This study deals with the recycling of Recycled Asphalt Pavement (RAP) and Reclaimed Asphalt Shingles (RAS) in asphalt and it is focused at binder scale. The aim is to investigate the physico – chemical and rheological properties of the two basic binders i.e. RAP’s and RAS’s binder and to examine how they change in a virgin binder blended with these two recycled binders in increasing percentage. Moreover, the applicability of a model to link bitumen and asphalt mix mechanical behavior is explored. Asphaltanalyzer and rotary evaporator are used to recover bitumen from RAP and RAS, a specific procedure for RAS extraction is successfully implemented. Twelve mixes with increasing percentage of recycled binders are realized according to Reclaimed Binder Ratio (RBR) of the corresponding asphalt mixes. A specific procedure for manually mixing the three binders together is proposed. A maximum of 6% wt. of RAS binder is blended together with virgin binder, while 40% wt. maximum for RAP binder and for the mixes with RAP and RAS together. Needle Penetration and Ring and Ball tests are carried out on the basic binders and on the mixes according with European characterization. The variation of results in the mixes is studied. The viscoelastic behavior of virgin binder, RAP binder and of all the mixes is studied by performing complex modulus tests at different temperatures and frequencies using Dynamic Shear Rheometer (DSR). Rheological tests are carried out also on RAS binder alone but without success due to its very hard nature. The shift in temperature of virgin binder master curve due to the hardening effect of recycling binders has been quantified. A comparison of the Viscoelastic behavior in term of Complex modulus and phase angle between mixes and virgin binder is carried out considering the Complex Coefficient of Mobilization (CCM). Fourier Transform Infrared Spectroscopy (FTIR) tests have been carried out to study the aging condition of RAP and RAS binders and their aging effect in the mix containing virgin binder also. C=O and S=O indexes have been calculated considering the internal procedure developed at IRC. Glass transition temperature (Tg) of the basic binders and the mixes has been determined by Differential Scanning Calorimetry (DSC) machine, with classic test and modulated too. The trend in the Tg’s of the mixes is studied. A blending equation has been proposed and adapted to this study by adding a third term for considering also the contribution of the third bitumen in the mix. It is proposed linear for the time independent parameters, while linear in logarithmic scale for time dependent parameters. The applicability of this blending law has been verified for the European test results translating probably the homogeneity of the realized mixes. The equation has been also applied to the chemical test results i.e. FTIR and DSC. It shows good prediction for FTIR results (especially for C=O index) translating perhaps the chemical compatibility of the binders and quite good results for modulated DSC. The viscoelastic mechanical behavior of virgin and RAP binders and of all the mixes have been modelled by using 2S2P1D model. At first normalized curves are realized to study the shape parameters variation with respect to virgin binder, then the parameters of the model have been calibrated for all the mixes. The applicability of the same blending law as before was successfully verified for describing the 2S2P1D parameters variation in mixes with increasing percentage of RAP binder. Finally, a transformation for linking the viscoelastic properties of binders and those of asphalt mixes have been considered. It has been investigated the virgin binder and virgin asphalt mix. A lack of superposition between the experimental value of virgin asphalt mix and the predicted value by the transformation has been found. The reason for this discrepancy is probably related to a limitation of DSR device at low temperature. Two different approaches have been proposed to overcome this issue. A first approach in which it is defined and calibrated a factor of increasing of the glass modulus G0 obtained by DSR machine. The second approach consists in recalibrate the parameters of the model 2S2P1D of virgin binder by considering the same shape parameters of the virgin asphalt mix and modifying the other parameters to have an acceptable superposition. The applicability of these two approaches has been successfully verified for virgin asphalt mix. The factor α within the equation of the transformation, probably relates to the considered mix design or aging during mixing, has been calculated too.
Questo studio riguarda il riciclaggio di Recycled Asphalt Pavement (RAP) e Reclaimed Asphalt Shingles (RAS) nei conglomerati bituminosi e si focalizza su scala legante. L’obbiettivo è si investigare dal punto di vista fisico – chimico e reologico le proprietà dei due leganti di base i.e. legante da RAP e da RAS e di esaminare come tali proprietà cambiano in un bitume vergine mescolato (blending) con questi due leganti riciclati in percentuale crescente. Inoltre, è stata esplorata l’applicabilità di un modello che lega il comportamento meccanico del bitume con quello del corrispondente conglomerato bituminoso. Sono stati utilizzati l’asphaltanalizzatore e evaporatore rotativo per recuperare il bitume da RAP e RAS, ed è stata implementata con successo una procedura specifica per l’estrazione del RAS. Sono stati realizzati dodici mix con percentuali crescenti di riciclato considerando la frazione di bitume riciclato (RBR) dei corrispondenti conglomerati bituminosi. È stata proposta una procedura specifica per mescolare manualmente i tre leganti. Un massimo pari a 6% in peso di bitume da RAS è stato mescolato con il bitume vergine, mentre 40% massimo in peso per il bitume da RAP e per i mix realizzati con i due bitumi insieme. Sono stati condotti test di penetrazione all’ago e palla anello sia sui bitumi di base, sia sui vari mix bituminosi in accordo con la caratterizzazione europea. È stata studiata la variazione dei risultati nei mix. Si è studiato il comportamento viscoelastico del bitume vergine, del bitume da RAP e di tutti i vari mix conducendo test di modulo complesso a differenti temperature e frequenze utilizzando il reometro a taglio dinamico (DSR). Sono stati effettuati test reologici anche sul bitume da RAS da solo ma senza successo a causa della natura molto dura di questo bitume. È stato quantificato lo spostamento in temperatura della curva maestra del bitume vergine a causa dell’effetto indurente dell’aggiunta di bitume riciclato. È stata condotta una comparativa del comportamento viscoelastico in termini di modulo complesso e angolo di fase tra i vari mix e il bitume vergine considerando il coefficiente complesso di mobilitazione (CCM). Sono stati condotti test usando la spettroscopia infrarossa a trasformata di Fourier (FTIR) per studiare le condizioni di invecchiamento dei bitumi da RAP e da RAS e il loro effetto di invecchiamento nei vari mix con anche bitume vergine all’interno. Sono stati calcolati gli indici C=O and S=O considerando una procedura interna sviluppata presso l’istituto di ricerca in costruttibilità (IRC). È stata determinata la temperatura di transizione vetrosa (Tg) dei bitumi di base e dei vari mix utilizzando la macchina della calorimetria differenziale a scansione (DSC), con test classici e modulati. È stato studiato anche il trend nella variazione delle Tg dei vari mix. È stata proposta un’equazione di blending e adattata a questo studio aggiungendo un terzo termine per tenere conto anche del contributo del terzo bitume nel mix. Tale equazione è stata proposta in forma lineare per i parametri indipendenti dalla temperatura, mentre in forma lineare in scala logaritmica per i parametri dipendenti dalla temperatura. È stata verificata l’applicabilità di questa legge di blending per i risultati dei test europei a significare probabilmente l’omogeneità dei mix bituminosi realizzati. L’equazione è stata applicata anche ai test chimici i.e. FTIR e DSC. Essa ha mostrato una buona predizione per i risultati della FTIR (soprattutto per quanto riguarda l’indice C=O) traducendo forse la compatibilità chimica dei bitumi e ha mostrato risultati abbastanza buoni per il DSC modulato. Il comportamento meccanico del bitume vergine, del bitume da RAP e di tutti i mix è stato modellizzato utilizzando il modello 2S2P1D. Dapprima sono state costruite le curve normalizzate per studiare la variazione dei parametri di forma rispetto a quelli del bitume vergine, dopo si sono potuti calibrare i parametri del modello per tutti i vari mix. È stata verificata con successo l’applicabilità della stessa legge di blending proposta prima per descrivere la variazione dei parametri del modello 2S2P1D nei mix contenenti percentuali crescenti di bitume da RAP. Infine, è stata considerata una trasformazione per legare le proprietà viscoelastiche dei bitumi e quelle dei conglomerati bituminosi. Si è considerato il bitume vergine e il conglomerato bituminoso vergine. È stata riscontrata una mancanza di sovrapposizione tra i valori sperimentali del conglomerato bituminoso vergine e i valori predetti dalla trasformazione. La ragione di tale discrepanza è probabilmente legata a un limite della macchina DSR alle basse temperature. Sono stati proposti due approcci per superare questa problematica. Un primo approccio nel quale viene definito e calibrato un fattore di incremento del modulo vetroso G0 ottenuto dalla macchina DSR. Un secondo approccio che consiste nel ricalibrare i parametri del modello 2S2P1D del bitume vergine considerando gli stessi parametri di forma del conglomerato bituminoso vergine e modificando gli altri parametri per avere una sovrapposizione accettabile. È stata verificata con successo l’applicabilità di questi due approcci per il conglomerato bituminoso vergine. È stato anche calcolato il fattore α presente all’interno dell’equazione di trasformazione, legato probabilmente allo specifico mix design considerato o all’invecchiamento durante la fase di mixing.
Physicochemical and rheological characterisation of bituminous binders recovered from reclaimed asphalt pavement (RAP) and reclaimed asphalt shingles (RAS). Modelling for predicting the behaviour of the final recycled asphalt mixture
GUGLIELMI, MATTEO
2018/2019
Abstract
This study deals with the recycling of Recycled Asphalt Pavement (RAP) and Reclaimed Asphalt Shingles (RAS) in asphalt and it is focused at binder scale. The aim is to investigate the physico – chemical and rheological properties of the two basic binders i.e. RAP’s and RAS’s binder and to examine how they change in a virgin binder blended with these two recycled binders in increasing percentage. Moreover, the applicability of a model to link bitumen and asphalt mix mechanical behavior is explored. Asphaltanalyzer and rotary evaporator are used to recover bitumen from RAP and RAS, a specific procedure for RAS extraction is successfully implemented. Twelve mixes with increasing percentage of recycled binders are realized according to Reclaimed Binder Ratio (RBR) of the corresponding asphalt mixes. A specific procedure for manually mixing the three binders together is proposed. A maximum of 6% wt. of RAS binder is blended together with virgin binder, while 40% wt. maximum for RAP binder and for the mixes with RAP and RAS together. Needle Penetration and Ring and Ball tests are carried out on the basic binders and on the mixes according with European characterization. The variation of results in the mixes is studied. The viscoelastic behavior of virgin binder, RAP binder and of all the mixes is studied by performing complex modulus tests at different temperatures and frequencies using Dynamic Shear Rheometer (DSR). Rheological tests are carried out also on RAS binder alone but without success due to its very hard nature. The shift in temperature of virgin binder master curve due to the hardening effect of recycling binders has been quantified. A comparison of the Viscoelastic behavior in term of Complex modulus and phase angle between mixes and virgin binder is carried out considering the Complex Coefficient of Mobilization (CCM). Fourier Transform Infrared Spectroscopy (FTIR) tests have been carried out to study the aging condition of RAP and RAS binders and their aging effect in the mix containing virgin binder also. C=O and S=O indexes have been calculated considering the internal procedure developed at IRC. Glass transition temperature (Tg) of the basic binders and the mixes has been determined by Differential Scanning Calorimetry (DSC) machine, with classic test and modulated too. The trend in the Tg’s of the mixes is studied. A blending equation has been proposed and adapted to this study by adding a third term for considering also the contribution of the third bitumen in the mix. It is proposed linear for the time independent parameters, while linear in logarithmic scale for time dependent parameters. The applicability of this blending law has been verified for the European test results translating probably the homogeneity of the realized mixes. The equation has been also applied to the chemical test results i.e. FTIR and DSC. It shows good prediction for FTIR results (especially for C=O index) translating perhaps the chemical compatibility of the binders and quite good results for modulated DSC. The viscoelastic mechanical behavior of virgin and RAP binders and of all the mixes have been modelled by using 2S2P1D model. At first normalized curves are realized to study the shape parameters variation with respect to virgin binder, then the parameters of the model have been calibrated for all the mixes. The applicability of the same blending law as before was successfully verified for describing the 2S2P1D parameters variation in mixes with increasing percentage of RAP binder. Finally, a transformation for linking the viscoelastic properties of binders and those of asphalt mixes have been considered. It has been investigated the virgin binder and virgin asphalt mix. A lack of superposition between the experimental value of virgin asphalt mix and the predicted value by the transformation has been found. The reason for this discrepancy is probably related to a limitation of DSR device at low temperature. Two different approaches have been proposed to overcome this issue. A first approach in which it is defined and calibrated a factor of increasing of the glass modulus G0 obtained by DSR machine. The second approach consists in recalibrate the parameters of the model 2S2P1D of virgin binder by considering the same shape parameters of the virgin asphalt mix and modifying the other parameters to have an acceptable superposition. The applicability of these two approaches has been successfully verified for virgin asphalt mix. The factor α within the equation of the transformation, probably relates to the considered mix design or aging during mixing, has been calculated too.File | Dimensione | Formato | |
---|---|---|---|
Matteo Guglielmi Final Report.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
8.93 MB
Formato
Adobe PDF
|
8.93 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/150774