Portland Cement
than that of a control and was still considered satisfactory (DF Association; 2005.
above 80). [17] ACI. ACI committee report 555, removal and reuse of hardened concrete.
Detroit: American Concrete Institute; 2001.
 A replacement amount of 100% recycled materials was deter- [18] Karthik Obla, Haejin Kim, Colin Lobo. Crushed returned concrete as aggregates
mined to have detrimental effects on a concrete when com- for new concrete. s.l.: RMC Research and Education Foundation, NRMCA
pared with a control mixture. The compressive strength of a Research Laboratory; 2007.
[19] Hogan FJ, Muesel JW. Evaluation for durability and strength development of a
concrete composed of only slag cement is significantly lower ground granulated blast furnace slag. Cement, Concr Aggr 1981;3(1):40–52.
at all age groups. An ultimate strength of 4200 psi (29.0 MPa) [20] Richardson David N. Strength and durability characteristics of a 70% ground
with 6.5% air content was achieved at 90-days of age. The per- granulated blast furnace slag concrete mix. Rolla: Missouri Transportation
Institute & Missouri Department of Transportation; 2006.
meability of this concrete was significantly lower than the con- [21] Sippel, Steven Cramer, Chad. Effects of ground granulated blast furnace slag in
trol mixture. The freeze–thaw durability of this concrete was Portland cement concrete. Madison: Department of Civil and Environmental
very low. This concrete achieved a normal strength classifica- Engineering, University of Wisconsin-Madison; 2005. p. 62 [Final report of
findings].
tion at 90-days of age and if slow strength gain is not a concern,
[22] ASTM C-989. Standard specification for slag cement for use in concrete and
as well as climate, this concrete may be a suitable candidate. mortars. West Conshohocken: American Society for Testing Materials; 2010.
Coupled with the immense benefit of using a concrete com- [23] Reiner Mark. Technology, environment, resource and policy assessment of
posed of entirely recycled materials, there is most certainly sustainable concrete in urban infrastructure. Denver: University of Colorado
Denver; 2007.
applications where this concrete could be used successfully. [24] Shi C, Stegemann J, Caldwell R. Effect of supplementary cementing materials
on the specific conductivity of pore solution and its implications on the rapid
chloride permeability test (AASHTO T277 and ASTM C1202). ACI Mater J
1998:389–94 [s.l.: ACI, July–August].
Acknowledgments [25] Lane DS, Ozyildirim Celik. Ph.d. Combinations of pozzolans and ground
granulated blast furnace slag for durable hydraulic cement concrete.
Charlottesville: Virginia Transportation Research Council & US Department
The authors would like to acknowledge CTL Thompson, Materi- of Transportation Federal Highway Administration; 1999.
als Engineering, Inc. in Denver, CO for their assistance in testing the [26] Weihua Jin, Christian Meyer, Stephan Baxter. Glascrete – concrete with glass
ASR Specimens. The authors would like to thank Rocky Mountain aggregate. ACI Mater J 2000:208–13.
[27] Detwiler Rachel J. PCA’s guide specification for concrete subject to alkali-silica
Bottling, Inc. and Oxford Recycling Inc. for their generous dona- reactions: mitigation measures. Skokie: Portland Cement Association; 2003
tions of RCA and waste glass. In addition, the authors would like [PCA research and development information].
to thank Holcim, Inc., the U.S.B.R. and LEHIGH Cement for their [28] McLellen G, Higgins DD. Effectiveness of slag cement in preventing alkali-silica
reaction: ten-year results. Farmington Hills: American Concrete Institute;
generous donations of cementitious materials used in this study. 2009 [ACI SP-263-3].
[29] Mindess, Young, Darwin. Concrete. In: Horton Marcia, editor. 2nd ed. Upper
Saddle River: Pearson Education, Inc.; 2003. p. 644.
References [30] ASTM C 1567. Determining the potential alkali-silica reactivity of
combinations of cementitious materials and aggregate. West Conshohocken:
[1] <www.minerals.usgs.gov>. United States geological survey.<http://www. American Society for Testing Materials; 2009.
minerals.usgs.gov/minerals/pubs/commodity/cement/mcs-2010-cemen.pdf>; [31] ASTM C 33. Standard specification for concrete aggregates. West
2011. [cited: 9.09.11.]. Conshohocken: American Society for Testing Materials; 2008.
[2] <www.cement.org>. Portland Cement Association. <http://www.cement.org/ [32] ASTM C 702. Standard practice for reducing samples of aggregate to testing
smreport09/sec_page3_1.htm>; 2011. [cited: 9.09.11.]. size. West Conshohocken: American Society for Testing Materials; 2007.
[3] <http://www.epa.gov>. United States environmental protection agency. <http:// [33] ASTM C 136. Standard analysis of fine aggregate and coarse aggregates. West
www.epa.gov/climatechange/emissions/downloads11/US-GHG-Inventory- Conshohocken: American Society for Testing Materials; 2006.
2011-Annex-1.pdf> ;2011. [cited: 9.09.11.]. [34] ASTM C 1202. Standard test method for electrical indication of concrete’s
[4] Verdugo, Cesar David. The practicality, versatility and feasibility of utilizing ability to resist chloride ion penetration. West Conshohocken: American
recycled glass as a concrete aggregate. Gainsville: University of Florida, Civil Society for Testing Materials; 2010.
Engineering Department; 2009. [35] ASTM C 666. Standard test method for resistance of concrete to rapid freezing
[5] <en.wikipedia.org>. Wikipedia, the free encyclopedia. <http://en.wikipedia.org and thawing. West Conshohocken: American Society for Testing Materials;
/wiki/Soda-lime_glass>; 2011. [cited: 3.04.11.]. 2008.

You might also like