Next Article in Journal
MACLA-LSTM: A Novel Approach for Forecasting Water Demand
Next Article in Special Issue
Developing Sustainable Asphalt Mixtures Using High-Density Polyethylene Plastic Waste Material
Previous Article in Journal
Biobjective Optimization Model Considering Risk and Profit for the Multienterprise Layout Design in Village-Level Industrial Parks in China
Previous Article in Special Issue
Investigation of Using Crushed Glass Waste as Filler Replacement in Hot Asphalt Mixtures
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 4
10.3390/su15043622
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Structural Performance of Foamed Asphalt Base in a Full Depth Reclaimed and Sustainable Pavement

by
Konstantinos Gkyrtis
*,
Christina Plati
and
Andreas Loizos
Laboratory of Pavement Engineering, National Technical University of Athens (NTUA), 15772 Athens, Greece
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(4), 3622; https://doi.org/10.3390/su15043622
Submission received: 30 January 2023 / Revised: 10 February 2023 / Accepted: 15 February 2023 / Published: 16 February 2023
(This article belongs to the Special Issue Green Infrastructure and Recycled Materials Sustainability)
Browse Figures
Versions Notes

Abstract

:
Environmental awareness and budgetary constraints in road construction and rehabilitation require innovative and at the same time environmentally friendly solutions in favor of resource conservation, in any type of pavement intervention. In the meantime, surface interventions in severely deteriorated pavements are insufficient, increasing as such the recycling potential of existing materials in Full Depth Reclamation (FDR). This study addresses the efficiency of an FDR process in which the recycled base course of an existing semi-rigid pavement is stabilized with Foamed Asphalt (FA). The adverse effects of FA curing, which have a significant impact on pavement evaluation, particularly in the early stages of a pavement’s life cycle, prompted the conduct of a comprehensive testing campaign at two different periods, shortly after pavement rehabilitation and five years thereafter. The campaign included nondestructive testing at the aforementioned periods using Ground Penetrating Radar (GPR) and the Falling Weight Deflectometer (FWD), which facilitate the structural evaluation of the FA base. Using the collected nondestructive testing data, the principles of Multi-Layer Elastic Theory (MLET) and Genetic Algorithms (GA) are used to investigate the performance of the FA base. Both analysis tools captured the FA strength increase in the second period, but an increased variance of the FA modulus was observed mainly in the first test period, probably due to the curing effect. In addition, GA showed an advantage over other, more conventional tools for back-analysis of pavement stiffness, resulting in an interesting correlation potential of FA modulus to a deflection-based parameter reflecting the condition of the FA layer. Overall, the study contributes to the development of a practical methodology suitable for the evaluation of non-conventional and sustainable pavement structures.

1. Introduction

In the framework of sustainable pavement development, reusing existing materials instead of virgin materials has gained increasing popularity worldwide because of its potential to simultaneously preserve and sustain limited natural resources [1,2]. In the meantime, budget restrictions for road construction, reconstruction, and maintenance require innovative yet environmentally friendly solutions to guarantee road users an acceptable service level [3]. Reutilized materials can be used both in new pavement construction as well as when undertaking rehabilitation actions in severely deteriorated pavement structures. For these types of pavements, surface interventions are probably insufficient, leaving margin for a hidden structural issue to reappear again in the pavement’s surface. Therefore, an in-depth rehabilitation is probably needed. However, the demand for purely new high-performance materials is a critical concern for the pavement engineering community in the context of energy consumption, resource conservation, and overall pavement sustainability [1]. As such, recycling in pavements has gained a lot of attention in the last decades focusing on new opportunities and construction challenges for both bound and unbound layers [1,4].
Amongst the available techniques for recycling existing pavement materials, Cold In-Place Recycling (CIR) is an effective means for substantially shortening the construction phase and ensuring cost effectiveness [5,6]. The technique is also a sustainable engineering solution that significantly reduces storage costs, energy and fuel consumption, and lowers overall gas and noise emissions, as there is no need for in-plant production or transportation to the paving site [7,8,9,10]. If the recycling depth extends below the bottom of the asphalt layer, CIR is commonly referred to as Full Depth Reclamation (FDR). The FDR technique aims to structurally strengthen base/subbase layers of existing pavement structures facing severe damage [5,11]. FDR can be implemented in many manners and one of these is with the use of Foamed Asphalt (FA) for the stabilization of existing pavement materials.
By definition, FA is produced by injecting water into hot bitumen (at 2–3% of bitumen weight), which expands in volume and subsequently decreases in viscosity, which is ideal for FA to adequately coat aggregates at ambient temperatures [6]. Due to its sustainable potential, FA has also been recently investigated for stabilizing asphalt layers during pavement rehabilitation. Mixing at ambient temperatures leads to Warm-Mix Asphalt (WMA) technology, which can be efficiently incorporated into rehabilitated asphalt layers, resulting in durable and resilient performance of the reconstructed layers [12]. The importance of incorporating fillers into the FA mix design to increase strength has also been recognized [13].
However, the authors focused on the use of FA to stabilize the base course when rehabilitating pavements with full-depth interventions. The reason for this choice is that once the remaining layers of the pavement are strengthened to provide structural stability, any type of future and/or regular interventions can be focused on the upper pavement layers to restore serviceability, an approach that considers feasibility and cost effectiveness. Indeed, there is literature evidence that for severely damaged pavements requiring extensive rehabilitation, FA can be considered as a promising option to stabilize the base course [3,14,15].
An important issue concerning FA-stabilized materials is the absence of a mix design methodology suitable for CIR mixes [16]. Some researchers have pinpointed the moisture susceptibility of Foamed Asphalt base materials [17], especially prior to Asphalt Concrete (AC) overlay construction [18], while others have concluded that FA outperforms bitumen emulsion for base stabilization in terms of increased moisture and frost resistance [8,18]. Skotnicki et al. [14] argued in favor of higher strength values of FA mixtures compared to those containing asphalt emulsion. In addition, the incorporation of cement as filler into the FA mixtures has been considered as necessary to reduce the moisture susceptibility. Mugume [15] reported that cement rates lower than 0.5% are in general non-preferable. Therefore, it can be stated that the reduced susceptibility to frost and moisture damage of properly designed mixtures coupled with the stabilization contribution of FA raise the sustainability and the suitability of the FDR-FA technique versus a non-stabilized FDR technique. Moreover, achieving water stability can be also accompanied by a satisfactory performance against fatigue failure, which is a critical pavement performance indicator too [12].
The overall mechanical behavior of FA-stabilized base materials is somewhere between the characteristics of AC and non-stabilized aggregate materials. Many researchers have focused on the advanced laboratory characterization of these stabilized base materials (either foamed or emulsion-based) [3,14,15]. Their emphasis is being put on determining fundamental engineering properties in terms of stiffness indicators for incorporation into mix design procedures [3,8,18,19,20,21]. Among these stiffness indicators, the triaxial dynamic modulus, indirect tensile strength, and resilient modulus have been mainly utilized to evaluate temperature sensitivity, fatigue resistance, and permanent deformation rates. There is available evidence that FA stabilizers allow the mixture to exhibit mechanical behaviors superior to more conventional aggregate materials and even approach the behavior of AC materials [14,15,19,21].
Even though significant research efforts have been carried out for laboratory material characterization, limited documentation exists in respect to overall in situ performance of FA-stabilized materials throughout their lifecycle [14,22,23,24]. This may be because of aspects mainly related to construction issues or the curing effect of the stabilized material that raises many difficulties in the framework of a more rational in situ structural evaluation, especially during the early life of rehabilitated pavement structures [25]. For instance, inadequate water reduces the mixture’s workability during compaction, whereas an excessive amount of water leads to lower density and poor early strength thereby retarding the curing process [15]. Further, once the pavement layer is constructed, core extraction in many cases is not possible during the early life stages, as moisture within the material may be not fully absorbed and thus material strength may not reach its maximum value.
Therefore, traditional evaluation methods based on ground-truth data (e.g., coring, etc.), apart from being destructive and location-specific, are also insufficient and unsuitable for FDR-FA pavements. There is no practicality nor engineering justification in repeating the coring process to evaluate the strength of FA mixtures until the curing process is completed. Moreover, laboratory testing does not necessarily reflect the material behavior under real traffic conditions. These remarks form the rationale behind this study, since according to relevant interests for evaluating FDR-FA structures, a shift to in situ Non Destructive Testing (NDT) evaluation techniques has been noticed for those types of non-conventional pavements. In particular, the Falling Weight Deflectometer (FWD) remains the standard approach for the evaluation of pavement structural integrity [26,27], so there is great interest in evaluating its capability to assess non-conventional pavement structures, including FDR-FA pavements. Based on the limited relevant experience from the analysis of FWD testing data, the present research studies the feasibility of developing a practical approach for the in situ structural evaluation of an FA-stabilized base layer of an FDR rehabilitated pavement section based on deflectometric data collected though a thorough experimental campaign followed by the necessary analysis with proper tools [28,29].

2. Background

Despite recent advances related to continuous deflectometric testing at traffic speeds, FWD remains the standard approach for measuring surface deflections worldwide and assessing the structural integrity of pavements at both network and project levels [30,31]. A typical deflection bowl is shown in Figure 1. Hakim and Brown [30] defined three categories for analyzing deflections from FWD, based on the level of accuracy required, the complexity of the pavement, and the amount of information available. Category 1 refers to the use of Deflection-Based Parameters (DBPs) to derive pavement condition; Category 2 includes a standard back-analysis process to represent a layered pavement; and Category 3 is recommended for more detailed analysis to account for more complex pavement types or site conditions.
For the first category, Table 1 provides a list of commonly used DBPs along with the pavement area for which they provide structural evidence. The subscripted characters in the “mathematical expression” column correspond to the distance from the FWD loading plate at which a deflection value is recorded.
By using DBPs at the network level, the pavement structural condition can be rapidly screened leading to a more enhanced maintenance prioritization and decision making process [32]. A semi-empirical, semi-mechanistic approach has been developed for pavements structural analysis, according to which the supplementary use of deflection bowl parameters along with visual inspection surveys facilitate a benchmarking assessment of pavement structures indicating areas for potential rehabilitation needs [31]. In addition, a new methodology has been recently developed for pavement network level assessment based on the FWD deflections without the necessity of knowing layer thicknesses [33].
Beyond this, multiple correlations between DBPs and semi-rigid base strength have been reported. In particular, the curvature index of d300 − d450 is considered as the most sensitive index for semi-rigid base modulus, with a threshold value of 6 μm being proposed to discriminate intact and distressed areas [34]. The index d200 − d600 has also been proposed for semi-rigid base bearing capacity evaluation [35]. It is to be noted that these observations may not be suitable for cases where different materials and environmental conditions are likely to be faced and as such, they should not be generally adopted without proper validation. However, as the current trend for pavement condition assessment is the measurement of deflections at traffic speeds with advanced deflectometers [36], research interests related to DBPs will consistently remain ongoing.
From a more mechanistic view, material moduli have to be estimated from FWD raw data utilizing closed form solutions, and iterative and database searching back-calculation methods [37]. Most back-analysis procedures rely on the basic principles of Multi-Layer Elastic Theory (MLET). The most conventionally utilized method is based on iterative optimization procedures where pavement responses are calculated using MLET [38]. According to MLET, each pavement layer is assumed to behave in a linearly elastic manner, regardless of temperatures or loading frequencies, which are known to mainly affect the performance of bound materials. MLET is a reference theory for analyzing pavement designs where the underlying layer always has a lower stiffness, as is the case with a typical flexible pavement. However, MLET can be used for other types of pavement designs, but additional knowledge may be required (e.g., laboratory data, material testing, correlation analysis, etc.) to support the interpretation of MLET analysis with sufficient engineering judgment.
Although popular, the procedure of back-analysis may suffer from several issues including among others, algorithms for AC temperature adjustment [39], the potential impact of stiff layers, and the non-uniqueness of the derived solutions, as different layer stiffness combinations may lead to similar predicted deflection bowls. As a result, back-calculated moduli may be procedure dependent, in that different procedures may lead to different stiffness profiles for the same pavement structure [40]. In addition, when stiff layers exist between weaker layers, back-analysis may become a challenging issue [28]. In particular, for the case of an FDR-FA pavement, where the AC and FA materials exhibit similar stiffness moduli at intermediate temperatures [19], back-analysis may lead to the overestimation of AC modulus and the underestimation of the FA modulus [19]. FDR-FA pavements are not always fully aligned with MLET assumptions and thus, engineering judgment is required for the analysis of in situ collected data in regard to the calculated layer moduli.
Alternatively, the introduction of GA within back-calculation procedures has gained increased popularity for pavement analysis. GA is a robust and randomized search algorithm that optimizes the search domain for back-calculation [41]. Extensive research has been carried out for the optimal selection of GA parameters upon improving back-analysis accuracy [29,42,43]. GA seems to outperform conventional back-analysis providing a more global search capability that overcomes the problem of local minima convergence. There are also relatively recent examples of GA use in semi-rigid pavements. On this basis, in a previous research study [44], a preliminary investigation of the GA-based optimization tool for the assessment of an FDR-FA pavement structure was performed and the conclusions were encouraging. As such, this study aims to further assess in situ evaluation challenges and data interpretation for the condition analysis of FDR-FA pavements based on the GA tool.

3. Test Site, Materials, and Testing

In respect to the test site, the pavement’s investigation length was approximately 2.5 km. A high-traffic severely deteriorated pavement of semi-rigid structure (Figure 2a) in Southern Europe was rehabilitated and the FDR-FA pavement section under investigation (i.e., after rehabilitation) is illustrated in Figure 2b.
According to the FDR principles, all AC layers and part of the underlying layer (i.e., the CBM layer in this study) have to be treated [11]. In particular, after the AC layers were milled, a significant portion of the stabilized base was pulverized without pre-milling. According to the mix design, the base material was stabilized with 2.25% Foamed Asphalt and 1% cement. These design values were defined as optimal after conducting a laboratory experiment on fabricated specimens tested for their indirect tensile strength [45] and unconfined compressive strength [46]. It is worth noting that the chosen values are also in agreement with other recent studies [15]. Since the purpose of this study goes beyond the design requirements of FA mixes, the reader may refer to [47] for more details.
After designing the FA mixture, the analytical design for the rehabilitated pavement demanded a 9 cm AC overlay to be laid on the FA base surface. A uniform thickness for the FA base was iteratively adjusted to 28 cm to fulfill structural capacity requirements.
The in situ experiment with NDT took place during two separate periods; the first one (Period A) was shortly after pavement rehabilitation (i.e., five days after the compaction of AC layers was completed and before the pavement was opened to traffic)—representing the early life performance of the FA-stabilized base material, and the second one (Period B) was five years after pavement rehabilitation—representing the long-term performance. NDT included the FWD and the Ground Penetrating Radar (GPR) for the estimation of layers’ thicknesses. GPR tests at Period A revealed that the as-built thicknesses of AC and FA layers confidently matched the design values and the thickness of the remaining CBM layer was found to be around 20 cm with little variance.
During both test periods, a 50 kN FWD load was applied on the pavement surface at the outer wheel-path of the heavy-duty traffic lane and deflections were recorded through nine sensors. In the meantime, temperature records within the AC layer through properly drilled holes led to a range of 16–20 °C and 23–25 °C during Periods A and B, respectively. Based on the in situ data, FA-stabilized base moduli were back-calculated with two analysis tools and were comparatively assessed. Thereafter, the potential of utilizing DBPs as reliable indexes for the assessment of the structural condition of the FA-stabilized base material was investigated. A flowchart for the study’s research steps is given in Figure 3.
The utilized tools to estimate FA moduli (EFA) included the EVERCALC software (version 5.0) [28] and the BackGenetic3D (version 1.0) [29]. The latter works on the principles of the Genetic Algorithm (GA) theory.

4. Analysis and Results

4.1. Data Processing

For the analysis purpose, atypical deflection basins with abrupt or abnormal shapes were excluded from further analysis. After that, 37 deflection basins were analyzed representing identical test locations for both test periods. In general, deflections from all the sensors were found to be lower in Period B than in Period A, although temperatures were higher during Period B. An average deflection basin per period is illustrated in Figure 4 together with the corresponding AC temperature range that was recorded. Although the Coefficient of Variation (CV %) of deflections per sensor was around 33–43% for both periods, the purpose of this figure is to roughly demonstrate that pavement condition of the whole length under investigation is in good structural condition as pavement deflections are kept at relatively low levels.
On average, measured deflections were reduced during Period B according to a ratio of approximately 0.89 with a coefficient of variation of 21%. This ratio corresponds to the mean reduction in deflection records considering data from the entire deflection basin, i.e., all FWD sensors. The calculated ratio of 0.89, which corresponds to an average 11% reduction in measured deflections during the transition from Period A to Period B, clearly demonstrates the significant effect of the FA material curing on the overall improvement in pavement performance. Based on temperature discrepancies between the two test periods, a one-to-one comparison may not be fully correct. Yet, the actual reduction should be even higher if temperature normalization had been considered for measured deflections. As such, the 11% reduction may be considered as conservative.

4.2. Moduli Assessment

Back-analysis was performed for each of the 37 test locations for both investigatory periods implementing both analysis tools. In the absence of a calibrated algorithm for temperature-based normalization, all FA moduli were evaluated according to their raw values, so there is no bias in the evaluation of stiffness. This means that the moduli were calculated at the temperature measured in situ at each test location. Figure 4b shows that the observed temperatures in both time periods fall in the range of 16–25 °C, which corresponds to a moderate temperature spectrum for Southern Europe. At such temperatures, the viscoelastic nature of bound materials is not normally a critical issue for pavement analysis [26].
Given these remarks, the root-mean-square values of the calculated deflections centered around 2–3% for both analysis tools. The back-analysis results for FA modulus are summarized in Table 2 and Figure 5 in the form of descriptive statistics (i.e., median and average values) and box plots, respectively. Descriptive statistics provide an overview of the field FA moduli on the purpose of pavement evaluation. At the next stage of the correlation analysis, each individual value of FA modulus is to be used.
From Table 2, it can be seen that during Period A, the use of EVERCALC resulted in moduli estimation with higher range and higher variance (CV = 45.3%). On the contrary, the GA tool seems to be more consistent in back-calculating the FA modulus at Period A, as a significantly lower variance has been detected, indicating increased reliability in FA material characterization. During Period B, both processes provided results with increased variance indicating potential stiffness fluctuations within the length of 2.5 km five years after the pavement rehabilitation.
In addition, it can be observed that the FA moduli for Period B have increased indicating the curing effect on material characteristics and thus, overall pavement performance improvement. This is also supported by the measured deflections during the two periods. The FA base outperforms conventional granular base materials according to the range of FA moduli of 1000–4000 MPa considering analysis results from both investigation periods. From the median moduli values observed, it could be argued that in situ moduli values seem to be in a similar order of magnitude with other moduli values that have been also detected in other relevant research for an FA-stabilized base [19]. For comparison, typical back-calculated moduli for gravel and crushed stone materials are in the range of 100–300 MPa [48], which is far below the observed range for FA moduli.
Figure 6 illustrates the correlation of estimated moduli from EVERCALC (EEVER) and GA tool (EGEN) for both investigatory periods. During Period A (Figure 6a), the achieved correlation based on linear regression is poor (R2 = 0.23), indicating inconsistencies between the two methodologies for FA modulus back-calculation. On the contrary, it seems that the correlation of EVERCALC and GA moduli is better (R2 = 0.80) during Period B (Figure 6b).
The two sets of moduli per period were subjected to a t-test analysis (two-sample assuming equal variances), which is presented in Table 3. The null hypothesis assumes that there is no significant deference between the two processes for moduli estimation. The test was performed for a 95% confidence level, and the null hypothesis is accepted when |tstat| < tcritical. It can be seen that for Period A, the null hypothesis is rejected, and for Period B, the null hypothesis is accepted indicating that the achieved correlation has become more robust. This probably implies that during Period B, both processes may be able to accommodate the long-term stiffness profile of an FDR-FA pavement, when the FA-stabilized material has fully undergone the curing process.
In any case, the differentiation of R2 values between the two test periods stimulated the authors to further assess the correlations between FA moduli and DBPs.

4.3. Correlations between FA Moduli and DBPs

As an additional analysis step, the reliability of the two analysis theories under consideration was evaluated, by investigating correlations between DBPs and FA moduli estimations. A sufficient number of DBPs was considered for the base layer structural evaluation. The bottom of the FA layer is located at a depth of approximately 33 cm, and thus, deflections d200 and d300 were anticipated to have an impact in the estimation of FA modulus. Stimulated from past studies [34,35], multiple deflection indicators were considered, including at first differences in deflections, namely, d200 − d300, d200 − d450, d200 − d600, d200 − d900, d200 − d1200, d200 − d1500, d200 − d1800, d300 − d450, d300 − d600, and d450 − d600. Thereafter, three additional dimensionless indexes were considered that are reported to provide information for the condition of the middle layers as defined in [49]. These include the shape factor F2 = (d300 − d900)/d600, the area factor A2 = 6 × (d300 + 2d450 + d600)/d0, and the area index AI2 = (d300 + d600)/(2d0). An overview of the achieved correlations is presented in Figure 7. A power law model was used for correlating FA modulus with each DBP.
No significant correlation could be achieved with the EEVER for either of the investigation periods. In particular, during Period A, it was not possible to define a structural index for the assessment of the structural condition of the FA material, as R2 < 0.20 for all DBPs considered. During Period B, the maximum correlation with the EEVER was achieved for the deflection indexes d200 − d450 and d300 − d450, where R2 = 0.63, which cannot be considered as satisfactory. On the contrary, improved correlations were achieved through the GA, especially for the indicators d200 − d450 and d200 − d600. These improved correlations fall in the range R2 = 0.65–0.75 during Period A and in the range of R2 = 0.70–0.93 during Period B.
Based on these observations, three deflection indicators (with the higher R2 values) were selected for further statistical tests. FA moduli predicted from the indexes d200 − d450, d200 − d600, and d300 − d450 through power law regression analysis were compared with back-calculated moduli and the results from the t-test are presented in Table 4. Based on the poor correlations that were observed with EEVER during Period A, this case was excluded from the t-test analysis.
The null hypothesis is accepted for all cases (|tstat| < tcritical), indicating that any differences in the predicted moduli are not statistically significant. As such, for the FDR-FA pavement under investigation, it seems that FA moduli can be confidently estimated through DBPs through the GA tool for Period A and through both tools for Period B.
Considering the R2 values, the indicator d200 − d450 is selected for the estimation of FA moduli for both periods through GA. Similar remarks are available in the international literature too [34,35]. The selection of the indicator d200 − d450 as a more representative deflection indicator for the assessment of FA structural condition is further justified in this study, as the thickness of the FA layer is around 26–28cm, which nearly corresponds to this basin area. Correlations between EGEN and d200 − d450 are presented for the investigatory Periods A and B in Figure 8.
In Period A, the achieved correlation is lower than its respective value in Period B. This observation can be potentially attributed to the curing effect of the FA material, which has not been completed during Period A. On the other hand, assuming that the curing process has been fully completed in Period B, the material behavior is probably more consistent and thus DBPs can be used more confidently for the structural evaluation of FA material in the long term.

5. Conclusions

FDR with Foamed Asphalt (FA) as a base stabilizer is a relatively new rehabilitation technique with improved sustainable characteristics. Based on a given FA mix design for stabilizing the base course of a semi-rigid pavement that was rehabilitated, this study contributes to the evaluation of the field performance of the FA material with an in situ test campaign using NDT measurements and the related analysis. The main results are as follows:
  • The FA modulus was estimated to be approximately 1500 and 2500 MPa for Periods A (early life performance) and B (long-term performance), respectively, in accordance with past studies. The increase of FA moduli values in Period B produces evidence in support of the FA curing effect on improving overall pavement performance.
  • Regarding the back-analysis processes, it was observed that the EEVER of FA material during Period A exhibited increased variance, which was not expected for a relatively short investigatory length shortly after pavement rehabilitation. In addition, no significant correlations were observed between DBPs and EEVER. As such, the use of EVERCALC upon FA moduli estimation may not be suitable enough.
  • On the contrary, it seems that the GA tool was more consistent in terms of FA modulus estimation based on the lower variance observed in the back-calculated moduli of Period A. From the regression analysis, the effect size of the correlation coefficient was acceptable for the case of EGEN and DBPs-estimated FA moduli.
  • The potential of using DBPs as reliable indexes for evaluating the structural conditions of the FA-stabilized base materials was demonstrated. In more detail, the deflection index d200 − d450 was found to be an optimal indicator for the assessment of the FA-stabilized base structural condition, as it was significantly correlated with FA modulus for both investigation periods.
  • As a result, GA seems to exhibit an advantage over other more conventional tools for the back-analysis of pavement stiffness and the suitability assessment of DBPs for FA modulus estimation at non-conventional and sustainable pavement structures.
Overall, it is believed that the methodological framework presented in this paper could be easily applicable in other FDR-FA pavements, where different traffic or environmental conditions are expected to be encountered.

Author Contributions

Conceptualization, K.G., C.P. and A.L.; Methodology, K.G., C.P. and A.L.; Analysis, K.G.; Writing—original draft preparation, K.G.; Writing—review and editing, K.G., C.P. and A.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Zhao, Y.; Goulias, D.; Tefa, L.; Bassani, M. Life Cycle Economic and Environmental Impacts of CDW Recycled Aggregates in Roadway Construction and Rehabilitation. Sustainability 2021, 13, 8611. [Google Scholar] [CrossRef]
  2. Tebaldi, G.; Dave, E.; Hugener, M.; Falchetto, A.C.; Perraton, D.; Grilli, A.; Lo Presti, D.; Pasetto, M.; Loizos, A.; Jenkins, K.; et al. Cold Recycling of Reclaimed Asphalt Pavements. In Testing and Characterization of Sustainable Innovative Bituminous Materials and Systems; Partl, M., Porot, L., Di Benedetto, H., Canestrari, F., Marsac, P., Tebaldi, G., Eds.; RILEM State-of-the-Art Reports; Springer: Cham, Switzerland, 2018; Volume 24, pp. 239–296. [Google Scholar]
  3. Baldo, N.; Rondinella, F.; Daneluz, F.; Pasetto, M. Foamed Bitumen Mixtures for Road Construction Made with 100% Waste Materials: A Laboratory Study. Sustainability 2022, 14, 6056. [Google Scholar] [CrossRef]
  4. Thives, L.P.; Ghisi, E. Asphalt mixtures emission and energy consumption: A review. Renew. Sustain. Energy Rev. 2017, 72, 473–484. [Google Scholar] [CrossRef]
  5. Bocci, M.A.; Bocci, M.; Grilli, A.; Cardone, F.; Ferrotti, G. Full-depth reclamation for the rehabilitation of local roads: A case study. Int. J. Pavement Eng. 2014, 15, 191–201. [Google Scholar] [CrossRef]
  6. Mondal, P.G.; Kuna, K.K. Influence of bitumen foam characteristics on moisture susceptibility of foamed bitumen stabilised mixes. Int. J. Pavement Eng. 2022. [Google Scholar] [CrossRef]
  7. Sangiorgi, C.; Tataranni, P.; Simone, A.; Vignali, V.; Lantieri, C.; Dondi, G. A laboratory and filed evaluation of Cold Recycled Mixture for base layer entirely made with Reclaimed Asphalt Pavement. Constr. Build. Mater. 2017, 138, 232–239. [Google Scholar] [CrossRef]
  8. Iwański, M.; Chomicz-Kowalska, A. Application Of The Foamed Bitumen And Bitumen Emulsion To The Road Base Mixes In The Deep Cold Recycling Technology. Balt. J. Road Bridge Eng. 2016, 11, 291–301. [Google Scholar] [CrossRef]
  9. Wayne-Lee, K.W.; Mueller, M.; Singh, A. Cold in-place recycling as a sustainable pavement practice. J. Civ. Eng. Archit. 2016, 8, 680–692. [Google Scholar]
  10. Chen, Z.; Yi, J.; Zhao, H.; Luan, H.; Xu, M.; Zhang, L.; Feng, D. Strength development and deterioration mechanisms of foamed asphalt cold recycled mixture based on MD simulation. Constr. Build. Mater. 2020, 269, 121324. [Google Scholar] [CrossRef]
  11. Gonzalo-Orden, H.; Linares-Unamunzaga, A.; Pérez-Acebo, H.; Díaz-Minguela, J. Advances in the Study of the Behavior of Full-Depth Reclamation (FDR) with Cement. Appl. Sci. 2019, 9, 3055. [Google Scholar] [CrossRef] [Green Version]
  12. Xie, J.; Huang, W.; Hu, B.; Xiao, Z.; Hassan, H.M.Z.; Wu, K. Study on the Road Performance of Foamed Warm-Mixed Reclaimed Semi-Flexible Asphalt Pavement Material. Materials 2021, 14, 5379. [Google Scholar] [CrossRef] [PubMed]
  13. Deb, P.; Singh, K.L. Effect of Curing on Failure Characteristics of Cold Mix Asphalt Containing Different Fillers. Iran. J. Sci. Technol. Trans. Civ. Eng. 2023. [Google Scholar] [CrossRef]
  14. Skotnicki, L.; Kuźniewski, J.; Szydlo, A. Stiffness Identification of Foamed Asphalt Mixtures with Cement, Evaluated in Laboratory and In Situ in Road Pavements. Materials 2020, 13, 1128. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Mugume, R.B. Investigation of Foamed Bitumen Mixes Using Reclaimed Asphalt Pavement Materials for Cold Recycling Technology. Int. J. Pavement Res. Technol. 2022, 15, 98–110. [Google Scholar] [CrossRef]
  16. Bazrafshan-Moghadam, B.; Farhad-Mollashahi, H. Suggesting a Simple Design Method for Cold Recycled Asphalt Mixes with Asphalt Emulsion. J. Civ. Eng. Manag. 2017, 23, 966–976. [Google Scholar] [CrossRef] [Green Version]
  17. Chen, D.H.; Bilyeu, J.; Scullion, T.; Nazarian, S.; Chiu, C.T. Failure Investigation of a Foamed-Asphalt Highway Project. J. Infrastruct. Syst. 2006, 12, 33–40. [Google Scholar] [CrossRef]
  18. Kim, Y.; Im, S.; Lee, H. Impacts of Curing Time And Moisture Content on Engineering Properties of Cold In-Place Recycling Mixtures Using Foamed or Emulsified Asphalt. J. Mater. Civ. Eng. 2011, 23, 542–553. [Google Scholar] [CrossRef]
  19. Khosravifar, S.; Schwartz, C.W.; Goulias, D.G. Mechanistic structural properties of foamed asphalt stabilised base materials. Int. J. Pavement Eng. 2015, 16, 27–38. [Google Scholar] [CrossRef]
  20. Kim, Y.; Lee, H.D. Performance Evaluation of Cold In-place Recycling Mixtures Using Emulsified Asphalt Based on Dynamic Modulus, Flow Number, Flow Time, and Raveling Loss. KSCE J. Civ. Eng. 2012, 16, 586–593. [Google Scholar] [CrossRef]
  21. Fu, P.; Jones, D.; Harvey, J.T.; Bukhari, S.A. Laboratory Test Methods for Foamed Asphalt Mix Resilient Modulus. Road Mater. Pavement Des. 2009, 10, 188–212. [Google Scholar] [CrossRef]
  22. Khosravifar, S.; Schwartz, C.W.; Goulias, D.G. Foamed Asphalt Stabilized Base: A Case Study. In Airfield and Highway Pavement 2013: Sustainable and Efficient Pavements; ASCE: Reston, VA, USA, 2013; pp. 106–117. [Google Scholar]
  23. Ramanujam, J.M.; Jones, J.D. Characterization of foamed-bitumen stabilization. Int. J. Pavement Eng. 2007, 8, 111–122. [Google Scholar] [CrossRef]
  24. Louw, S.; Wu, R.; Hammack, J.; Jones, D. Development of an Improved Test Setup for Measuring the Resilient Modulus of Stabilized Pavement Materials. Transp. Res. Rec. 2019, 2673, 304–313. [Google Scholar] [CrossRef]
  25. Loizos, A.; Papavasiliou, V. Effect of Temperature Fluctuations on the Bearing Capacity of Cold In-Depth Recycled Pavements. Sustainability 2022, 14, 426. [Google Scholar] [CrossRef]
  26. Gkyrtis, K.; Plati, C.; Loizos, A. Mechanistic Analysis of Asphalt Pavements in Support of Pavement Preservation Decision-Making. Infrastructures 2022, 7, 61. [Google Scholar] [CrossRef]
  27. Marecos, V.; Fontul, S.; Antunes, M.L.; Solla, M. Evaluation of a highway pavement using non-destructive tests: FallingWeight Deflectometer and Ground Penetrating Radar. Constr. Build. Mater. 2017, 154, 1164–1172. [Google Scholar] [CrossRef]
  28. WSDOT. Everseries User’s Guide—Pavement Analysis Computer Software and Case Studies; Washington State Department of Transportation: Washington, DC, USA, 2005. [Google Scholar]
  29. Pan, E.; Chen, E.; Alkasawneh, W. Layered flexible pavement studies: Challenges in forward and inverse problems. Int. J. Pavement Res. Technol. 2008, 1, 12–16. [Google Scholar]
  30. Hakim, B.; Brown, S.F. Pavement Analysis Using the FWD: Practical Difficulties and Proposed Simplification. In Proceedings of the 10th International Conference on Asphalt Pavement, Quebec, QC, Canada, 12–17 August 2006; pp. 59–68. [Google Scholar]
  31. Horak, E. Benchmarking the structural condition of flexible pavements with deflection bowl parameters. J. S. Afr. Inst. Civ. Eng. 2008, 50, 2–9. [Google Scholar]
  32. Van-Geem, C.; Nigro, P.; Berlemont, B. The Use of Deflection Measurements in Pavement Management of the Primary Road Network of Wallonia, Belgium. In Proceedings of the 9th International Conference on Managing Pavement Assets (ICMPA), Washington, DC, USA, 18–21 May 2015. [Google Scholar]
  33. Kavussi, A.; Abbasghorbani, M.; Moghadas-Nejad, F.; Bamdad-Ziksari, A. A new method to determine maintenance and repair activities at network level pavement management using falling weight Deflectometer. J. Civ. Eng. Manag. 2017, 23, 338–346. [Google Scholar] [CrossRef]
  34. Cong, L.; Lytton, R.L.; Xie, Z. Sublayer Strength Evaluation with FWD in Semi-rigid Base Asphalt Pavement Rehabilitation Project. In Proceedings of the GeoShanghai International Conference, Shanghai, China, 3–5 June 2010; pp. 392–397. [Google Scholar]
  35. Cui, X.; Dong, Q.; Ni, F.; Liang, X. Evaluation of Semi-rigid Base Performance Through Numerical Simulation and Data Mining of Pavement Deflection Basin. In Proceedings of the GeoShanghai 2018 International Conference: Transportation Geotechnics and Pavement Engineering, Singapore, 27–30 May 2018; pp. 364–371. [Google Scholar]
  36. Elbagalati, O.; Elseifi, M.A.; Gaspard, K.; Zhang, Z. Implementation of the Structural Condition Index into the Louisiana Pavement Management System Based on Rolling Wheel Deflectometer Testing. Transp. Res. Rec. 2017, 2641, 39–47. [Google Scholar] [CrossRef]
  37. Goktepe, S.; Agar, E.; Lav, A.H. Advances in backcalculating the mechanical properties of flexible pavements. Adv. Eng. Softw. 2006, 37, 421–431. [Google Scholar] [CrossRef]
  38. Guzina, B.B.; Osburn, R.H. Effective Tool for Enhancing Elastostatic Pavement Diagnosis. Transp. Res. Rec. 2002, 1806, 30–37. [Google Scholar] [CrossRef]
  39. Akbarzadeh, H.; Bayat, A.; Soleymani, H.R. Analytical Review of the HMA Temperature Correction Factors from Laboratory and Falling Weight Deflectometer Tests. Int. J. Pavement Res. Technol. 2012, 5, 30–39. [Google Scholar]
  40. Zang, G.; Li, L.; Chen, Z.; Sun, L. A Nondestructive Evaluation Method for Semi-Rigid Base Cracking Condition. In Proceedings of the 96th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 8–12 January 2017. [Google Scholar]
  41. Georgouli, K.; Pomoni, M.; Cliatt, B.; Loizos, A. A simplified approach for the estimation of HMA dynamic modulus for in service pavements. In Proceedings of the 6th International Conference on Bituminous Mixtures and Pavements (ICONFBMP), Thessaloniki, Greece, 10–12 June 2015; pp. 661–670. [Google Scholar]
  42. Fwa, T.F.; Tan, C.Y.; Chan, W.T. Backcalculation analysis of pavement layer moduli using genetic algorithms. Transp. Res. Rec. 1997, 1570, 134–142. [Google Scholar] [CrossRef]
  43. Amaranatha-Reddy, M.; Sudhakar-Reddy, K.; Pandey, B.B. Selection of Genetic Algorithm Parameters for Backcalculation of Pavement Moduli. Int. J. Pavement Eng. 2004, 5, 81–90. [Google Scholar] [CrossRef]
  44. Plati, C.; Georgouli, K.; Cliatt, B.; Loizos, A. Incorporation of GPR data into genetic algorithms for assessing recycled pavements. Constr. Build. Mater. 2017, 154, 1263–1271. [Google Scholar] [CrossRef]
  45. ASTM D6931-17; Standard Test Method for Indirect Tensile (IDT) Strength of Asphalt Mixtures. ASTM International: West Conshohocken, PA, USA, 2017.
  46. ASTM D1074-17; Standard Test Method for Compressive Strength of Asphalt Mixtures. ASTM International: West Conshohocken, PA, USA, 2017.
  47. Loizos, A.; Papavasiliou, V.; Plati, C. Investigating in situ stress-dependent behaviour of foamed asphalt-treated pavement materials. Road Mater. Pavement Des. 2012, 13, 678–690. [Google Scholar] [CrossRef]
  48. Ji, R.; Siddiki, N.; Nantung, T.; Kim, D. Evaluation of resilient modulus of subgrade and base materials in Indiana and its implementation in MEPDG. Sci. World J. 2014, 2014, 372838. [Google Scholar] [CrossRef] [Green Version]
  49. Horak, E.; Emery, S. Falling Weight Deflectometer Bowl Parameters as Analysis Tool for Pavement Structural Evaluations. In Proceedings of the 22nd ARRB conference: Research into Practice: Proceedings, Canberra, Australia, 29 October–2 November 2006. [Google Scholar]
Sustainability 15 03622 g001 550
Figure 1. FWD testing: (a) a typical array and (b) a typical deflection bowl.
Figure 1. FWD testing: (a) a typical array and (b) a typical deflection bowl.
Sustainability 15 03622 g001
Sustainability 15 03622 g002 550
Figure 2. (a) Initial pavement section and (b) rehabilitated pavement section.
Figure 2. (a) Initial pavement section and (b) rehabilitated pavement section.
Sustainability 15 03622 g002
Sustainability 15 03622 g003 550
Figure 3. Outline of research methodology (AC: Asphalt Concrete, FA: Foamed Asphalt, CBM: Cement Bound Material, NDT: Non-Destructive Testing, FWD: Falling Weight Deflectometer, GPR: Ground Penetrating Radar).
Figure 3. Outline of research methodology (AC: Asphalt Concrete, FA: Foamed Asphalt, CBM: Cement Bound Material, NDT: Non-Destructive Testing, FWD: Falling Weight Deflectometer, GPR: Ground Penetrating Radar).
Sustainability 15 03622 g003
Sustainability 15 03622 g004 550
Figure 4. (a) Average deflection basins, (b) box plots for AC temperatures.
Figure 4. (a) Average deflection basins, (b) box plots for AC temperatures.
Sustainability 15 03622 g004
Sustainability 15 03622 g005 550
Figure 5. Box plots of FA moduli: (a) Period A and (b) Period B.
Figure 5. Box plots of FA moduli: (a) Period A and (b) Period B.
Sustainability 15 03622 g005
Sustainability 15 03622 g006 550
Figure 6. Correlations between back-calculated FA moduli: (a) Period A and (b) Period B (FA: Foamed Asphalt, EEVER: modulus from the EVERCALC tool, EGEN: modulus from the GENETIC tool).
Figure 6. Correlations between back-calculated FA moduli: (a) Period A and (b) Period B (FA: Foamed Asphalt, EEVER: modulus from the EVERCALC tool, EGEN: modulus from the GENETIC tool).
Sustainability 15 03622 g006
Sustainability 15 03622 g007 550
Figure 7. Values of R2 for Foamed Asphalt modulus prediction based on Deflection-Based Parameters (MLET: Multi-Layer Elastic Theory, GA: Genetic Algorithms).
Figure 7. Values of R2 for Foamed Asphalt modulus prediction based on Deflection-Based Parameters (MLET: Multi-Layer Elastic Theory, GA: Genetic Algorithms).
Sustainability 15 03622 g007
Sustainability 15 03622 g008 550
Figure 8. Correlations between Deflection-Based Parameter and FA modulus from the GA: (a) Period A and (b) Period B (FA: Foamed Asphalt, GA: Genetic Algorithms, EGEN: modulus from the GENETIC tool).
Figure 8. Correlations between Deflection-Based Parameter and FA modulus from the GA: (a) Period A and (b) Period B (FA: Foamed Asphalt, GA: Genetic Algorithms, EGEN: modulus from the GENETIC tool).
Sustainability 15 03622 g008
Table
Table 1. Typical DBPs.
Table 1. Typical DBPs.
No.IndexesMathematical ExpressionComments
1Central (maximum) deflection
(d0-μm)		-Overall pavement condition
2Surface Curvature Index
(SCI-μm)		d0 − d300Surface layer condition
3Base Damage Index
(BDI-μm)		d300 − d600Surface and intermediate layers condition
4Base Curvature Index
(BCI-μm)		d600 − d900Intermediate layers condition
5AREA parameter
(AREA-dimensionless)		6(d0 + 2d300 + 2d600 + d900)/d0Overall pavement condition
6Area Under Pavement Profile
(AUPP-μm)		0.5(5d0 − 2d300 − 2d600 − d900)Upper layers condition
7Deflection at the outer geophone
(d1800-μm)		-Subgrade condition
Table
Table 2. Overview of back-analysis results for Foamed Asphalt base layer.
Table 2. Overview of back-analysis results for Foamed Asphalt base layer.
Statistics for FA ModulusPeriod APeriod B
EEVEREGENEEVEREGEN
Min modulus (MPa)500848500765
Median modulus (MPa)1633142723172592
Max modulus (MPa)3711247587087267
Average modulus (MPa)1833149628442849
CV (%)45.3%28.0%80.2%61.4%
Table
Table 3. Results from t-test for back-calculated moduli.
Table 3. Results from t-test for back-calculated moduli.
ParametersPeriod APeriod B
Variable 1
EEVER		Variable 2 EGENVariable 1
EEVER		Variable 2 EGEN
Mean (MPa)1833149628442849
Variance688,248175,5165,205,3513,055,880
Observations37373737
df72 72
t Stat2.2070 −0.0108
p (T ≤ t) one-tail0.0153 0.4957
t Critical one-tail1.6663 1.6663
p (T ≤ t) two-tail0.0305 0.9914
t Critical two-tail1.9934 1.9934
Table
Table 4. Results from t-test for DBP-predicted moduli.
Table 4. Results from t-test for DBP-predicted moduli.
Period A—GENETICPeriod B—GENETICPeriod B—EVERCALC
Index: d200 − d450Variable 1: EGEN-AVariable 2: Ed200 − d450Variable 1: EGEN-BVariable 2:
Ed200 − d450		Variable 1: EEVER-BVariable 2:
Ed200 − d450
Mean E (MPa)149614812849280428442532
Variance175,516125,2523,055,8802,627,6715,205,3512,412,400
Observations373737373737
df72 72 72
t Stat0.1661 0.1159 0.6865
p (T ≤ t) one-tail0.4343 0.4540 0.2473
t Critical one-tail1.6663 1.6663 1.6663
p (T ≤ t) two-tail0.8685 0.9080 0.4946
t Critical two-tail1.9935 1.9935 1.9935
Index: d200 − d600Variable 1: EGEN-AVariable 2: Ed200 − d600Variable 1: EGEN-BVariable 2:
Ed200 − d600		Variable 1: EEVER-BVariable 2:
Ed200 − d600
Mean E (MPa)149614802849279428442516
Variance175,516124,3603,055,8802,553,6815,205,3512,296,038
Observations373737373737
df72 72 72
t Stat0.1761 0.1405 0.7286
p (T ≤ t) one-tail0.4303 0.4443 0.2343
t Critical one-tail1.6663 1.6663 1.6663
p (T ≤ t) two-tail0.8607 0.8887 0.4686
t Critical two-tail1.9935 1.9935 1.9935
Index: d300 − d450Variable 1: EGEN-AVariable 2: Ed300 − d450Variable 1: EGEN-BVariable 2:
Ed300 − d450		Variable 1: EEVER-BVariable 2:
Ed300 − d450
Mean E (MPa)149614802849280428442562
Variance175,516125,3233,055,8802,940,4105,205,3512,963,207
Observations373737373737
df72 72 72
t Stat0.1749 0.1126 0.5991
p (T ≤ t) one-tail0.4308 0.4553 0.2755
t Critical one-tail1.6663 1.6663 1.6663
p (T ≤ t) two-tail0.8617 0.9107 0.5510
t Critical two-tail1.9935 1.9935 1.9935
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Gkyrtis, K.; Plati, C.; Loizos, A. Structural Performance of Foamed Asphalt Base in a Full Depth Reclaimed and Sustainable Pavement. Sustainability 2023, 15, 3622. https://doi.org/10.3390/su15043622

AMA Style

Gkyrtis K, Plati C, Loizos A. Structural Performance of Foamed Asphalt Base in a Full Depth Reclaimed and Sustainable Pavement. Sustainability. 2023; 15(4):3622. https://doi.org/10.3390/su15043622

Chicago/Turabian Style

Gkyrtis, Konstantinos, Christina Plati, and Andreas Loizos. 2023. "Structural Performance of Foamed Asphalt Base in a Full Depth Reclaimed and Sustainable Pavement" Sustainability 15, no. 4: 3622. https://doi.org/10.3390/su15043622

APA Style

Gkyrtis, K., Plati, C., & Loizos, A. (2023). Structural Performance of Foamed Asphalt Base in a Full Depth Reclaimed and Sustainable Pavement. Sustainability, 15(4), 3622. https://doi.org/10.3390/su15043622

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop