IntroductionConstruction and demolition (C&D) wastes consist essentially of the debris resulting from activities ranging from excavation and site preparation to the construction, maintenance, rehabilitation, and demolition of buildings and other civil infrastructures. Vast amounts of C&D wastes are produced every year by the construction industry in all developed and developing countries alike, which has raised significant environmental concerns at an international level. In Europe, the vast quantities of C&D wastes generated annually have led the European Commission to identify these materials as a priority waste stream for reuse and recycling. In fact, when measured in volume, C&D waste is the largest waste stream in the European Union, accounting for about 1/3 of all waste generated (European Commission 2018). In this context, C&D waste reuse and recycling is of global importance, since it can contribute not only to more efficient waste management in the construction sector, reducing the waste disposal volumes to landfill, but also to attenuate the environmental impacts associated with the overexploitation of natural resources (Arulrajah et al. 2013; Vieira and Pereira 2015b; Arulrajah et al. 2017, 2020).The use of recycled C&D aggregates as an alternative backfill in the construction of geosynthetic-reinforced soil structures, such as embankments, slopes, and retaining walls, has recently been pointed out as a potential application that can offer significant project cost savings (i.e., by avoiding the use of more expensive conventional backfill materials), apart from contributing to sustainability and environmental protection (Santos et al. 2013, 2014; Vieira et al. 2016, 2020a, b).Geosynthetics, such as geogrids and geotextiles, are polymeric materials that have been extensively used for soil reinforcement and stabilization over the last four decades (Greenway et al. 1999; Allen and Bathurst 2002; Allen et al. 2002; Benjamim et al. 2007; Palmeira 2009; Ferreira et al. 2013; Rahman et al. 2014; Ferreira et al. 2015, 2016a, 2020a; Maghool et al. 2020; Karnam Prabhakara et al. 2021a, b). When properly designed and installed, geosynthetics provide a cost-effective alternative to more traditional solutions (Rowe and Jones 2000; Indraratna et al. 2010; Christopher 2014; Ferreira et al. 2016b, c; Nimbalkar and Indraratna 2016; Ferreira et al. 2020b). However, the viscoelastic nature of polymeric geosynthetics and their susceptibility to creep may influence the performance of the reinforced structures where they are installed, which is why the assessment of the long-term mechanical properties of geosynthetics is of primary interest when these materials are used in permanent reinforcement applications (Lopes et al. 1994; Leshchinsky et al. 1997; Li and Rowe 2001; Kongkitkul et al. 2007; Bathurst et al. 2012; Miyata et al. 2014; Ferreira et al. 2021). If the durability and long-term behavior of geosynthetics are important factors when conventional backfill materials are used, their assessment becomes even more relevant when alternative (i.e., nonconventional) materials, such as recycled wastes, are considered.The long-term tensile behavior of geosynthetics is often evaluated through creep and creep rupture testing (Allen and Bathurst 1996; Zornberg et al. 2004; Bueno et al. 2005; Miyata et al. 2014; Pinho-Lopes et al. 2018; Deng and Huangfu 2021). As per the International Standard ISO 13431:1999 (ISO 1999), the tensile creep behavior can be assessed by loading geosynthetic specimens with a constant static force (defined as a percentage of the maximum tensile strength), under constant conditions of temperature and humidity for a period of 1,000 h, during which the elongation of the specimens is recorded. To analyze the creep rupture behavior, the specimens are also loaded with a constant static force, which is maintained until the specimen ruptures, and the time to rupture is determined. The latter method has typically been used to characterize the creep rupture envelopes, which define the time to rupture of a given geosynthetic when subjected to a constant (sustained) load. The creep rupture envelope is then used to estimate the creep reduction factor, which is a key parameter in the design of geosynthetic-reinforced structures to account for the effects of sustained load on the long-term tensile strength of the reinforcement.Another factor that is usually considered when assessing the design value of the long-term tensile strength of a geosynthetic is the strength loss caused by the chemical and biological degradation of the polymers. The chemical and biological degradation of geosynthetics is particularly relevant when these materials are applied in landfill barrier systems (Rowe and Sangam 2002).The principal cause of chemical degradation of geosynthetics made of polyethylene and polypropylene is oxidation. The oxidation degradation of geosynthetics in field conditions depends on the chemical constituents of the surrounding media, the available oxygen concentration, and temperature. As pointed out by Hsuan et al. (2008), the presence of transition metal ions (e.g., cobalt, manganese, iron, and copper) can accelerate the oxidation of polyolefins (i.e., polyethylene and polypropylene), which are the most widely used polymers in geosynthetics. However, since geosynthetics made from polyolefins contain antioxidant stabilizers, it is anticipated that the oxidation of the polymer does not begin until nearly all of the antioxidants have been consumed (Hsuan et al. 2008).The hydrolytic susceptibility of polyester (PET) geosynthetic products has also been the subject of significant research. In fact, the main cause of chemical degradation of PET geosynthetics is by hydrolysis. A maximum pH value of 9 is recommended for PET geosynthetics used in critical applications, whereas a lower limit of pH=3 is considered for both polyolefin and PET geosynthetic products (FHWA 2010).The biological degradation, which arises from the polymer being attacked by microorganisms, has also been considered as a matter of concern. However, according to Koerner et al. (2017), unless biologically sensitive additives, such as low-molecular-weight plasticizers are included in the polymer formulation, geosynthetic resins are insensitive to bacteria and fungi.Geosynthetics have been exhumed after decades of service with marginal reductions on their properties (Koerner et al. 2017). Nevertheless, understanding the potential effects of the use of alternative backfill materials, such as recycled C&D aggregates on the short- and long-term strength properties of geosynthetics is of great interest.In this study, the long-term tensile response of a high-strength geotextile typically used for soil reinforcement is investigated by means of a series of creep rupture tests carried out on fresh specimens as well as specimens that were previously exposed to recycled C&D aggregates and a natural soil. To this end, two small trial embankments were constructed (using a recycled C&D material and a clayey sand as backfill), whereby the embedded geotextile specimens were subjected to real environmental conditions for a period of 24 months. The creep strain and rupture behavior of fresh geotextile specimens was evaluated and compared with that of the exhumed specimens in order to assess the potential chemical and environmental degradation induced by the exposure to the aforementioned backfill materials. Based on the obtained results, the creep reduction factors were then estimated and compared with those generally reported in the literature.Materials and MethodsMaterialsThe geosynthetic used in the current study was a uniaxial high-strength composite geotextile (also termed as a geocomposite reinforcement) with a mass per unit area of 340 g/m2. This material consists of high modulus PET yarns attached to a continuous filament nonwoven polypropylene (PP) geotextile backing. The nominal tensile strength (machine direction) and corresponding elongation as per the manufacturer specifications are 75 kN/m and 10%, respectively.As mentioned, two trial embankments were constructed using different backfill materials: a recycled C&D aggregate and a clayey sand. The recycled C&D aggregate is a fine-grained material obtained from the processing of C&D wastes coming primarily from the maintenance and demolition of residential buildings. The recycling process was carried out at a specialized recycling plant and involved an initial sorting phase, in which contaminants such as wood, steel, and plastic were removed, followed by crushing and grain size separation. Table 1 indicates the relative proportion of constituents obtained according to the European Standard EN 933-11:2009 (CEN 2009) by manual sorting of particles larger than 4 mm. As given in Table 1, this recycled C&D material is mainly composed of concrete and mortar products, unbound aggregates, and soil. The particle size distribution curves of the recycled C&D material and the soil (clayey sand) used in this study are presented in Fig. 1. It may be argued that this sand is not the ideal reference soil, since its gradation is different from that of the C&D material and comprises a significantly higher amount of fines, but it was not possible to find a more suitable soil at the time this study was conducted. Nevertheless, the main purpose of using a reference material (i.e., to understand whether the recycled C&D aggregates are susceptible to induce any additional degradation when compared to that caused by a natural soil) can still be achieved.Table 1. Constituents of the recycled C&D materialTable 1. Constituents of the recycled C&D materialConstituentsValueConcrete, concrete products, mortar, concrete masonry units, Rc (%)36.8Unbound aggregate, natural stone, hydraulically bound aggregate, Ru (%)33.7Clay masonry units, calcium silicate masonry units, aerated nonfloating concrete, Rb (%)10.8Bituminous materials, Ra (%)0.5Glass, Rg (%)1.0Soils, Rs (%)17.1Other materials, X (%)0.1Floating particles, FL (cm3/kg)7.80The analysis of the leaching behavior of recycled C&D materials to be used in geotechnical applications is of critical importance as it enables evaluating the potential for groundwater contamination and the presence of chemical substances that could induce the degradation of geosynthetics. Hence, laboratory leaching tests were performed on the recycled C&D material used in the construction of the trial embankment, as well as on C&D material collected 24 months after the embankment construction (i.e., subjected to weather conditions), following the procedure described in the European Standard EN 12457-4:2002 (CEN 2002). The concentration of each contaminant was then compared with the acceptance criteria for inert landfill as per the European Council Decision 2003/33/EC (European Commission 2003).It was found that only the concentration of sulfate for the recycled C&D material in its initial state (2,100 mg/kg of dry matter) exceeded the maximum value established by the European legislation for inert landfill (1,000 mg/kg of dry matter), whereas all the other pollutants were significantly below the threshold values. The relatively high concentration of sulfate in recycled aggregates is often associated with the presence of gypsum drywall, a typical component of mixed recycled aggregates (Jang and Townsend 2001). It should be noted that according to the European legislation (European Commission 2003), if the waste material does not comply with the sulfate threshold value, it can still be considered as meeting the acceptance criteria provided that the leaching does not exceed 6,000 mg/kg at a liquid to solid (L/S) ratio of 10 l/kg. However, the recycled C&D material collected 24 months after the embankment construction exhibited a significantly lower concentration of sulfate (630 mg/kg of dry matter), which may be attributed to the reduction in leached concentration over time as a result of rainwater. Further details about the leaching tests and associated results can be found in a previous publication (Vieira and Pereira 2021).Trial EmbankmentsTo examine the potential chemical and environmental degradation of the geosynthetic due to a 24-month exposure to the C&D material, a small trial embankment was constructed using recycled C&D material as backfill. An additional similar embankment was also constructed using a clayey soil for comparison purposes. Both embankments were 3.0 m long, 2.0 m wide, and 0.45 m high.The geosynthetic specimens were placed within the embankments at two different levels vertically spaced at 0.20 m. After cleaning the foundation from existing vegetation, a 0.05-m-thick layer of backfill material was placed and compacted [Fig. 2(a)] and some geosynthetic specimens were installed without overlapping. These specimens were covered with a first layer of backfill placed manually to prevent mechanical damage [Fig. 2(b)]. The remaining volume of backfill material was evenly spread and compacted to reach a 0.20-m-thick layer. Additional geosynthetic specimens were positioned [Fig. 2(c)] and the final 0.20-m-thick backfill layer was then placed and compacted. Finally, the lateral slopes of the embankments were compacted and covered with coarse recycled C&D aggregates to prevent erosion by rain water [Fig. 2(d)]. It is worth mentioning that the compaction was performed to facilitate the embankment construction and prevent wind and rain erosion, and not to simulate the compaction procedures taking place in a real-life embankment. Therefore, a lightweight compaction process was adopted to minimize geosynthetic damage during installation (i.e., mechanical damage), whose effects were outside the scope of this study.The exhumation of geotextile specimens after a 24-month exposure period was performed with special care to avoid any damage [Fig. 3(a)]. After removing the coarse aggregates placed on the lateral slopes along with the existing vegetation, the backfill materials were manually removed. The visual inspection of the geotextile specimens showed that some of them had been crossed by plant roots up to a few millimeters in diameter, regardless of the backfill type [Fig. 3(b)]. Exhumed geosynthetic specimens were then stored in the laboratory at a constant temperature of 20°C before testing.Creep Rupture TestsThe long-term tensile behavior of the geotextile was characterized by conventional creep rupture tests carried out in accordance with the recommendations in the International Standard ISO 13431:1999 (ISO 1999) using specimens of a technically representative width. A technically representative width is defined by ISO 13431:1999 (ISO 1999) as a small width (i.e., less than 200 mm), which exhibits tensile strength/strain properties per unit width under identical test conditions, within ±5% for tensile strength and ±20% for elongation at maximum load, of the values obtained according to ISO 10319. Because of limitations associated with the maximum axial load that can be applied in the creep rupture tests, it was not possible to use 200-mm-wide specimens in these tests (the required axial loads would exceed the maximum capacity of the test system). Therefore, tensile tests were previously performed on narrower geotextile specimens (100 mm), which confirmed the validity of using specimens of a technically representative width of 100 mm for the creep rupture tests.The geotextile specimens (100 mm wide and 300 mm long) were loaded in the machine direction with a constant static force, under constant conditions of temperature (20°C), until the specimen ruptured. The tensile creep load was applied by the use of weights acting through a system of levers. The tests were performed at different load levels expressed as a percentage of the maximum tensile strength of the geotextile under intact conditions (TCR/Tmax), which was previously determined by wide-width tensile tests according to ISO 10319:2015 (ISO 2015). Following the ISO 13431:1999 (ISO 1999), the load levels were selected from a range between 50% and 90% of the measured tensile strength of the geotextile. A video extensometer was used to monitor the geotextile strain during the creep rupture tests, using two reference points fixed to the specimens and 200 mm apart (Fig. 4). This system enabled the strains to be continuously monitored throughout the test. The time at which tensile creep rupture occurred was automatically recorded by the system.A total of 40 geotextile specimens (fresh specimens and specimens that were previously exposed to the different backfill materials for 24 months) were tested in this current study. For intact conditions (i.e., fresh specimens), six distinct load levels in the range of 67.7%–89.8% were employed and at least three geotextile specimens were tested for any specific load level (to ensure repeatability of the test results). In the case of the exhumed specimens, due to limitations in material availability, four load levels were considered (ranging from 62.7%–73.4% for exhumed specimens from the C&D material and 65.3%–72.9% for exhumed specimens from the clayey sand) and at least two specimens were tested at a given load level. It should be noted that all of the aforementioned load levels were defined with respect to the maximum tensile strength of fresh specimens determined by wide-width tensile tests. Thus, a given load level (TCR/Tmax) corresponds to the same applied load for both the fresh and exhumed specimens.Results and DiscussionShort-Term Tensile ResponseThe short-term tensile behavior of intact and exhumed geotextile specimens was evaluated by wide-width tensile tests performed according to the International Standard ISO 10319:2015 (ISO 2015). These tests were carried out at a strain rate of 20%/min using a universal testing machine. The associated results were discussed in detail in a previous paper (Vieira and Pereira 2021), but a summary of the main data is also presented herein given its relevance for this current study. Table 2 presents the results obtained for five geotextile specimens tested under intact conditions, in terms of the maximum tensile force (Tmax) and corresponding strain (εTmax), as well as the secant stiffness modulus at 2% strain (J2%) and the secant stiffness modulus at maximum load (JTmax). In turn, the results from tests carried out on exhumed geotextile specimens after a 24-month exposure to the C&D material and the clayey sand are summarized in Table 3. The average tensile load-strain curves of the geotextile for each test condition are shown in Fig. 5.Table 2. Results of tensile tests carried out on fresh specimens of geotextileTable 2. Results of tensile tests carried out on fresh specimens of geotextileSpecimenTmax (kN/m)εTmax (%)J2% (kN/m)JTmax (kN/m)Specimen 170.19.2693765Specimen 267.610.0605676Specimen 373.710.7648691Specimen 469.09.7742714Specimen 572.79.2549793Mean value70.69.7647728Confidence interval of 95%70.6±3.29.7±0.8647±93728±62Table 3. Results of tensile tests carried out on exhumed specimens of geotextile after a 24-month exposure periodTable 3. Results of tensile tests carried out on exhumed specimens of geotextile after a 24-month exposure periodSpecimenTmax (kN/m)εTmax (%)J2% (kN/m)JTmax (kN/m) Specimen 151.610.3605502 Specimen 254.79.7552566 Specimen 352.48.2666640 Specimen 450.78.8639580 Specimen 552.89.8528541 Mean value52.49.3598566 Confidence interval of 95%52.4±1.99.3±1.0598±51566±45 Specimen 151.510.2538507 Specimen 252.910.2584517 Specimen 347.810.6574453 Specimen 455.38.1633679 Specimen 550.88.6794587 Mean value51.79.5624549 Confidence interval of 95%51.7±3.49.5±1.3624±88549±77From the comparison of the results presented in Tables 2 and 3, it can be concluded that the exposure of the geotextile to both the recycled C&D material and the clayey sand led to a considerable degradation of the short-term tensile strength of the material. The reduction in the peak tensile strength of the exhumed specimens may be attributed to material handling and construction procedures, as well as the chemical and environmental degradation induced by the exposure of the specimens to the backfill materials. Vieira and Pereira (2021) evaluated the short-term tensile behavior of geotextile specimens exhumed right after the construction of the embankment built with recycled C&D material. They reported that the specimens that were exhumed immediately after embankment construction exhibited a loss of tensile strength of about 17%, in comparison with that of the fresh specimens. This reduction in the measured tensile strength was attributed to the less effective binding of the PET yarns to the nonwoven geotextile backing, caused by material handling and installation. According to the results presented in Tables 2 and 3, the decrease in tensile strength for the geotextile specimens exhumed from the C&D material was approximately 26% (i.e., 9% greater than that measured for the specimens exhumed right after installation). Thus, 9% of the tensile strength reduction resulted from the exposure to the C&D material during 24 months (i.e., chemical and environmental degradation), whereas the remaining reduction resulted from handling and construction procedures. On the other hand, the tensile strength loss after a 24-month exposure to the recycled C&D material (26% on average) was comparable to that induced by the clayey sand (27% on average), indicating that the use of the recycled backfill material did not cause any additional degradation.The results in Tables 2 and 3 also indicate that the tensile stiffness at 2% strain (J2%) was not significantly affected by the geotextile exposure to the different backfill materials, with the mean values for the exhumed specimens falling within the confidence interval for intact specimens. This small variation of tensile stiffness under low strains can be clearly observed in Fig. 5, which also shows that the strain at maximum load was rather similar irrespective of the test conditions. As a result, the secant stiffness at maximum load (JTmax) decreased significantly when the materials were exposed to the backfill materials for a period of 24 months (up to 25%).The average load-strain curve of the geotextile specimens after exposure to the clayey sand was slightly different from those of the remaining specimens, particularly in terms of tensile stiffness for strains exceeding 3% (Fig. 5). This is likely related to the higher amount of fine particles of the clayey sand that remained embedded within the geotextile pores after exhumation. As indicated earlier in Fig. 1, the fines content of the clayey sand (28.8%) was considerably higher than that of the recycled C&D material (11.7%), leading to more significant accumulation of fine particles within the geotextile. This evidence is further clarified in Fig. 6, which shows the scanning electron microscopy (SEM) images of fresh and exhumed geotextile specimens obtained through a high resolution environmental scanning electron microscope with X-ray microanalysis and electron backscattered diffraction analysis, Quanta 400 FEG ESEM/EDAX Genesis X4M (Field Electron and Ion Company, Hillsboro, Oregon) from the Materials Centre of the University of Porto. Figs. 6(a, c, and e) illustrate a PET yarn connected to the nonwoven geotextile backing (magnification ×50) for a fresh specimen, an exhumed specimen from the C&D material, and an exhumed specimen from the clayey sand, respectively, whereas Figs. 6(b, d, and f) show the individual wires of a PET yarn (magnification ×500) for the different exposure conditions. Irrespective of the exposure condition, the connections of the PET yarn to the nonwoven geotextile were still visible [Figs. 6(a, c, and e)]. Furthermore, as shown in Figs. 6(c–f), a significant amount of fine particles remained attached to the geotextile after exhumation, particularly in the case of the specimen that was previously exposed to the clayey sand [Figs. 6(e and f)].Long-Term Tensile ResponseCreep Strain BehaviorFig. 7 depicts the curves of axial strain plotted against log time from representative creep rupture tests carried out on fresh geotextile specimens, as well as specimens that were exhumed after a 24-month exposure to the recycled C&D material and the clayey sand under real environmental conditions. As mentioned earlier, these tests were performed under different load levels expressed as a percentage of the maximum tensile strength of fresh specimens (TCR/Tmax). It is noteworthy that one of the tests represented in this figure (corresponding to an exhumed specimen from the C&D material under TCR/Tmax=63.2%) is still ongoing.The results presented in Fig. 7 show that, when subjected to identical axial loads (TCR/Tmax) in the range of 67.7%–72.9%, the geotextile specimens tested under intact conditions exhibited ductile behavior with axial strains at failure exceeding 10%, whereas the specimens that were previously exposed to the backfill materials exhibited brittle behavior. For fresh geotextile specimens under load levels within the aforementioned range, the creep strains tended to increase almost linearly with time (in a semilog scale) at the initial stage of the test (primary creep), but the rupture was preceded by a substantial increase in the creep rate (tertiary creep). However, for the exhumed specimens, most creep curves show nearly linear creep strains (in a semilog scale) up to failure. The results in Fig. 7 also indicate that the behavior of fresh geotextile specimens tends to change from ductile to brittle when higher axial loads are imposed (i.e., for TCR/Tmax in the range of 77.5%–88.2%).As provided in Tables 2 and 3 and Fig. 5, the tensile strength of the exhumed specimens was lower than that of the fresh specimens. Hence, for the exhumed specimens, a given imposed creep load was actually closer to their ultimate tensile strength. Specifically, all the exhumed specimens were subjected to axial loads equal to or greater than 84.5% of their maximum tensile strength, which exceeds the level of 77.5% beyond which the fresh specimens started to exhibit brittle behavior. Therefore, the reason for the brittle behavior of the exhumed specimens appears to be that the imposed load levels (in relation to their maximum tensile strength) exceeded the threshold value beyond which the material starts to exhibit brittle response.Fig. 8(a) presents the initial axial strains for all of the geotextile specimens subjected to creep rupture testing. The initial strain was determined as the average of the strain readings obtained during the initial 60 s elapsed from the end of the loading time (i.e., from the moment when the full creep load was applied). It can be seen that the initial axial strains for fresh specimens, exhumed specimens from the recycled C&D material, and exhumed specimens from the natural soil subjected to the different axial loads ranged from 7.0% to 9.6%, 6.8% to 8.2%, and 6.7% to 9.4%, respectively.In general, the initial strains tended to increase with the applied axial load, as expected. For fresh specimens, the increase in the initial strains with the applied load was particularly significant for higher TCR/Tmax values (i.e., from 77.8% to 89.8%). The specimens that were previously exposed to the backfill materials exhibited initial deformations comparable to those of the fresh specimens when subjected to similar axial loads. It can also be observed that the geotextile specimen exhumed from the clayey sand and tested under TCR/Tmax=67.7% underwent a significant higher deformation than the remaining specimens [Fig. 8(a)]. This is possibly associated with local damage that occasionally takes place under field conditions, affecting the structure and integrity of the specimens.Comparing the initial strains from the creep rupture tests with the strains recorded during the wide-width tensile tests reported earlier [Fig. 8(b)], it can be concluded that the creep tests led generally to lower deformations under the same axial loads. These differences are likely related to the different systems used to apply the tensile loads in these tests, the rate at which the load is applied, as well as the distinct specimen configurations. However, as mentioned previously, the strains induced by the application of the creep loads tended to increase with the load magnitude, thereby following a similar trend to that of the tensile tests.The axial strains at failure obtained from the creep rupture tests are illustrated in Fig. 9(a). In turn, Fig. 9(b) compares the strains at failure from the creep rupture tests with the evolution of strains recording during the wide-width tensile tests. The total strains at failure varied from 8.6% to 13.6% for fresh geotextile specimens, from 7.6% to 9.7% for specimens that were exposed to the C&D material, and from 7.3% to 10.1% for specimens that were exposed to the natural clayey sand prior to creep testing. The exhumed specimens exhibited comparable deformations at failure, which were generally lower than those of the fresh specimens when subjected to identical axial loads. This finding suggests that the exposure to the backfill materials reduced the deformability of the geotextile, and thereby its capacity to elongate before failing in tension. This is possibly attributed to the fine soil particles embedded within the geotextile pores, which may have reduced the capability of the PET yarns to stretch. Moreover, the scatter of results was generally higher for fresh specimens tested under lower load levels (i.e., 67.7%–73.4%).As shown in Fig. 9(b), about half of the fresh geotextile specimens subjected to creep rupture testing reached strains at failure exceeding the average strain at failure obtained from the tensile tests. Conversely, most of the geotextile specimens that were previously exposed to the backfill materials experienced creep rupture at axial strains below the average strain at failure recorded in the corresponding tensile tests. Hence, the maximum strain that a geotextile can withstand when subjected to a specific constant load may be lower than the maximum strain obtained during a wide-width tensile test. In fact, the strain at failure for this particular geotextile under sustained loading was up to 24.5% lower than that measured from tensile tests on intact specimens (the conventional procedure).Fig. 10 plots the incremental geotextile strains due to sustained (i.e., creep) loading, obtained by subtracting the initial strains from the total strains at failure. In general, the incremental strains for the fresh specimens were significantly higher than those for the exhumed specimens under the same axial load. For the exhumed specimens, the incremental strains did not significantly change under different TCR/Tmax values or backfill types. However, for the fresh specimens, the incremental strains tended to decrease progressively with increasing axial load. This is because higher imposed load levels led generally to higher initial strains. As a result, lower incremental strains were experienced up to creep rupture.Creep Rupture BehaviorCreep rupture envelopes for polyester geosynthetic reinforcement products are usually characterized using a log-linear equation (ISO 2007; Bathurst et al. 2012). Fig. 11 shows the applied creep loads, expressed as a fraction of the maximum tensile strength of the geotextile determined from wide-width tensile tests on fresh specimens (TCR/Tmax), plotted against the time to rupture for all of the specimens subjected to creep rupture testing (semilogarithmic plot). Also shown in this figure are the creep rupture envelopes (log-linear approximations to the data) and the corresponding equations and correlation coefficients (R2).The results presented in Fig. 11 indicate that for short-term conditions, the fresh specimens exhibited higher strength properties than the specimens that were previously exposed to the filling materials for 24 months, which is in agreement with the results from the wide-width tensile tests discussed previously. However, for long-term conditions, and if the extrapolation of the creep rupture data to predict the lifetime corresponding to a particular applied load and/or the long-term available tensile strength for a given design lifetime is considered, the exhumed specimens lead to higher estimates of the long-term strength. This essentially stems from the slope of the creep rupture envelope obtained for fresh specimens being higher than those corresponding to the exhumed specimens. The interception of the creep rupture envelope corresponding to the fresh specimens with the curves for the exhumed specimens from the C&D material and the clayey sand occurs at creep loads (TCR/Tmax) of 60.8% (after 4.3 years) and 63.3% (after 1.6 years), respectively. These results suggest that, under long-term conditions, the exhumed geotextile specimens may exhibit higher performance than the fresh specimens. This is possibly associated with the beneficial effect of the intrusion of soil particles into the geotextile pores, leading to increased long-term strength and/or to design lifetimes exceeding those estimated for the fresh specimens when subjected to relatively low sustained loads (i.e., creep loads below the aforementioned interception values). However, further tests involving longer durations would be required to validate this assumption.Fig. 11 also shows that the scatter of results for exhumed geotextile specimens was considerably higher than that for fresh specimens, hence resulting in creep rupture envelopes characterized by lower correlation coefficients. A similar trend was also reported by other researchers in earlier studies (Pinho-Lopes et al. 2018). This suggests that the effects induced by the exposure of geosynthetics to backfill materials under real field conditions can be somewhat heterogeneous. On the other hand, the results and associated variability for specimens that were previously subjected to the C&D material were comparable to those of specimens exhumed from the clayey sand, implying that the C&D material did not induce any additional degradation when compared to that resulting from the exposure to the natural soil.Creep Reduction FactorAccording to the Technical Report ISO/TR 20432:2007 (ISO 2007), the long-term design tensile strength (TD) of geosynthetics used for soil reinforcement should be obtained by separately accounting for the detrimental effects of several influential factors on the tensile strength and can be expressed as (1) TD=TcharRFCR×RFID×RFW×RFCH×fswhere Tchar = characteristic strength (i.e., a statistic value typically derived from the mean tensile strength of the geosynthetic subtracted by two standard deviations); RFCR = creep reduction factor (to cover the effect of sustained static load); RFID = reduction factor for mechanical damage (or installation damage); RFW = reduction factor for weathering (to account for weathering during exposure prior to installation or of permanently exposed geosynthetics); RFCH = reduction factor to account for the strength loss resulting from chemical and biological degradation; and fs = safety factor to account for the extrapolation uncertainty, particularly in extrapolation over long durations.The creep rupture behavior of the high-strength geotextile used in this study was characterized in the previous section by measuring the time to rupture for each specimen subjected to a given sustained load. The results can be extrapolated to predict longer lifetimes at lower loads and thereby estimate the creep reduction factor, which is required to limit the load acting on the reinforcement to a level that will prevent creep rupture over the design life of the structure.As per the ISO/TR 20432:2007 (ISO 2007), a condition on the extrapolation of the creep rupture envelope is that there is no evidence to believe that the creep rupture behavior will change over time. This implies checking that at long durations there is no (1) abrupt change in the gradient of the creep rupture envelope; (2) abrupt change in the strain to failure; and (3) significant change in the nature of the fracture surface.In this study, the creep reduction factors for the tensile strength of the geotextile for a given design period (RFCR(t=td)) were computed based on ISO/TR 20432:2007 (ISO 2007) by considering the data obtained from the wide-width tensile tests and the creep rupture tests, as given by (2) RFCR(t=td)=TmaxTCR,R(t=td)where Tmax = maximum tensile strength obtained from wide-width tensile tests performed on fresh geotextile specimens (mean value); and TCR,R(t=td) = load leading to geosynthetic rupture (also known as the creep rupture load or creep limit) for a given design lifetime (t=td), predicted by extrapolating the corresponding creep rupture envelope.Table 4 lists the values of the creep rupture load and creep reduction factor for fresh and exhumed geotextile specimens considering design lifetimes of 30 and 75 years. The design lives of geosynthetic-reinforced soil structures may range from several years, for temporary structures, to 75–120 years for permanent structures. The selected design lifetimes of 30 and 75 years were aimed at enabling the comparison of the creep reduction factors with those from databases available in the literature (Bathurst et al. 2012; Miyata et al. 2014; Pinho-Lopes et al. 2018). It is noteworthy that, for design purposes, European and North American practices (ISO 2007; WSDOT 2009; AASHTO 2013) recommend not to extrapolate the creep rupture envelope by more than two log cycles beyond the longest observed time to creep rupture (i.e., the test with the longest duration). While the available creep rupture data for fresh specimens and exhumed specimens from the clayey sand would not fully meet this criterion, the creep reduction factors have been estimated for the purposes of this investigation.Table 4. Estimated creep rupture loads (TCR,R) and creep reduction factors (RFCR) for design lifetimes of 30 and 75 yearsTable 4. Estimated creep rupture loads (TCR,R) and creep reduction factors (RFCR) for design lifetimes of 30 and 75 yearsSpecimenTCR,R(t=30 years) (kN/m)RFCR(t=30 years)TCR,R(t=75 years) (kN/m)RFCR(t=75 years)Fresh39.61.7838.01.86Exhumed–C&D material41.31.7140.51.74Exhumed–clayey sand42.91.6442.41.67It can be observed from Table 4 that the computed creep rupture load for a design life of 30 years increased about 4.3% and 8.3% for exhumed specimens from the C&D material and the natural soil, respectively, in comparison to that predicted on the basis of the results for fresh specimens. Accordingly, the creep reduction factor corresponding to the geotextile tested under intact conditions (RFCR(t=30 years)=1.78) exceeded those obtained for the exhumed specimens.As expected, similar trends were also observed when considering a longer design life of 75 years. The creep rupture load for t=75 years increased about 6.6% and 11.6% for exhumed specimens from the C&D material and the natural soil, respectively, in relation to that for the fresh specimens. The highest creep reduction factor (RFCR(t=75 years)=1.86) was obtained for specimens tested under intact conditions. Therefore, the use of fresh specimens for creep rupture testing (i.e., the usual approach), rather than specimens that were previously exposed to the backfill materials under real environmental conditions can be considered as a conservative (i.e., safe) procedure with regard to the analysis of the long-term strength of this geosynthetic.The creep rupture factors obtained herein are comparable to those reported by other researchers in previous related studies. Bathurst et al. (2012) provided a summary of computed creep reduction factors for a wide range of geosynthetic products. For a design lifetime of 75 years, the creep reduction factors ranged from 1.36 to 1.67 for PET geosynthetics, from 1.76 to 2.8 for PP geotextiles, and from 2.48 to 3.12 for high-density polyethylene (HDPE) geogrids (Bathurst et al. 2012). Miyata et al. (2014) collected creep test data from a large database of geogrid products and found that the creep reduction factors for a design life of 75 years (computed using reference tensile strengths based on 10% strain/min tests) were in the range of 2.07–3.03 for uniaxial HDPE geogrids, 1.55–4.26 for PP geogrids, and 1.20–1.92 for woven and knitted PET geogrids. Pinho-Lopes et al. (2018) obtained a creep reduction factor of 2.08 for a woven PP geotextile and a lower creep reduction factor of 1.68 for a PET geogrid considering a design lifetime of 30 years.The differences in the results obtained from different studies involving distinct geosynthetic materials is not surprising. The reference tensile strength used to calculate the creep reduction factor may change from country to country and may also be influenced by the axial strain rate imposed in the tensile tests (Miyata et al. 2014). Furthermore, as noted by Bathurst et al. (2012), several factors such as the manufacturing process, constituent polymer type and grade, thickness of axially loaded members for integral punched and drawn polyolefins, temperature, and load level are all susceptible to affect the creep behavior of geosynthetic products to a significant extent. Therefore, the creep strain and rupture behavior of geosynthetics should ideally be evaluated using the specific geosynthetic product to be used in the project.References AASHTO. 2013. Standard practice for determination of long-term strength for geosynthetic reinforcement. AASHTO R64-13. Washington, DC: AASHTO. Allen, T. M., and R. J. Bathurst. 1996. “Combined allowable strength reduction factor for geosynthetic creep and installation damage.” Geosynth. Int. 3 (3): 407–439. https://doi.org/10.1680/gein.3.0069. Allen, T. M., and R. J. Bathurst. 2002. “Observed long-term performance of geosynthetic walls and implications for design.” Geosynth. Int. 9 (5–6): 567–606. https://doi.org/10.1680/gein.9.0228. Allen, T. M., R. J. Bathurst, and R. R. Berg. 2002. “Global level of safety and performance of geosynthetic walls: An historical perspective.” Geosynth. Int. 9 (5–6): 395–450. https://doi.org/10.1680/gein.9.0224. Arulrajah, A., M. Naeini, A. Mohammadinia, S. Horpibulsuk, and M. Leong. 2020. “Recovered plastic and demolition waste blends as railway capping materials.” Transp. Geotech. 22 (Mar): 100320. https://doi.org/10.1016/j.trgeo.2020.100320. Arulrajah, A., J. Piratheepan, M. M. Disfani, and M. W. Bo. 2013. “Geotechnical and geoenvironmental properties of recycled construction and demolition materials in pavement subbase applications.” J. Mater. Civ. Eng. 25 (8): 1077–1088. https://doi.org/10.1061/(ASCE)MT.1943-5533.0000652. Arulrajah, A., E. Yaghoubi, Y. C. Wong, and S. Horpibulsuk. 2017. “Recycled plastic granules and demolition wastes as construction materials: Resilient moduli and strength characteristics.” Constr. Build. Mater. 147 (Aug): 639–647. https://doi.org/10.1016/j.conbuildmat.2017.04.178. Bathurst, R. J., B. Q. Huang, and T. M. Allen. 2012. “Interpretation of laboratory creep testing for reliability-based analysis and load and resistance factor design (LRFD) calibration.” Geosynth. Int. 19 (1): 39–53. https://doi.org/10.1680/gein.2012.19.1.39. Benjamim, C. V. S., B. S. Bueno, and J. G. Zornberg. 2007. “Field monitoring evaluation of geotextile-reinforced soil-retaining walls.” Geosynth. Int. 14 (2): 100–118. https://doi.org/10.1680/gein.2007.14.2.100. Bueno, B. S., M. A. Costanzi, and J. G. Zornberg. 2005. “Conventional and accelerated creep tests on nonwoven needle-punched geotextiles.” Geosynth. Int. 12 (6): 276–287. https://doi.org/10.1680/gein.2005.12.6.276. CEN (European Committee for Standardization). 2002. Characterisation of waste—Leaching—Compliance test for leaching of granular waste material and sludges—Part 4. EN 12457-4. Brussels, Belgium: CEN. CEN (European Committee for Standardization). 2009. Tests for geometrical properties of aggregates—Part 11: Classification test for the constituents of coarse recycled aggregate. EN 933-11. Brussels, Belgium: CEN. Christopher, B. R. 2014. “Cost savings by using geosynthetics in the construction of civil works projects.” In Proc., 10th Int. Conf. on Geosynthetics (10ICG), 21–25. Essen, Germany: German Geotechnical Society. European Commission. 2003. Council decision 2003/33/EC establishing criteria and procedures for the acceptance of waste at landfills pursuant to Article 16 of and Annex II to Directive 1999/31/EC. Brussels, Belgium: Official Journal of the European Union. Ferreira, F. B., A. T. Gomes, C. S. Vieira, and M. L. Lopes. 2016a. “Reliability analysis of geosynthetic-reinforced steep slopes.” Geosynth. Int. 23 (4): 301–315. https://doi.org/10.1680/jgein.15.00057. Ferreira, F. B., P. M. Pereira, C. S. Vieira, and M. L. Lopes. 2021. “Time-dependent response of a recycled C&D material-geotextile interface under direct shear mode.” Materials 14 (11): 3070. https://doi.org/10.3390/ma14113070. Ferreira, F. B., C. S. Vieira, and M. L. Lopes. 2013. “Analysis of soil-geosynthetic interfaces shear strength through direct shear tests.” In Proc., Int. Symp. on Design and Practice of Geosynthetic-Reinforced Soil Structures, 44–53. Lancaster, PA: DEStech Publications. Ferreira, F. B., C. S. Vieira, and M. L. Lopes. 2015. “Direct shear behaviour of residual soil–geosynthetic interfaces—Influence of soil moisture content, soil density and geosynthetic type.” Geosynth. Int. 22 (3): 257–272. https://doi.org/10.1680/gein.15.00011. Ferreira, F. B., C. S. Vieira, and M. L. Lopes. 2016b. “Soil-geosynthetic interface strength properties from inclined plane and direct shear tests—A comparative analysis.” In Proc., 6th Asian Regional Conference on Geosynthetics: Geosynthetics for Infrastructure Development, 925–937. New Delhi, India: Indian Chapter of International Geosynthetics Society. Ferreira, F. B., C. S. Vieira, and M. L. Lopes. 2020a. “Pullout behavior of different geosynthetics-Influence of soil density and moisture content.” Front. Built Environ. 6 (12): 1–13. https://doi.org/10.3389/fbuil.2020.00012. Ferreira, F. B., C. S. Vieira, M. L. Lopes, and D. M. Carlos. 2016c. “Experimental investigation on the pullout behaviour of geosynthetics embedded in a granite residual soil.” Eur. J. Environ. Civ. Eng. 20 (9): 1147–1180. https://doi.org/10.1080/19648189.2015.1090927. Ferreira, F. B., C. S. Vieira, M. L. Lopes, and P. G. Ferreira. 2020b. “HDPE geogrid-residual soil interaction under monotonic and cyclic pullout loading.” Geosynth. Int. 27 (1): 79–96. https://doi.org/10.1680/jgein.19.00057. FHWA. 2010. Design and construction of mechanically stabilized earth walls and reinforced soil slopes. FHWA-NHI-10-024. Washington, DC: FHWA. Greenway, D., J. R. Bell, and B. Vandre. 1999. “Snailback wall—First fabric wall revisited at 25 year milestone.” In Proc., Geosynthetics ′99: Specifying Geosynthetics and Developing Design Details, 905–919. Roseville, MN: Industrial Fabrics Association International. Hsuan, Y., H. Schroeder, K. Rowe, W. Müller, J. Greenwood, D. Cazzuffi, and R. Koerner. 2008. “Long-term performance and lifetime prediction of geosynthetics.” In Proc., 4th European Conference on Geosynthetics. Austin, TX: International Geosynthetics Society. Indraratna, B., S. Nimbalkar, D. Christie, C. Rujikiatkamjorn, and J. S. Vinod. 2010. “Field assessment of the performance of a ballasted rail track with and without geosynthetics.” J. Geotech. Geoenviron. Eng. 136 (7): 907–917. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000312. ISO. 1999. Geotextiles and geotextile-related products—Determination of tensile creep and creep rupture behaviour. ISO 13431. Geneva: ISO. ISO. 2007. Guidelines for the determination of the long-term strength of geosynthetics for soil reinforcement. ISO/TR 20432. Geneva: ISO. ISO. 2015. Geosynthetics—Wide-width tensile test. ISO 10319. Geneva: ISO. Karnamprabhakara, B. K., U. Balunaini, A. Arulrajah, and R. Evans. 2021b. “Axial pullout resistance and interface direct shear properties of geogrids in pond ash.” Int. J. Geosynth. Ground Eng. 7 (2): 22. https://doi.org/10.1007/s40891-021-00266-x. Karnam Prabhakara, B. K., U. Balunaini, and A. Arulrajah. 2021a. “Development of a unique test apparatus to conduct axial and transverse pullout testing on geogrid reinforcements.” J. Mater. Civ. Eng. 33 (1): 04020406. https://doi.org/10.1061/(ASCE)MT.1943-5533.0003497. Koerner, R. M., Y. G. Hsuan, and G. R. Koerner. 2017. “Lifetime predictions of exposed geotextiles and geomembranes.” Geosynth. Int. 24 (2): 198–212. https://doi.org/10.1680/jgein.16.00026. Kongkitkul, W., F. Tatsuoka, and D. Hirakawa. 2007. “Creep rupture curve for simultaneous creep deformation and degradation of geosynthetic reinforcement.” Geosynth. Int. 14 (4): 189–200. https://doi.org/10.1680/gein.2007.14.4.189. Leshchinsky, D., M. Dechasakulsom, V. N. Kaliakin, and H. I. Ling. 1997. “Creep and stress relaxation of geogrids.” Geosynth. Int. 4 (5): 463–479. https://doi.org/10.1680/gein.4.0102. Li, A. L., and R. K. Rowe. 2001. “Influence of creep and stress-relaxation of geosynthetic reinforcement on embankment behaviour.” Geosynth. Int. 8 (3): 233–270. https://doi.org/10.1680/gein.8.0194. Maghool, F., A. Arulrajah, M. Mirzababaei, C. Suksiripattanapong, and S. Horpibulsuk. 2020. “Interface shear strength properties of geogrid-reinforced steel slags using a large-scale direct shear testing apparatus.” Geotext. Geomembr. 48 (5): 625–633. https://doi.org/10.1016/j.geotexmem.2020.04.001. Pinho-Lopes, M., A. M. Paula, and M. L. Lopes. 2018. “Long-term response and design of two geosynthetics: Effect of field installation damage.” Geosynth. Int. 25 (1): 98–117. https://doi.org/10.1680/jgein.17.00036. Rahman, M. A., A. Arulrajah, J. Piratheepan, M. W. Bo, and M. A. Imteaz. 2014. “Resilient modulus and permanent deformation responses of geogrid-reinforced construction and demolition materials.” J. Mater. Civ. Eng. 26 (3): 512–519. https://doi.org/10.1061/(ASCE)MT.1943-5533.0000824. Rowe, R. K., and C. J. Jones. 2000. “Geosynthetics: Innovative materials and rational design.” In Proc., ISRM Int. Symp. Lisbon, Portugal: International Society for Rock Mechanics. Santos, E., E. Palmeira, and R. Bathurst. 2014. “Performance of two geosynthetic reinforced walls with recycled construction waste backfill and constructed on collapsible ground.” Geosynth. Int. 21 (4): 256–269. https://doi.org/10.1680/gein.14.00013. Santos, E. C. G., E. M. Palmeira, and R. J. Bathurst. 2013. “Behaviour of a geogrid reinforced wall built with recycled construction and demolition waste backfill on a collapsible foundation.” Geotext. Geomembr. 39 (Aug): 9–19. https://doi.org/10.1016/j.geotexmem.2013.07.002. Vieira, C. S., F. B. Ferreira, P. M. Pereira, and M. L. Lopes. 2020a. “Pullout behaviour of geosynthetics in a recycled construction and demolition material—Effects of cyclic loading.” Transp. Geotech. 23 (Jun): 100346. https://doi.org/10.1016/j.trgeo.2020.100346. Vieira, C. S., and P. M. Pereira. 2015a. “Damage induced by recycled construction and demolition wastes on the short-term tensile behaviour of two geosynthetics.” Transp. Geotech. 4 (Sep): 64–75. https://doi.org/10.1016/j.trgeo.2015.07.002. Vieira, C. S., and P. M. Pereira. 2015b. “Use of recycled construction and demolition materials in geotechnical applications: A review.” Resour. Conserv. Recycl. 103 (Oct): 192–204. https://doi.org/10.1016/j.resconrec.2015.07.023. Vieira, C. S., and P. M. Pereira. 2021. “Short-term tensile behaviour of three geosynthetics after exposure to recycled Construction and Demolition materials.” Constr. Build. Mater. 273 (Mar): 122031. https://doi.org/10.1016/j.conbuildmat.2020.122031. Vieira, C. S., P. M. Pereira, F. B. Ferreira, and M. L. Lopes. 2020b. “Pullout behaviour of geogrids embedded in a recycled construction and demolition material. Effects of specimen size and displacement rate.” Sustainability 12 (9): 3825. https://doi.org/10.3390/su12093825. Vieira, C. S., P. M. Pereira, and M. L. Lopes. 2016. “Recycled construction and demolition wastes as filling material for geosynthetic reinforced structures. Interface properties.” J. Cleaner Prod. 124 (Jun): 299–311. https://doi.org/10.1016/j.jclepro.2016.02.115. WSDOT. 2009. Standard practice for determination of long-term strength for geosynthetic reinforcement. WSDOT T925. Olympia, WA: WSDOT.