Cement and Concrete Research 30 (2000) 497 ± 499

Technical Note
The use of waste ceramic tile in cement production
È nal
Nuran Ay*, MevluÈt U
Ceramic Engineering Department, Anadolu University, Yunusemre Kampusu, 26470 Eskisehir, Turkey
Received 29 August 1999; accepted 3 January 2000

Abstract

In ceramic tile production, because of various reasons, unsold fired products come out. These are waste tiles and only a little part of them
are used. Remainings create environmental problems. If these waste tiles are used in cement production, this pollution decreases. In this
study, usage of waste tile as pozzolan was studied. Waste tile was added into Portland cement in 25%, 30%, 35%, and 40% weight ratios.
Pozzolanic properties of waste tile and setting time, volume stability, particle size, density, specific surface area, and strength of cement
including waste tile were investigated. The test results indicated that the waste tiles show pozzolanic properties, and chemical and physical
properties of the cement including tile conforms to cement standard up to the addition of 35% waste tile. D 2000 Elsevier Science Inc. All
rights reserved.

Keywords: Waste ceramic tile; Physical properties; Pozzolan; Cement

1. Introduction as a pozzolan. Can ground tile be used as a pozzolan? In

this study, the answer to this question was investigated.
When they are used alone, pozzolans do not show
binding properties. Pozzolans gain binding property as a
result of its chemical reaction with water when used with 2. Experimental work
lime or cement [1± 6]. Pozzolans are divided into two
groups: natural pozzolans and artificial pozzolans. Lime- 2.1. Materials
pozzolans are the oldest construction materials [7± 9].
One important artificial pozzolan is fired clay. It was In this study, waste glazed ceramic tiles, which are
used with lime before cement was invented [2,10]. The scrap from a ceramic tile plant and Portland cement,
clay minerals have a crystal structure. Therefore, clays in
raw form do not possess pozzolanic properties. However,
by heat treatment, clays become highly pozzolanic. Heat Table 1
treatment destroys the crystal structure of clays and Chemical analysis of tile and cement
forms a quasi-amorphous or disordered alumina silicate Materials Tile (%) Cement (%)
structure [11]. The reactions occurring between a pozzo- SiO2 63.29 20.52
lanic structure and a cement surface is investigated in Al2O3 18.29 5.46
various research [12 ± 16]. It is found that silicate hy- Fe2O3 4.32 3.64
drates produced in reactions resemble Portland cement CaO 4.46 65.04
MgO 0.72 1.35
hydration product.
P2O5 0.16 0.04
Raw materials of ceramic tile are clays, quartz, and K2 O 2.18 0.68
feldspar. Tile is fired at 1100± 1200°C. Brick and roof tiles Na2O 0.75 0.17
are clay products. Brick and roof tile powder can be used SO3 0.10 2.18
Clÿ 0.005 0.002
TiO2 0.61 0.27
SrO2 0.02 0.03
* Corresponding author. Tel.: +90-222-3350580, ext. 6354; fax: +90-
Mn2O3 0.05 0.07
222-3239501.
L.O.I. 1.61 1.16
E-mail address: [email protected] (N. Ay).

0008-8846/00/$ ± see front matter D 2000 Elsevier Science Inc. All rights reserved.
PII: S 0 0 0 8 - 8 8 4 6 ( 0 0 ) 0 0 2 0 2 - 7
498 È nal / Cement and Concrete Research 30 (2000) 497±499
N. Ay, M. U

Table 2 Table 4
Particle size analysis of samples Normal consistency water, initial and final setting time of samples
% Under Tile (%) Water (ml) Initial setting time (h) Final setting time (h)
Size (mm) Tile, 25% Tile, 30% Tile, 35% Tile, 40% 25 90 + 28 2.50 3.40
30 90 + 27 2.50 3.35
200 ± ± 99.91 99.91
35 90 + 24 3.00 4.00
90 95.99 94.81 94.69 94.79
40 90 + 22 3.00 4.05
45 76.44 75.20 75.38 75.60
32 63.74 62.97 63.43 63.65
3 10.21 10.04 10.51 10.05

Table 5
Volume stability results
were used. Table 1 gives the chemical analysis of tile
Tile (%) a (mm) b (mm) c (mm) cÿa (mm)
and cement. Chemical analysis was done by ARL 8680
X-ray spectrofotometre. 25 16 18 19 3
30 14 15 15 1
35 15 16 16 1
2.2. Preparation of test specimens and method 40 15 15 15 0

The waste tiles were crushed in a jaw crusher and

ground in a ball mill for 55 min. The ground tile was
subjected to an activity test. A pozzolanic activity test was TS 25 [17] and A.S.T.M C 618 [2]. From this table it is
done by the Rilem Cemburean method. Portland cement obvious that waste tile can be used as pozzolan.
was blended with ground tile by weight ratios 25%, 30%, Drinking water was used in the preparation of
35%, and 40% in a ball mill for 30 min. Normal consis- cement including waste tile (300 g). Initial and final
tency was determined by Vicatring. The initial and final setting time and the amount of needed water are given
setting times of the specimens were measured in normal in Table 4. Results of the volume stability test are
laboratory environments. Volume stability was measured by given in Table 5. These results conform to TS 26 [18]
the Le ChatelieÂr method. Particle size analysis of each and TS 12144 [19].
specimen is given in Table 2. Particle size analysis was The density of cement including waste tile was mea-
done by using a Malvern Mastersizer-E particle sizer. sured and it was determined that sample density decreases
Density was determined by using a Le ChatelieÂr retort with increasing tile ratio. As it is seen in Table 6, specific
and Quanto Chrome multi Picnometre. Specific surface surface area conforms to TS 26. Specific surface area is
area was measured with TONI Technik Blaine equipment. higher than 2800 g/cm2.
Four different batches were prepared for the strength test. In The results of the bending and compressive strength
each batch, one unit of cement including tiles was mixed tests are given in Table 7. Strength results of samples
with three units of sand and one and a half unit of water. containing 25%, 30%, and 35% waste tile have enough
The mortar was placed in prismatic moulds having dimen- strength. Samples containing 40% tile have enough 7-
sions of 160  40  40 mm. Prismatic specimens were day strength but it failed the 28-day strength test.
subjected to 1, 2, 7, and 28 days compression and bending From the test results obtained in this study, it is
test by using a test machine (TONI COM, TC III (300 kN)). concluded that up to 35% waste tile can be added
to cement.

3. Results and discussion

4. Conclusion
Pozzolanic properties of ceramic tiles were measured.
Results are in Table 3. Pozzolanic properties conform to The test results obtained in this study show that:

- Ground waste tile has pozzolanic properties,

Table 3
Comparison of pozzolanic properties of waste tile and TS 25
TS 25 Tile
Table 6
(SiO2+Al2O3+Fe2O3), %  70.00 86.01 Density and specific surface area results of samples
MgO, %  5.00 0.72
Tile % Density (g/cm2) Specific surface area (cm2/g)
SO3, %  3.00 0.1
L.O.I., %  10.00 1.61 25 3.00 3732
7 days strength (N/mm2) 30 2.99 3513
Bending  10.00 12.8 35 2.98 3685
Compressive  40.00 54.8 40 2.92 3813
 È nal / Cement and Concrete Research 30 (2000) 497±499
N. Ay, M. U 499

Table 7
Bending and compressive strength test results of samples
1 day 2 days 7 days 28 days
Bending Compressive Bending Compressive Bending Compressive Bending Compressive
strength strength strength strength strength strength strength strength
Tile (%) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (N/mm2)
25 2.5 8.6 3.4 14.7 5.0 27.8 7.1 38.4
30 2.6 10.2 3.8 15.7 5.3 27.8 6.8 38.0
35 2.1 7.2 2.9 12.0 4.8 26.8 6.1 33.7
40 2.0 6.9 2.9 11.6 4.6 22.2 6.2 32.2

- Waste tile can be added into cement up to 35% [8] P.C. Hazra, V.S. Krisnaswamy, Natural pozzolans in India, their utility,
weight ratio, distribution and petrography, Records of the Geological Survey of
Indian 87 (4) (1987) 675 ± 706.
- Adding waste tile into cement reduces cost. [9] R.J.S. Spence, Lime and Surkhi Manufacture in Indian, in: R.J.S.
Spence (Ed.), Lime and Alternative Cements, Proceedings of One
Acknowledgments day Seminar on Small-scale Manufacturing of Cementitious Materi-
als, Intermediate Technology Development Group, London, 1974, pp.
The authors would like to thank Eskisehir Cement 18 ± 21.
[10] Vitruvius, The Ten Books of Architecture, Dover, NY, 1960 Bk. II,
Factories T.A.S for equipment for this project. Chap. VI.
[11] T.Y. Erdogan, Admixture for Concrete, Middle East Technical Univ.
Press, Ankara, Turkey, 1997.
References [12] B. Caroll, E. Freeman, The behaviour of colloidal silicate solutions as
revealed by adsorption indicators, J Phys Chem 58 (4) (1954) 335 ± 339.
[1] ASTM C 125, Standard Terminology Relating to Concrete and Con- [13] P. Barret, D. Menetrier, B. Cottin, Study of silica ± lime solution reac-
crete Aggregate, 1994 Annual Book of ASTM Standards. tion, Cem Concr Res 7 (1) (1977) 61 ± 67.
[2] ASTM C 618, Standard Specification for Coal Fly Ash and Raw or [14] R. Dron, LaÂctivite Pouzzolanique, Bulletin de Liason des Laboratoires
Calcined Natural Pozzolan for Use as a Mineral Admixture in Port- des Ponts et ChausseÂes 93 (1978) 66 ± 69.
land Cement Concrete, 1994 Annual Book of ASTM Standards. [15] K.M. Alexander, Reactivity of ultra fine powders produced from sili-
[3] S.N. Ghosh, Advances in Cement Technology, 1st edn. Pergamon, cious rock, J Am Concr Inst 57 (5) (1960) 557 ± 569.
Oxford, 1983. [16] A.L. Fraay, J.M. Bijen, Y.M. de Haan, The reaction of fly ash in
[4] F. Massazza, Puzzolanik Cimento Seminari, Ankara. concrete, a critical examination, Cem Concr Res 19 (1989) 235 ± 246.
[5] P.K. Mehta, Puzzolanic and cementitious by products as mineral ad- [17] TS 25, Trass, Turkish Standards Institute, Ankara, Turkey, 1975 (in
mixtures for concrete, fly ash, silica fuÈme, slag and other mineral by- Turkish).
products in concrete, ACI SP 79 (1) (1983) 1 ± 46. [18] TS 26, Cimento ± Trasli Cimento, TuÈrk Standartlari EnstituÈsuÈ, Ankara,
[6] P.K. Mehta, Concrete Structure, Properties and Materials, Prentice- Turkey, 1992 (in Turkish).
Hall, London, 1986. [19] TS 12144, Cimento ± Puzolanik, TuÈrk Standartlari EnstituÈsuÈ, Ankara,
[7] R. Malinowski, Y. Garfinkel, Prehistory of concrete, Concrete Inter- Turkey, 1997 (in Turkish).
national 13 (3) (1991) 62 ± 68.