7th UK CARE Annual General Meeting

UK Chinese Association of Resources and Environment

Greenwich, 15 September 2007

USE OF WASTE GLASS AS AGGREGATE IN CONCRETE

Liang, Hong1; Zhu, Huiying2; Byars, Ewan A.3

1
the University of Edinburgh, UK; h.liang@ed.ac.uk
2
Deputy General Engineer at GuangXi Research & Design Institute of Building Science, China
3
the University of Sheffield, UK

ABSTRACT
The increasing awareness of glass recycling speeds up inspections on the use of waste glass with
different forms in various fields. One of its significant contributions is to the construction field where
the waste glass was reused for value-added concrete production. Literature survey indicates that
the use of waste glass as aggregates in concrete was first reported over 50 years ago. The
concomitant alkali-silica reaction (ASR) by using glass in concrete and its unique aesthetic
properties have been investigated since then. However, no complete solution to ASR has been
found and the application of glass in architectural concrete still needs improving. Laboratory
experiments were conducted in the University of Sheffield to further explore the use of waste glass
as coarse and fine aggregates for both ASR alleviation as well as the decorative purpose in
concrete. This paper presents mainly the latter aspect, in which study, both fresh and hardened
properties of architectural concrete were tested. Results demonstrate that the use of waste glass as
aggregate facilitates the development of concrete towards a high architectural level besides its high
performances, thereafter, the increasing market in industry.
KEYWORDS
alkali-silica reaction (ASR), aesthetic properties, glass aggregate
INTRODUCTION
Waste glass is a major component of the solid waste stream in many countries [1]. It can be found
in many forms, including container glass, flat glass such as windows, bulb glass and cathode ray
tube glass. At present, although a small proportion of the post consumer glass has been recycled
and reused, a significant proportion, which is about 84% of the waste glass generated in UK [2], is
sent to landfill. Glass is a 100% recyclable material with high performances and unique aesthetic
propertieswhich make it suitable for wide-spread uses. Besides, the current recycling state and
legislative forces pose great pressures on glass recycling and reusing. The use of glass as
aggregates in concrete has great potential for future high quality concrete development.

However, deleterious alkali-silica reaction (ASR) might occur in glass concrete due to its high silica
constituent. Some solutions have been formed to alleviate ASR, but these solutions have some
limitations which made it still particularly important to investigate the utilisation of glass in concrete.
These limitations include the long-term inspecting of the effectiveness of ASR suppressants; the
effectiveness of the combined use of ASR suppressants; the effective dosage of suppressants in
mitigating ASR.
In order to develop concrete with high aesthetic standard and less ASR potential, a series of
laboratory experiments were conducted with the following objectives: (a) to investigate ASR in
concrete using certain suppressants; (b) to develop concrete mixes using glass as a coarse
aggregate replacement; and (c) to obtain a high architectural level by using glass as decorative

aggregates. The first stage ASR study confirms the effectiveness of certain suppressants in
mitigating ASR of glass concrete. Based on this research finding, assumptions are made where the
ASR effect has been reduced to a certain extent that could be outside the consideration in the
second stage architectural study. The paper here describes the architectural study of waste glass in
concrete, encompassing the fresh and hardened properties tests, along with the aesthetic standard
improvement by introducing pigments. The space between lines must be single; the space between
paragraphs, one blank line.
EXPERIMENTAL WORK
As aforementioned, within the series of experiments, the architectural concrete study is of main
concern in this paper, where, glass was used both as coarse and fine aggregates with the purpose
of achieving a high performance and aesthetic level of concrete. Details of the experimental work
are given in the following sections.
Materials
1. Cement
In the architectural concrete study, white Portland cement (WPC) was used. This special cement
has very low alkali content, thus could potentially mitigate the ASR. Meanwhile, the white colour
shows its benefit for universal matching with different coloured glasses when glass concretes are
made.
2. Cement replacement materials (CRM)
Based on the ASR study tests results, 10% Metakaolin (MK) and 20% super-classified pulverised fly
ash (SPFA) were used to replace part of the cement, both within the recommended range, and were
verified to behave better in inhibiting ASR compared with other CRMs such as ground granulated
blast furnace slag (GGBS).
3. Coarse aggregate (CA)
Different types of glass of varying sizes and colours were used instead of the 5-10mm normal
aggregate (NA) in traditional concrete control mixer. The specific gravity of glass is 2.50.
4. Fine aggregate (FA)
Dolomite, which has a pure white colour, was widely used in the architectural concrete
development. Its shape and size have potential benefit in obtaining a good particle size distribution
in glass concrete. 1-3mm glass particles were also used as fine aggregate in glass concrete to
make a suitable colour distribution in some mixes.
5. Other materials
Pigments were added to produce coloured architectural concrete. Beside, superplasticiser (SP) was
used to reduce the amount of water required and to obtain good workability.
TEST METHODS
1. Fresh concrete property tests
Three tests were carried out to examine the fresh concrete workability, air content as well as the
density. Among them, the workability of fresh concrete was examined by the slump test. The test
set-up is conducted as recommended in BS EN 12350-2:2000 [3]. In the air content test, the
pressure gauge method [4] was adopted. In the density test, fresh concrete is compacted into a rigid
and watertight container of known volume V and known mass M 2 and is then weighed M 1 . The
density is calculated by Formula (1)
D

M 2 M 1
M

V
V

(1)

2. Hardened concrete property tests

The compressive strength test adopts the method specified in BS EN 12390-3: 2002 [5]. The
CARE 2007 Greenwich

specimens were loaded to failure in a compression testing machine and the maximum load was
recorded. For the tensile strength test, the method used is as specified in BS EN 12390-6:2000 [6].
The water absorption test aims at inspecting the capability of the architectural concrete to absorb
water. The method is based on the BS 1881-122:1983 [7].
RESULTS AND DISCUSSIONS
A number of coloured glass concretes using different colours and particle sizes of glass aggregates,
white cement with CRM materials and pigments were developed at laboratory and some of the
samples were given in Fig. 1. These samples demonstrate the achievements of the use of coloured
glasses in improving aesthetic standard for concrete.
Fresh concrete property tests
The results of the fresh concrete properties test are given in Table 1. A satisfactory agreement has
been found from these results.

Figure 1. - Polished coloured glass concrete

Table 1 - Fresh concrete property test results

3

Mix Code

Slump (mm)

Air Content (%)

Density (kg/m )

LH-A-A

48.00

1.00

2.30

LH-A-B

56.00

3.60

2.22

LH-A-C

110.00

2.10

2.26

120.00

1.50

2.29

LH-A-D
Mix No. Description:
LH-A-A
LH-A-B
LH-A-C
LH-A-D

CARE 2007 Greenwich

Glass Aggregate and MK without Pigment

Glass Aggregate and MK with Pigment
Glass Aggregate and SPFA without Pigment
Glass Aggregate and SPFA with Pigment

1. Slump test results

As shown in Table 1, the slumps are in the BS recommended range of 10-200mm [3], indicate that
the test results are valid in this experiment.

Figure 2. - Slump of architectural concrete

It can be seen from Fig. 2 that the workability of glass concrete containing SPFA was much higher
(about twice as much) than that of the concretes containing MK as CRM. This is quite reasonable,
since, MK is a pozzolanic material obtained through thermal activation of kaolinite. Hence, it could
react rapidly with the calcium hydroxide in the cement paste, converting it into stable cementitious
compounds, thus refining the microstructure of concrete, thereby reducing its workability as well as
permeation properties. In the meanwhile, the workability is governed mainly by the surface area and
shape of pozzolan. The mix with SPFA would have more powdered form of ingredients and thus
less resistance due to its smoother surface, leading to the higher workability.
For concretes with the same kind of CRM, the workability is very similar with or without pigment
because of the small amount of pigments used in the trial mixes.
2. Air Content Test Results
As specified in BS, the air content should be in the range of 0% to 10% [4]. The test results in Table
1 shows that the results conform to the specified range and the test is satisfactory.
3. Density Test Results
The densities are approximately 2300kg/m3 and conform to the assumption of mix design.
Hardened concrete property tests
1. Compressive strength test
1) Failure mode
A typical failure mode of the architectural glass concrete tested for compressive strength is shown
in Fig. 3. The four exposed faces were cracked approximately equally, with little damage to faces in
contact with the platens. Similar failure mode has been found in the BS [5] whereas a satisfied test
has been verified.

CARE 2007 Greenwich

Figure 3. - Typical failure mode of architectural glass concrete

2) Compressive strength development

Two sets of glass concrete specimens were used for each mix code, the mean compressive
strength was measured and calculated at 7, 14 and 28 days individually and the results are given in
Table 2 and Fig. 4.
Table 2 - Mean compressive strength (MPa)
Mix Code

7 days

14 days

28 days

LH-A-A

39

58

59

LH-A-B

47

54

54

LH-A-C

36

39

44

LH-A-D

35

42

49

Compressive strength (MPa)

60
55
50
45
40

LH-A-A

LH-A-B

LH-A-C

LH-A-D

35

30
0

10

15

20

25

30

Age duration (days)

Figure 4. - Mean Compressive Strength

High compressive strength concrete with the value above 40MPa can be obtained by using
coloured glassed as aggregates in concrete, Table 2 and Fig. 4. As expected and commonly, the
compressive strength at earlier ages is much smaller than that at later ages and, after 7 days, the
rate of strength development decreases. This is consistent with the findings by Neville [9] that the
use of CRM can reduce the early strength and increase the later age strength.
CARE 2007 Greenwich

In addition, the strengths of mixes LH-A-A and LH-A-B are higher (around 20%) than the other two
mixes. This means that using MK as CRM can increase concrete strength more significantly than
using SPFA. This is attributed to the fact that MK has a much larger surface area than SPFA.
Therefore, more particles will react and be consumed finally in the hydration process. Besides,
glass is found to be another influence factor due to its effect on the surface area. 60% of 3-6mm
glass and 40% of 6-12mm glass were used in the first two mixes, while 100% of 6-12mm glass were
used in the last two mixes. The first two mixes have more powdered form of constituents and thus a
larger surface area can be resulted in; hence, affecting the strength development.
Moreover, the w/c ratio could affect the strength development of concrete. Although the initial w/c
ratio is the same for the four architectural concrete mixes, the absorption of water by the aggregate
can affect the final w/c ratio, which then leads to the strength difference. The higher the w/c ratio is,
the lower the strength. Therefore, for the first two mixes, a larger amount of aggregate absorbs a
greater quantity of water. The effective w/c ratio of these two mixes has been reduced and thus,
higher strength would be resulted in.
2. Tensile Strength Test
The tensile strength results of glass concrete are shown in Fig. 5.

Figure 5. - Mean tensile strength

It can be seen from Fig. 5 that the trend of the value of the tensile strength is similar to that of the
compressive strength for the four architectural glass concretes tested. This indicates that the
effective w/c ratio has a similar influence to the tensile strength as to the compressive strength and
this is likely related to the water absorption performance of concrete, which will be explained further
in the water absorption test section below.
3. Water Absorption Test
The water absorption is calculated as the increase in mass resulting from immersion expressed as
a percentage of the mass of the dry specimen [10] and the test results are shown in Fig. 6.

CARE 2007 Greenwich

Figure 6. - Water absorption

It can be seen from Fig. 6 that the first two concrete mixes have similar water absorption
performance as well as the second two mixes and the former absorbs more water than the later.
This discrepancy could be explained by the fact that, the first two mixes have larger surface areas
due to the use of MK and smaller glass particle size than the second two mixes.
As mentioned in section 3.2.1, the absorption of water is related to the strength development.
Similar trends of water absorption and compressive strength development could be found when
compared Fig. 4 to Fig. 6. LH-A-B has the lowest water absorption as well as the lowest strength
among the four mixes. Such is the case for LH-A-A, LH-A-C and LH-A-D. This finding also confirms
that higher water absorption would result in the reduction of the effective w/c ratio, and thus
increasing the strength.
CONCLUSIONS
The paper presents the effective utilisation of coloured glass aggregates in a range of architectural
concretes and their properties tests. The performance test results conducted in this research
confirm that the properties of those special mixed concretes are satisfactory. The properties tested
include workability, air content, density, compressive strength, tensile strength, and water
absorption. Moreover, it is found that water absorption is strongly related to the strength of the
concrete. Ultimately, glass is found to be an ideal material as a decorative aggregate in architectural
concrete with its satisfactory performances and aesthetic property improvement.
Acknowledgements
Acknowledgements are to the companies for supplying materials during the experiments. Thanks
are also given to the Department of Civil and Structural Engineering, University of Sheffield for
technical support.
References
1.

Shao Y, Lefort T, Moras S and Rodriguez D. Studies on Concrete Containing Ground Waste Glass. Cement and
Concrete Research, Vol. 30, pp 91-100, 2000.
2. Enviros. Recycled Glass Market Study & Standards Review. The Waste & Resources Action Programme, 2004.
3. British Standards Institution. Testing fresh concrete-Part 2: Slump test. BS EN 12350-2, BSI, London, 2000.
4. British Standards Institution. Testing fresh concrete-Part 7: Air content-Pressure methods. BS EN 12350-7, BSI,
London, 2000.
5. British Standards Institution. Testing hardened concrete-Part 3: Compressive strength of test specimens. BS EN
12390-3, BSI, London, 2002.
6. British Standards Institution. Tensile splitting strength of test specimens. BS EN 12390-6, BSI, London, 2000.
7. British Standards Institution. Testing concrete-Part 122: Method for determination of water absorption. BS
1881-122, BSI, London, 1983.
8. BRE. Alkali-Silica Reaction in Concrete: Detailed Guidance for New Construction. BRE Centre for Concrete
Construction, Digest 330 Part 2, 1999.
9. Neville, A M. Properties of Concrete. 4th ed. Harlow: Longman, 1995.
10. British Standards Institution. Testing concrete-Part 122: Method for determination of water absorption. BS
1881-122, BSI, London, 1983.

CARE 2007 Greenwich