Heat resistance of lightweight concrete with plastic aggregate from PET (polyethylene terephthalate)-mineral filler

1.   Introduction

Bottled drinking water is made based on the basic human need for drinking water that is easy to obtain, practical, clean, and healthy. Plastic was chosen as the packaging with its advantages, namely lightness, cheapness, comfortable shape, and recyclable. In its development, the demand for bottled water has increased from time to time. One million plastic bottles of bottled drinking water or soft drinks worldwide are sold to consumers every minute and increase by 20% by 2021 ^[1]. The increase will result in a figure of about 20000 bottles being purchased by consumers every second. In 2016 as many as 480 billion plastic bottles were sold worldwide, and this figure will be 583.3 billion in 2021. Meanwhile, Indonesia is the fourth largest plastic bottle producing country, with 4.82 billion bottles per year ^[2]. Weak waste management causes plastic bottle waste to be thrown away at final disposal or reaches the sea and threatens all marine animal life in it.

Generally, plastic bottles are made from highly recyclable polyethylene terephthalate (PET) type plastic. With a high, increasing number of products and single-use only, efforts to recycle are lagging. A material specialist who can find such materials must avoid single-use (and soon to be waste) products that do not decompose in nature. Simultaneously, ideas are also needed to reuse plastic waste that already exists (and with an increasing use) into useful materials.

Researchers in various parts of the world have reused PET plastic waste for building materials ^{[3,4,5,6,7,8,9,10,11]}. Mansour et al. ^[9] used original PET plastic bottles which functioned as a replacement for partition walls from traditional concrete blocks. Each PET plastic bottle is filled with dry sand, saturated sand, or filling only air, then stacked and bound using cement mortar. Each PET plastic bottle filled with dry sand, saturated sand, or no filling (air only), then assembled and bound using cement mortar. The results of his research show that the construction of plastic bottles filled with air can make one-story buildings with roof plates and has better thermal insulation than traditional concrete blocks. Another study uses recycled plastic waste from PET type mineral water bottles as coarse agammaegate in concrete, while fine agammaegate uses natural sand ^[3]. The research was conducted in several conclusions which are shown below. Concrete mixtures without chemical additives indicate that the compressive strength and tensile strength of the concrete are meager, namely 12 MPa and 1 MPa, respectively, due to the low adhesion strength between plastic agammaegate and cement mortar. In the next stage, determining the maximum plastic agammaegate size is 10 mm; the maximum cement water factor is 0.35 and using chemical admixture to adjust workability, showing that the strength of the concrete produced is up to 20 MPa. A similar study was carried out by making concrete mixtures with various W/C ratios, both for light agammaegates and natural agammaegates ^[5]. The results showed that PET agammaegate with adequate granulometry produced suitable quality mixtures with lower concrete density (68.88%) but with similar mechanical characteristics to natural concrete. At low W/C ratios (with a cement content of 300 kg/m³ or more), the mechanical strength of the concrete is determined by the coarse agammaegate characteristics. In contrast, at high W/C ratios, the characteristics of the cement paste are decisive.

Apart from being a substitute for coarse agammaegate, the use of fine agammaegate from PET plastic flakes as a substitute for fine natural agammaegate has been carried out ^[12,13]. The results show that the compressive strength, split tensile strength, and flexural tensile strength will increase at replacement levels up to 10%. More than this level, the mechanical characteristics of the concrete will decrease drastically. Research ^[13] also shows that the concrete modulus of elasticity will be smaller when using fine plastic agammaegate—the more the percentage, the more significant the decrease in the elastic modulus value.

Another benefit that can be used from the properties possessed by PET plastic is its tensile strength capacity, which is quite large, which varies from 263.72 to 550 MPa ^[14]. Plastic filaments extruded from PET plastic flakes were made which functioned as reinforcing fibers in the concrete mixture. The results show that at 1% fiber content (to cement volume), PET with high tensile strength (550 MPa) has the most favorable compressive strength and tensile strength when the W/C ratio is high (0.53). Meanwhile, PET with low tensile strength has the best compressive strength and tensile strength when the W/C ratio is low (0.38). The paper also states that studies using PET fiber as reinforcement indicate that the energy absorption capacity increases in the range from 400% to 700% compared to concrete without PET fibers, making it possible to use it for earthquake-resistant structures.

Other types of plastic besides PET have been carried out such as the use of E plastic, namely plastic that comes from the waste of used electrical or electronic devices such as used monitor tube wrappers ^[15]. The use of expanded polystyrene foam (EPS) plastic or better known as styrofoam ^[16,17,18], plastic types of polypropylene (PP) ^[19], plastic types of high-density polyethylene (HDPE) ^[20,21], plastic types of polyethylene (PE) and polyvinylchloride (PVC) ^[22], and even the use of used tires ^[23].

The use of plastic materials in concrete will reduce the strength of the concrete. One of the reasons is because the density of the concrete is also reduced. To overcome this, Kaseem et al. ^[24] suggest replacing cement with fly ash from 10% to 30%. His research showed that the 30% cement replacement was able to achieve the highest compressive strength, namely 21.28 MPa. This research also concluded that plastics could be used to develop low-strength concrete structures. Another disadvantage of using plastic agammaegate in concrete, either original or through further processing, is that it cannot absorb water, causing weak bonds between plastic agammaegate and mortar. This weakness can be explained below, namely that water that cannot be absorbed in the mortar-agammaegate interface will increase the W/C value, thereby reducing the concrete's quality in that zone. Also, plastic agammaegates generally have a smooth surface which causes the friction stress capacity between the mortar-agammaegates too low. Therefore, the interface zone of mortar-agammaegate in concrete with plastic agammaegate is a weak part of the binder-filler bonding system. To reduce these drawbacks researchers modified the water absorbency, surface roughness of plastic agammaegates, or both ^{[11,25,26,27,28,29]}.

Rumšys et al. ^[25] succeeded in grinding the plastic agammaegate surface mechanically by mixing the plastic agammaegate with sand in a mixer and rotating it for 5 minutes so that the plastic agammaegate's surface became a lot of scratches. However, the results showed that the surface roughness that occurred did not affect the strength of the concrete or water absorption. Another attempt to improve the plastic agammaegate surface was carried out by coating plastic agammaegate from PET bottles with ground granulated blast-furnace slag (GBFS) ^[26]. The results showed that the GBFS layer reacts with calcium hydroxide to form C–S–H bonds, strengthening the plastic agammaegate surface. Volcanic sand was chosen to coat polypropylene (PP) plastic agammaegate ^[27]. Results showed that the coating was able to increase the compressive strength of concrete compared to concrete with agammaegates without sand coating.

Another modification to improve water absorption or increase the surface roughness of plastic agammaegates is done by adding filler to the hot plastic melt during the agammaegate manufacturing process ^[11,29,30]. Fly ash with a carbon content of 12% is used as a filler mixed with HDPE type plastics to make artificial coarse agammaegate ^[29]. Three types of coarse agammaegate were prepared based on different proportions of filler-HDPE. Furthermore, using fine agammaegate from natural sand, five types of concrete are made based on the proportions of different materials. The results show that the mechanical characteristics and cracking properties of the concrete improve along with the fly ash content in the agammaegate. Meanwhile, Alqahtani et al. ^[30] using various types of fillers, namely: red sand, fly ash, and quarry fines, while the plastic waste used is of the type LLDPE (linear low density polyethylene).

Wiswamitra et al. ^[11] compared the use of various minerals, namely fine sand, fly ash, rice husk ash, and Portland pozolan cement as fillers in the manufacture of synthetic agammaegates (plastic agammaegates) with PET plastic bottle waste materials. Different types of plastic agammaegate are made based on different proportions of filler-PET. In the early stages of his research, the plastic coarse agammaegate that had been made was used to replace the natural coarse agammaegate. The results show that a higher compressive strength of concrete is shown in almost all types of fillers used, this is when compared to reference concrete (concrete with plastic agammaegate without filler). Synthetic agammaegate with a proportion of 1 rice husk ash:5 PET; and synthetic agammaegates with a proportion of 1 cement:3 PET; produces a high compressive strength of concrete and falls within the criteria for structural lightweight concrete.

The use of plastic as a material in concrete always raises curiosity about its resistance to heating. Various studies of heat resistance of concrete with plastic materials have been carried out. Koide et al. ^[3] concluded that the concrete with coarse agammaegate of PET plastic has heat resistance up to 60 ℃. Correia et al. ^[31] concluded in their research that during heating, the maximum temperature was higher for concrete with a plastic content than for concrete with natural agammaegates. Meanwhile, the decrease in compressive strength, modulus of elasticity, and split tensile strength in concrete with plastic content is higher than that of reference concrete.

This study will test the mechanical characteristics of lightweight concrete after heating, then compare it with reference concrete, namely concrete without heat treatment. The synthetic agammaegate used is a PET plastic filler mixed agammaegate divided into two types of agammaegate based on the filler used, namely rice husk ash and Portland pozzolana cement. This synthetic agammaegate uses proportions and preparation methods following the research of Wiswamitra et al. ^[11].

2.   Materials and methods

2.1.   Materials

The cement used is Portland pozzolanic cement (PPC) with a specific gravity of 3.15, with specifications that meet Indonesian standards, SNI 15-0302-2004 ^[32].

Fine agammaegate uses local sand obtained from dunes with a specific gravity of 2.14 and a Fineness Modulus (FM) of 2.68 with a gradation that meets the Zone 2 fine agammaegate gradations according to the grading limits for fine agammaegates listed in the Indonesian national standard, SNI 03-2834-2000 ^[33] (Table 1).

Natural coarse agammaegate using local gravel with a specific gravity of 2.68 with a size passing through sieve No. 3/8" (with size of sieve hole = 9.5 mm), stuck in sieve No. 4 (with the size of the sieve hole = 4.76 mm), and has an FM value of 6.00

For the manufacture of artificial agammaegate, the material used is plastic type PET (Polyethylene Terephthalate) without color from the packaging of beverage bottles in chopped form (Figure 1a), then the PET plate chopped is washed to remove impurities due to dirt or glue that may still stick, and finally dried in the sun. PET plastic has a specific gravity of 1.307, and a melting point of around 260 ℃. This study made two types of artificial agammaegate based on the type of filler mixed. The first filler is rice husk ash (Figure 1b) with a specific gravity of 2.73 which was previously sieved using sieve No. 16 (sieve hole size = 1.19 mm). The second filler is Portland pozzolana cement (Figure 1c) with a specific gravity of 3.15.

2.2.   Methods

This research was conducted in three stages (Figure 2), namely the manufacture of plastic agammaegates, casting of concrete specimens, and testing of mechanical characteristics which are described in more detail in the following section.

2.2.1.   Stage Ⅰ—production of type Ⅰ (PA Ⅰ) and type Ⅱ (PA Ⅱ) plastic aggregates

The plastic agammaegates are made like the flow chart shown in Figure 2b or in the sequence in Figure 3, with the first step being to melt the chopped PET plastic in a pan until all the plastic parts are melted. After the whole PET plastic becomes thick melt, then the filler powder (rice husk ash or cement) is poured into the pan, and the mixture is stirred again until the filler is evenly mixed. Then, the hot mix of the PET-filler mixture is poured into a wooden box mold and allowed to cool and harden. The hardened plastic block is then broken down using a hammer to make it smaller and put in the crusher. The resulting crusher fractions are then sieved to obtain fine and coarse agammaegate fractions. This synthetic agammaegate production follows the method of Wiswamitra et al. ^[11], which used filler of rice husk ash and Portland pozzolana cement. The ratio (weight ratio) of filler: PET plastic, namely one rice husk ash: five PET to produce plastic agammaegate type Ⅰ (PA Ⅰ) and one cement: three PET to create type Ⅱ plastic agammaegate (PA Ⅱ). Digital Microscope from Cooling Tech (Figure 4a) and Scanning Electron Microscope from Tabletop Microscope TM3030Plus (Figure 4c) was used to determine the microstructure synthetic agammaegates, before and after heating.

2.2.2.   Stage Ⅱ—casting concrete specimens

At this stage, the two types of plastic agammaegate that have been made (both coarse and fine) are used to create lightweight concrete. Five types of concrete are made. The first is concrete with fine natural agammaegate and coarse agammaegate from plastic agammaegate type Ⅰ/PA Ⅰ (rice husk ash agammaegate-PET), the second is concrete with both agammaegate fractions (fine and coarse) is PA Ⅰ, the third is concrete with fine natural agammaegate and coarse agammaegate from plastic agammaegate type Ⅱ/PA Ⅱ (cement-PET agammaegate), the fourth is concrete with both agammaegate fractions (fine and coarse) is PA Ⅱ. The fifth is reference concrete, namely concrete with natural agammaegate. To simplify the mention of concrete types, Table 2 shows the name for each concrete mixture, the artificial agammaegate's name, and the proportion of the synthetic agammaegate constituent. The test specimens are cylinders with a diameter of 10 cm and a height of 20 cm used for testing the compressive strength and tensile strength of the concrete; beam size 10 cm × 10 cm × 40 cm for flexural tensile testing. Provide different temperature treatments; various specimens are put in ovens with varying temperatures, namely 100 ℃, 200 ℃, 300 ℃, and 400 ℃ with a heating time of 2 h. One treatment as a reference is a concrete cylinder without heating. The oven equipment is shown in Figure 4b. Cylindrical specimens that do not undergo heating are assumed to receive a room temperature of 30 ℃ by the average temperature of one day at the research location. In Table 3, it can be seen the types of tests carried out, the kinds of temperature treatments, and the shape of the test specimens. The flexural tensile strength test is only carried out on samples without heating considering that the oven space capacity is insufficient to accommodate the specimen beam. Treatment of the concrete after casting is done by immersing all specimens in clean water for 14 d, then leaving the samples in open space until the samples are 28 d old and ready to be tested.

2.2.3.   Testing the mechanical characteristics of lightweight concrete

Several preparations were made before implementing the mechanical characteristic testing, including the application of sulfur capping to the cylindrical specimen and weighing the test object. Concrete compressive strength testing refers to Indonesian regulatory standards, SNI 03-1974-1990 ^[34]. The concrete compressive strength test set up is shown in Figure 5a. The split tensile strength test refers to the Indonesian national standard, SNI 03-2491-2002 ^[35], with the test set up shown in Figure 5b. The flexural tensile strength test is conducted as in Figure 5c.

3.   Results and discussion

3.1.   Stage I—production of type I (PA I) and type II (PA II) plastic aggregates

Two types of plastic agammaegate produced at this stage, namely type Ⅰ (PA Ⅰ) and type Ⅱ (PA Ⅱ), consisting of fine and coarse fractions, as shown in Figure 6. Figure 7 shows the maximum magnification of the surface or edge of each synthetic agammaegate using a Cooling Tech digital microscope with a magnication ratio of 40× to 1000×. The picture shows PA Ⅰ synthetic coarse agammaegate has a surface texture that is rougher and more porous than PA Ⅱ. The gradation of fine agammaegate of these two types of synthetic agammaegates is made the same as the gradation of natural sand used for reference concrete, namely zone 2 for fine agammaegate. Likewise, the synthetic coarse agammaegate size is made the same as the natural coarse agammaegate gradation, which passes through sieve No. 3/8" (with sieve hole size = 9.5 mm), stuck in sieve No. 4 (with sieve hole size = 4.76 mm). In Table 4, we can see the results of testing the characteristics of natural agammaegate and artificial agammaegate, where the addition of rice husk ash filler to PET plastic agammaegate with a ratio of 1:5 makes the absorption of synthetic agammaegate up to 2.43%, while agammaegate with cement filler, the agammaegate absorption only reaches 1.01%.

3.2.   Stage II—Casting concrete specimens

At this stage, a cylindrical specimen is produced with a diameter of 10 cm with a height of 20 cm and a concrete block measuring 10 cm × 10 cm × 40 cm. The consistency of fresh concrete for all types of concrete is carried out by adjusting the amount of water so that every concrete has a slump value that is not much different. All kinds of concrete have a slump value range from 45 to 6 mm. Table 5 shows the composition of the concrete mixture and the slump value for each type of concrete. The proportion of each concrete type follows the proportion of concrete by Wiswamitra et al. ^[11]. Figure 8 shows some of the test objects and the implementation of making test objects in the laboratory.

3.3.   Stage III—testing the mechanical characteristics of lightweight concrete

Complete concrete test results in concrete density and concrete mechanical characteristics after heating at various temperatures can be seen in Tables 6 and 7.

3.3.1.   The density, the compressive strength, and the cracks after heating

The results of the compressive test on the reference specimen (i.e., the sample with room temperature = 30 ℃) for all types of concrete and the results of the concrete density test show that all kinds of concrete with synthetic agammaegates produce concrete that falls within the criteria for structural lightweight concrete. ACI 213R-87 ^[36] requires that structural lightweight concrete is concrete that uses lightweight agammaegates with air-dry weight at 28 d ranging from 1440–1850 kg/m³ and with a compressive strength of more than 17.24 MPa. If the density and compressive strength of concrete are sorted from the largest to the smallest value, it will be seen in Figure 9. It appears that concrete with fine natural agammaegates (FN-CPRha and FN-CPCm) ranks first in terms of density and compressive strength, compared to concrete with all agammaegate fractions in the form of synthetic agammaegates (FC-PRha and FC-PCm). In the figure, it can be seen that the pattern of reduction in the compressive strength of concrete is sharper than the decrease in concrete density.

Meanwhile, concrete with PA Ⅱ produces a density and compressive strength of concrete that is higher than that of PA Ⅰ. Reference concrete, namely concrete with natural agammaegates, ranks first both in density and compressive strength of concrete. Figure 10 shows the value of concrete structural efficiency at the age of 28 d which is the quotient value of the compressive strength of the concrete with the value of the density of the concrete. The figure shows that concrete with fine natural agammaegate and coarse agammaegate from filler-PET (PA Ⅰ or PA Ⅱ) has a higher structural efficiency value than concrete using natural local agammaegates.

The test results on concrete after heating at various temperatures show that the compressive strength of the concrete decreases at higher heating temperatures. The decrease in the compressive strength of concrete with plastic agammaegates is sharper than that of reference concrete (FC-N), Figure 11. In more detail, it can be seen that the decrease in compressive strength can be seen from the heating temperature T = 100 ℃, both for reference concrete and concrete with artificial agammaegates. A significant reduction in compressive strength occurs in concrete with plastic agammaegates; this is due to the fine cracks caused by heating, which can be seen using a digital microscope, as shown in Figure 12. The crack width is measured from 0.004 to 0.011 mm. Cracks like these are not visible in reference concrete. Cracks in concrete due to heating can occur due to differences in the thermal expansion of the agammaegate and the cement matrix. The presence of heat causes the concrete to expand outward which creates internal tensile stresses which in turn cause internal cracks that extend to the surface ^[37]. Overheating will also cause chemical changes that cause microcracking so that the integrity and strength of the concrete decrease.

The reference concrete started to have fine cracks (seen with a digital microscope) when the heating temperature up to 200 ℃, while in concrete with synthetic agammaegates, the crack width became wider (visible to the naked eye) as shown in Figure 13 and the compressive strength of the concrete is decreasing. It can be seen that the cracks in concrete with PA Ⅱ (FN-CPCm and FC-PCm) are wider than concrete with PA Ⅰ (FN-CPRha and FC-PRha). At a heating temperature of 300 ℃, cracks in the concrete with plastic agammaegates were visible more and more clearly. The cracks in the cylindrical concrete were further clarified using a marker and photographed as shown in Figure 14. At heating temperature T = 400 ℃, the concrete specimen with plastic agammaegate began to burn and blackened, indicating that the plastic agammaegate from PET concrete had decomposed. It can also be seen in the post-heating concrete cylinder that the cracks in the concrete with PA Ⅱ are more numerous and wider than the concrete with the PA Ⅰ agammaegate. Figure 15 shows the fracture surface of the concrete cylinder in the specimen after heating at 400 ℃ which shows that the decomposition and evaporation process of PET plastic agammaegates due to heat that has exceeded its melting point occurs in the area around the perimeter of the concrete cylinder to a depth of ±2 cm. Figure 16 is the SEM photo results on the FN-CPRha specimens after heating at each temperature. The photo is focused only on the plastic agammaegate section, in this case, the type Ⅰ/PA Ⅰ plastic agammaegate (rice husk ash-PET agammaegate). In the series of photos, it can be seen that the PA Ⅰ agammaegate after heating at a temperature of 100 ℃ (Figure 16a), 200 ℃ (Figure 16b), 300 ℃ (Figure 16c), and 400 ℃ (Figure 16d), which shows that the higher the heating temperature, the plastic agammaegate shows various levels of deterioration, namely cracks, the rougher the surface is at the increasingly high, even decomposition occurs, especially in the agammaegate after heating at 400 ℃ which is indicated by the rough and wavy surface of the agammaegate.

3.3.2.   The splitting tensile strength

Data from the split tensile strength test for all types of concrete can be seen in Table 7 and illustrated in Figure 17. As in the compressive strength results, the split tensile strength shows a decreasing pattern with the same sequence of concrete types arranged from the largest to the smallest value. The more plastic content in the concrete (Figure 17a) and the higher the heating temperature (Figure 17b) the lower the tensile capacity of the concrete.

3.3.3.   The flexural strength

The flexural strength test was carried out using the third point loading method (Figure 5c). Figure 18 illustrates the flexural tensile strength value for each concrete, which shows a decreasing pattern, such as compressive strength and split tensile strength.

Figure 19 shows the fracture surface of the beam specimen after flexural tensile test. The type of FC-N reference concrete has the roughest fracture surface, followed by concrete using natural sand (FN-CPRha and FN-CPCm), and the smoothest is concrete with all synthetic agammaegates (FC-PRha and FC-PCm).

4.   Conclusions

After carrying out research on the heat resistance of lightweight concrete with plastic agammaegate from PET (Polyethylene Terephthalate)-mineral filler, several conclusions can be drawn:

●  Plastic agammaegate type Ⅰ (PA Ⅰ) produce a roughen and more porous surface of the agammaegate, has greater water absorption, has a smaller density and specific gravity than type Ⅱ plastic agammaegate (PA Ⅱ).

●  The use of PA Ⅰ causes the density and mechanical characteristics of concrete to be smaller than that of concrete using PA Ⅱ.

●  The use of PA Ⅰ and PA Ⅱ both as a substitute for coarse agammaegate fraction only or as a substitute for all-natural agammaegate fractions is capable of producing structural lightweight concrete.

●  The use of PA Ⅰ and PA Ⅱ as coarse agammaegates (while fine agammaegate uses natural sand) can produce greater structural efficiency of concrete compared to concrete with whole natural agammaegates (reference concrete).

●  Cracks in concrete due to heating will be less in number and with a smaller crack width when using PA Ⅰ than PA Ⅱ.

●  Decrease in the mechanical characteristics of concrete with plastic agammaegate and reference concrete due to heating, starting from the first temperature of treatment, which is 100 ℃.

●  The decrease in the compressive strength and tensile strength of concrete with plastic agammaegates due to heating looks more drastic than that of reference concrete.

●  The mechanical characteristics and density of concrete with plastic agammaegates will be smaller than that of reference concrete.

Acknowledgments

This research was made possible thanks to a grant from the Directorate General of Research and Development Strengthening; Ministry of Research, Technology and Higher Education of the Republic of Indonesia through the Doctoral Dissertation Research scheme.

Conflict of interest

All authors declare no conflicts of interest in this paper.