The concept of recycling old construction debris and materials and then using them as aggregate in new construction projects has been practised for decades. It was firstly known in the Romans time when they reused the stones and rubbles from old roads in building new ones. The strategy of recycling the construction waste (CW) and demolition materials (DM) was established in Europe just after the end of the Second World War in the late 1940 when recycled materials from demolished buildings were initially used as a base course for new roads and pavement applications (Olorusongo, 1999).
Economic aspects and environmental issues such as waste management and demolition works are the main advantages of introducing the concept of recycling the waste materials from construction and demolition projects and then using these recycled materials as aggregates in replacing natural coarse aggregate (NCA) and fine aggregate (NFA) in the production of new concrete for construction industry. In recent decades, the recycling industry of the CW has been increasing popularity and many governments have supported the research programmes that aim to improve the quality of the recycled materials and help to protect the environment by reducing the disposal of such wastes. Another reason for supporting this industry is the fact that the sources of high grade natural aggregate (NA) are depleting and the increase of the required quantities of aggregate for construction projects which make it important to find alternative sources of aggregate (Abdulrahman, 2009).
According to the European Demolition Association, 200 million tons of wastes are produced per annum in Europe with a 30% of these wastes can be recycled whereas in North America the recycling rate reaches up to 45% of the total annual waste (Buck, 1970).
In many countries, the percentages of CW and DM materials are generally considered to be high with a variation in the recycling rate from country to country. In the United Kingdom, for example, more than 50% of the CW and about 40% of the DM can be recycled while in the United State of America, the percentages of the materials that can be recycled are about half of those in the UK. Another country that depends on recycling waste materials as a source of aggregate is Spain where the rate of recycling reaches more than 70% for both CW and DM. Germany and Japan also have a good recycling rate since the percentages of the recycled materials are 60% and 65% respectively (Hendricks & Pietersen, 2000; Poon, 2000).
In order to improve recycling rate and the quality of the recycled materials to be suitable for high grade applications, many countries have established guidelines and technical specifications for promoting the use of the recyclable materials. Many researches have been taken place on several types of materials that can be recycled and then used in the production of high quality aggregate. It was found that various types of materials can be sorted and then recycled to be used in replacing the virgin aggregate in concrete such as timber, electricity fitting, steel, tubes, glass, florescent, brick, cables, and tiles. The main problem of using such materials is that the produced concrete is rarely used for high grade applications and can only be used for low grade projects such as pavement and base course of roads (Tam, et. al., 2007).
Literature review
Introduction
Influence of using recycled coarse aggregate on concrete properties
Most of the researches that have been taken place in the field of using recycled materials as aggregate in concrete have studied using these wastes as coarse aggregate (CA) rather than using them as fine aggregate (NFA). This might be due to the fact that CA quantities required in concrete for most construction projects are always higher than the quantities of the required NFA. Most of the studies on using RA in concrete have focused on the influence of the mechanical and geometrical properties of aggregate on strength properties of recycled aggregate concrete (RAC) mainly compressive strength. Some researchers studied using different percentages of recycled aggregate (RA) in concrete and then compare the results with normal concrete (NC) to specify the optimum percentage that results in best concrete properties whereas others preferred to study the concrete performance with using different types of waste materials as aggregate and also RA properties that might have the major influence on RAC properties.
Concrete properties with different percentages of RA
Changing the percentages of RA in concrete can significantly affect its general and mechanical properties as it was intensively studied by many researches who aimed to investigate the variation in concrete performance with using different percentages of RA.
Effect on mechanical and general properties
Olorunsogo and Padayachee (2002) carried out a study to investigate the influence of changing the percentages of RA replacing NA in concrete on its durability. Three different concrete mixes were made with 0%, 50% and 100% of RA and the results were compared at the ages 3, 7, 28 and 56 days. They concluded that the higher percentage of RA in concrete, the lesser RAC durability can be obtained and this can be attributed to the weakness of RA which contains cracks and fissures within the samples. This however, may lead to higher bond stress between RA and cement mortar in concrete than that made with NA.
A study by Gomez-Soberon was carried out 2002 to examine the influence of increasing the amount of RA in concrete on its porosity. It was revealed that the use of RA in concrete production leads to an increase in the porosity within RAC and this increase becomes higher with raising the percentages of RA replacing NA in concrete. This was accompanied by dropping in the mechanical properties of the produced concrete.
Tabash and Abdelfatah (2008) cited Frondistou-Yannas in 1977 who carried out a study aimed to compare compressive strength and modulus of elasticity between NC and RAC. The experiments indicated that if RAC is hardened in gravel at the expense of mortar, its compressive strength and modulus of elasticity will be at least 76% and 60% respectively of those of the conventional concrete.
Many studies investigated the optimum percentage of RA that can be used in replacing NA in concrete in order to achieve the best possible results. The experiments also aimed to compare the properties of RAC made with four different percentages of RA (0%, 25%, 50% and 100%). All the concrete mixes were designed to yield the same strength properties in order to compare the quantities of water required for each concrete mix to achieve the necessary strength characteristics. They illustrated that for a constant quantity of water for all the concrete mixes, increasing the percentage of RA in concrete will lead to a marked reduction in its compressive strength to reach its lowest value in concrete made with 100% of RA. The study suggested that in order to improve the compressive strength of RAC to be very close to that of conventional concrete, the quantities of water required for concrete should reduced with increasing the amount of RA (Etxeberria et al. 2007; Katz 2003).
In contrast, reducing the amount of water negatively affects the workability of RAC. Therefore, several studies suggested the use of RA in partially saturated condition in concrete to govern its fresh and hard properties. Furthermore, the high absorptivity of RA and also its coarser and more angular surfaces represent the major factors that lead to poorer workability in RAC. This means that the higher percentages of RA in concrete, the poorer workability can be achieved with the amount of water kept constant (Etxeberria et al. 2007, Tavacoli & Soroushian 1996).
Etxeberria et al. (2007) also cited Salem & Burdette (1998) who indicated that the higher percentages of RA replacing NA in concrete, the more quantities of cement are required in concrete in order to obtain the same workability and compressive strength as those of NC. They also suggested that in order to keep the workability and the compressive strength of RAC within the acceptable limits, the percentages of RA in concrete should be in the limit of 30%.
It was also figured out in several researches that the higher percentage of RA in concrete, the lesser compressive strength is resulted. There was a gradual reduction in compressive strength as the amount of RA increased but this reduction did not exceed the limit of 40%. Moreover, as it was mentioned previously, the best results for the compressive strength of RAC were obtained with a percentage of RA not more than 30% (De Oliveira & Vazquez 1996; Chen et al. 2003; Dhir et al. 1999).
Compressive strength with 100% RA
For full replacement of NA with water- cement ratio kept the same, the 28- days compressive strength of RAC was investigated by many researchers who revealed that the compressive strength of RAC in this case is approximately 20- 25% less than that of the NC with higher bond stress for RAC due to the coarser and more angular surfaces of the RA (Etxeberria et al. 2007, Tavacoli & Soroushian 1996).
Etxeberria et al. (2007) also found out that the concrete made with RA replacing the whole amount of the required NA (100% replacement) can only have a compressive strength higher than that of NC if the water- cement ratio is reduced to be lower than that used in the NC.
Effect of recycled materials properties on RAC
Since many studies have proved that concrete properties such as workability and compressive strength can be significantly affected by RA characteristics especially water absorption and resistance to fragmentation, most of the new researches were proposed to improve the quality of these materials in order to produce high quality concrete. Some of the RA characteristics which were investigated are particle shape and size distribution, density, absorptivity, porosity and carbonation depth. The relationship between such characteristics and the properties of RAC such as workability, compressive and tensile strength were also studied.
Assessing the correlation among RA properties and RAC
Tam et al. (2007) used regression analysis to test the correlation among the characteristics of recycled demolished concrete collected from different sites, their RA properties and the compressive strength of the produced RAC. The demolished concrete samples were firstly examined by testing and formulating the relationships among some of their characteristics such as density, absorptivity, porosity and carbonation depth. After treating the concrete debris samples, the properties of the resultant RA specimens were tested and the correlation between some of them was then analysed and formulated such as particle size distribution, density, porosity, absorption and particles shape as well as strength and toughness. Finally, the compressive strength of RAC was investigated for all samples.
They found that there is a strong relationship among concrete debris, RA samples and RAC made with these samples. The results showed that the quality of demolished concrete can significantly affect the properties of their RA and also the compressive strength of RAC. The analyses of the results also showed that the concrete performance can be predicted from the characteristics of the demolition and also from RA properties. From the results, it can be clearly seen that in order to produce high grade concrete, RA samples should be tested and improved in earlier stages before using them in concrete production and this can also help to save time and money.
Physical and mechanical properties of RA derived from laboratory concrete
Many other researches have been taken place to investigate the influence of parent concrete characteristics on the performance of RAC. One of these studies was done by Padmini et al. In 2009 to discuss the physical and mechanical properties of recycled materials extracted from virgin concretes made in laboratory with different compressive strengths and different maximum sizes of aggregate. The effects of these characteristics on the properties of RAC were then studied and formulated. Nine different concretes were designed using three maximum sizes of NA with three different compressive strength targets for each size of aggregate.
After reaching the 28 days strength targets, the parent concretes were crushed using a Jaw-crusher to produce nine different samples of RA and the properties of these samples were tested before using them in the production of RAC. The same process which was used in producing the parent concrete was also used to produce RAC to eliminate any effect that might result from the difference in the production method or test conditions. Therefore, RAC were designed to give the same workability and strength properties as in the virgin ones (Padmini et al. 2009).
For the mechanical properties of the RA samples, the results showed that the smaller maximum size of NA used in concrete, the lesser RA resistance against crushing, impact and abrasion are obtained. This can be attributed to the fact that crushed aggregate is generally coated by cement mortar which leads to less bulk density and specific gravity. The cement mortar also leads to more fine material when the aggregate samples are subjected to mechanical actions and this means that the values that can be obtained from crushing, impact and abrasion tests will be lesser than those of NA. The study also illustrated that due to the attached mortar, the water absorption of RA is higher than that of the virgin one and its value decreases with increasing the size of aggregate used in concrete. The decrease in the water absorption is due to the smaller surface area of aggregate that is available to adhere the cement mortar of parent concrete (Padmini et al. 2009).
Influence of water- cement ratio and size of aggregate
Padmini et al. (2009) also carried out another study aiming to analyse the correlation between water- cement ratio, aggregate- cement ratio and cement content and then formulate them in terms of the compressive strength of RAC and compare the results with those of NC. The three maximum sizes of aggregate which were used to produce the parent concretes are 10mm, 20mm and 40mm. The parent concretes were designed for three different targets of compressive strength with three different workability values for each one of them (25mm, 50mm and 150mm).
The experiments pointed out that in order to maintain a particular workability, RAC requires higher water- cement ratio when compared to NC to overcome the problem of the high absorptivity of RA. This however, leads to a reduction in the compressive strength of RAC which requires higher cement content to result in a compressive strength value close to that of parent concrete. The difference in the compressive strength was higher with increasing the required target of strength.
Padmini et al. (2009) concluded that the higher compressive strength of parent concrete, the higher water absorption for the RA samples derived from this concrete. However, the smaller maximum size of aggregate used in parent concrete, the higher water absorption of RA is obtained. From the last conclusion, it can be clearly seen that in order to overcome the problem of the high water absorption of RA, it is generally better to use higher maximum size of aggregate which also leads to an increase in the compressive strength of the produced RAC.
Effect of recycled materials source on RAC properties
A study was done by Tabash and Abdelfatah in 2008 to investigate the effects of using RA extracted from different sources on the mechanical properties of the produced concrete such as compressive strength and splitting tensile strength. In other words, the experiments were proposed to study the influence of the origin of recycled materials on the properties of RAC mainly their effects on compressive and tensile strength. Two different sources were used to produce tow samples of RA. One of these sources was a laboratory concrete had an identified compressive strength whereas the other sample was extracted from a dump site. Initially investigations into the RA properties were carried out including abrasion and soundness tests. The RA samples were then used in the production of new concretes with different targets of compressive strength. The results were compared with the compressive strength of concretes made with NA to achieve different compressive strengths.
The 28 days designed compressive strengths of the cylindrical concrete specimens of the two concretes were 30Mpa and 50Mpa respectively. It should be noticed that all the concretes were designed for a slump of 100mm. The results illustrated that in order to maintain the same workability, the quantity of water needed in the concrete made with RA should be 10% more than that required for NC.
The tests on RA indicated that the resistance of the RA samples is 30% less than that of the NA and this can be attributed to the fact that NA has more density and strength than those of RA. The experiments also revealed that for the abrasion test results, the higher strength of the demolished concrete that the RA samples were derived from, the lesser losses in RA were occurred and therefore the new concrete will have higher grade strength. It was also figured out that the compressive strength of the concrete made with RA extracted from strong demolished concrete is approximately close to that of NA (Tabash and Abdelfatah 2008).
However, in order to obtain these results, the quantity of water required in RAC should be reduced and this will result in poorer workability. Moreover, if the quantity of water is kept constant, the compressive strength of RAC will be 30% less than that of the conventional concrete according to these tests. The results also showed a reduction of 30% in the splitting tensile strength and this percentage was lower in the concrete produced by using RA extracted from stronger concrete debris. They concluded that in spite of the reduction in the concrete strength characteristics by about 10 - 25%, the values were within the acceptable limits (Tabash and Abdelfatah 2008).
Effect of the recycling process on RAC properties
Sago-Crentsil and Brown (2001) found out that the processing of recycling the waste materials can influence the mechanical properties of the resultant concrete. The results illustrated that RAC workability decreases with using RA samples produced in the laboratory which have coarser and more angular surfaces than those produced commercially. In contrast, this provided a higher bond stress between the aggregate and the concrete ingredients. The same study also indicated some differences in compressive and tensile strength between NA and RAC with better results obtained from NA as they were shown in other researches. There was also approximately a 12% reduction in the abrasion resistance of RA with an increase in the values of drying shrinkage.
Influence of using recycled fine aggregate (RFA) on concrete properties
In spite of the large number of previous researches that have been taken place to investigate and improve the quality of recycled waste materials in order to use them as aggregate in concrete, most of these researches focused on studying the use of demolition wastes as CA in concrete and a small number of studies have investigated using such materials as NFA. The rareness of the studies on RFA can be attributed to its high water absorption which can significantly affect the final concrete properties (Kasai 1994).
Replacement amount of RFA
A number of studies that investigated the effect of increasing the amount of RFA in concrete have concluded that as the percentage of RFA in concrete increases, its compressive strength gradually decreases and the reduction in all cases was less than 40%. It was also revealed that RAC made with an amount of RFA up to 20% has approximately the same compressive strength value as that of the conventional concrete (Khatib 2004; Dhir et al. 1999).
Khatib (2005) carried out a study to evaluate the effects of using different percentages of RFA in concrete on its compressive strength. A number of different concrete mixes were made with 0%, 25%, 50% and 100% of RFA with water -cement ratio kept the same for all of them so that the effect on compressive strength will only be due to the changes in the amount of RFA. The compressive strength of the different concretes was examined at ages 1, 7, 28 and 90 days of curing. It was found that increasing the percentages of RFA in concrete leads to a reduction of 15- 30% in compressive strength. However, the results showed that concrete made with a percentage of RFA up to 50% has long term compressive strength similar to that of NC. There was also a reduction of 25- 30% in the long term compressive strength of concrete made with 100% of RFA.
The study also investigated the effect of changing the percentages of replacement on concrete density. It was illustrated that the higher percentage of RFA replacing virgin fine aggregate in concrete, the lesser concrete density can be achieved. In contrast, during the curing period, the results showed a little increase in the RAC density (Khatib 2005).
The workability of RAC was also investigated for concrete made with different percentage of RFA. It was shown that for water- cement ratio of 0.5, RAC has very good workability for all percentages where a slump of 170- 190mm was achieved in the experiments. It was also noticed that the concrete workability was slightly affected by increasing the amount of RFA in concrete (Khatib 2005).
Influence of RFA sources
Many studies indicated that the sources of RFA can significantly affect the compressive strength of RAC. One of these studies was proposed to compare the compressive strength of concrete made with RFA samples extracted from two different sources. One of these samples was derived from crushed concrete whereas crushed brick was the source of the other sample. These samples were placed in concrete by different percentages (0%, 25%, 50% and 100%) with water- cement ratio kept constant for all concrete mixes. The results showed that for both samples, there was a gradual decrease in the compressive strength of RAC. The concrete made with a percentage of crushed concrete up to 25% exhibited a reduction of about 15% in compressive strength whereas for full replacement of RFA, the reduction reached 27%. However, there was only 4% reduction in the compressive strength of concrete made with 25%- 50% of crushed brick and the reduction was less than 10% at 100% of replacement. It was concluded that concrete made with crushed brick has higher compressive strength than that made with crushed concrete (Khatib 2005).
Ajdukiewicz and kliszczewicz (2002) investigated the effects of using RFA extracted from known compressive strength concrete on the mechanical properties of high grade concretes. The samples were produced from demolished concretes which have compressive strength between 40 â€" 70Mpa. The results showed a 10% reduction in the compressive strength of the concrete made with RFA with a reduction of 8 â€" 20% in the bond stress.
Evangelista and Brito (2007) evaluated the influence of using RFA extracted from laboratory concrete on the mechanical properties of concrete made with this aggregate such as compressive strength and splitting tensile strength. The results were then compared to that of the laboratory concrete. Natural concrete was firstly made with water- cement ratio of 0.52 and after 35 days of curing, this concrete was tested and then crushed using a jaw crusher to produce the fine aggregate samples required to produce RAC. The results showed a slight reduction in both compressive and tensile strength of concrete made with a percentage of recycled fine aggregate up to 30%. Beyond this percentage, the reduction was higher to reach up to 30% compared to the compressive strength of NA.
Influence of recycled material type
The effects of the type of recycled materials that can be used as fine aggregate (FA) on concrete properties such as workability and compressive strength have been investigated by several studies. Bai and Basheer (2003) carried out a study to test the compressive strength and workability of concrete incorporating furnace bottom ash (FBA) as RFA. Fine aggregate was replaced in concrete by 0%, 30%, 50%, 70%, and 100% of FBA with two different water- cement ratios (0.45 and 0.55). The results of the slump tests illustrated that in spite of the higher water absorption of FBA, there was an increase in RAC workability with increasing the percentage of replacement.
For RAC compressive strength, the results showed a clear decrease in the 28 days compressive strength as the percentage of FBA increased in concrete whereas for the long term improvement in compressive strength, the results were close to those of NA especially when concrete specimens were tested after 365 days of curing.
Another study also investigated the use of the same materials (FBA) in replacing NFA in concrete. The study suggested that the percentage of FBA that can be used in concrete should be less than 30% since the experiments showed that up to this percentage, the compressive strength of RAC was close to that of concrete made with NFA. Beyond this percentage, the results were not within the acceptable limits (Bai et al 2005).
In contrast, many other studies showed that the use of such materials as NFA in concrete production leads to light weight concrete with less density and high porosity and this type of concrete cannot be used for high grade construction projects (Collins and Sanjayan 1999, Kohn et al. 1999).
Influence of replacement amount with different RFA sources
Kou and Poon (2009) proposed a study to compare the compressive strength and workability of concrete made with three different samples of fine aggregate. The experiments were divided into three stages, in the first one, NC was made using river sand while in the second and third stages, two RACs were made by replacing the river sand by 0%, 25%, 50%, 75% and 100% of crushed rock and furnace bottom ash (FBA). For the three stages, the concrete mixes were designed using two ways. Firstly they were designed by keeping water- cement ratio at constant value of 0.53. Secondly, the concretes were designed to achieve the same workability target for a slump value of 60- 80mm. The compressive strength results were compared at the ages of 3, 7, 28 and 90 days.
The results of slump tests showed that for water- cement ratio kept constant, RAC made with FBA had a very good workability better than that of NC. This can be attributed to the absorptivity of FBA which results in more free water that becomes available to improve the workability of the fresh concrete. Also, FBA has more fine particles and smoother surfaces than those of the river sand which positively affect the RAC workability.
In contrast, the results indicated that the higher percentage of crushed stone used in concrete, the poorer workability can be achieved. The poor workability is due to angular particles shape and coarser surfaces of crushed stone when compared to the other two sources. Furthermore, RAC made with crushed stone required more water or higher water- cement ratio in order to achieve the same workability as those of NC and RAC made with FBA (Kou and Poon 2009).
In terms of compressive strength, there was a marked reduction in concrete made with FBA and the value of compressive strength continued to decrease with raising the percentage of replacement. This reduction is due to the smoother surfaces of the FBA particles which lead to weak pond between these particles and the cement mortar. In addition, for concrete made with crushed stone, the compressive strength was within the acceptable values up to 75% of replacement where it started to decrease rapidly to reach its lowest value at full replacement of RFA. Moreover, for a fixed water- cement ratio, concrete made with 100% of crushed stone had higher compressive strength value than that made with 100% of FBA and all the values were lower than those of NC (Kou and Poon 2009).
Similar results were also reported by many other researchers who found out that in order to keep the compressive strength of RAC within the acceptable limits, the cement content should be increased with the increase in the replacement level which means lower water- cement ratio than that of NC (Donza et al. 2002, Kim et al. 1997, Kou and Poon 2009).
With regard to the second part of the previous study where all the concrete mixes were designed to result in the same workability range, the results showed an increase in the compressive strength of RAC made with FBA for all replacement levels to be higher than that of NC which made with river sand. This can be attributed to the high water absorption of FBA which leads to decrease the water- cement ratio in concrete. Furthermore, for concrete made with crushed stone, the results were similar to those of the first part of the experiments when the water- cement ratio was kept constant (Kou and Poon 2009).
