materials Article
Evaluating and Modeling the Internal Diffusion Behaviors of Microencapsulated Rejuvenator in Aged Bitumen by FTIR-ATR Tests Junfeng Su 1,2, *, Yingyuan Wang 2 , Peng Yang 3 , Shan Han 2 , Ningxu Han 4, * and Wei Li 1 1 2 3 4
*
Tianjin Key Laboratory of Advanced Fibers and Energy Storage, Tianjin Polytechnic University, Tianjin 300387, China;
[email protected] Department of Polymer Materials, School of Material Science and Engineering, Tianjin Polytechnic University, Tianjin 300387, China;
[email protected] (Y.W.);
[email protected] (S.H.) School of Navigational Engineering, Guangzhou Maritime Institute, Guangzhou 510725, China;
[email protected] Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen 518060, China Correspondence:
[email protected] (J.S.);
[email protected] (N.H.); Tel./Fax: +86-22-2621-0595 (J.S.); +86-755-2653-7094 (N.H.)
Academic Editor: Jorge de Brito Received: 29 September 2016; Accepted: 11 November 2016; Published: 17 November 2016
Abstract: Microencapsulated rejuvenator has been attracted much attention for self-healing bitumen. The diffusion coefficient is one of the key parameters to estimate the feasibility of rejuvenator to age bitumen. The objective of this research was to evaluate diffusion behaviors of microencapsulated rejuvenator in aged bitumen by a FTIR-ATR method. Various microcapsule samples were mixed in bitumen to form thin films. The core material of microcapsules used as rejuvenator was diphenylsilane (DPS), its fairly specific absorption band at 843 cm−1 was selected as a marker band to calculate the diffusion coefficient (D). The microstructure parameters, including contents, mean size and mean shell thickness of microcapsules, were considered to understand the diffusion behaviors under different temperatures (20, 30, 40 and 50 ◦ C) in bitumen. The results showed that a larger mean size of microcapsules did not greatly affect the D values under the same temperature. In contrast, a higher mean shell thickness decreased the D values because of the decrement of damage probability of microcapsules under the same content. With the same microcapsule sample in bitumen, the D values presented a trend of linear increase when the content of microcapsules was increased. All these results indicated that the microstructure affected the diffusion behaviors based on the concentration of released rejuvenator. A preliminary model of diffusion behaviors of microencapsulated rejuvenator in bitumen was given based on the Arrhenius equation considering the microstructure of microcapsules, the amount of released rejuvenator and the age degree of bitumen. This model may be a guide to the construction and application of self-healing bitumen using microcapsules. Keywords: diffusion behaviors; bitumen; self-healing; microcapsule; rejuvenator
1. Introduction Bitumen can be defined as a self-healing material because it has the potential to restore stiffness and strength by closing the microcracks which occur when the pavement is subjected to traffic loads or under high temperature [1]. Its self-healing capability will decrease or disappear as the ageing problem of bitumen leads to pavement failure after years of usage including surface raveling and reflective cracking [2]. Rejuvenator use may be the only one method that can restore the original properties of the pavements because rejuvenating agents have the capability of reconstituting the binder’s chemical Materials 2016, 9, 932; doi:10.3390/ma9110932
www.mdpi.com/journal/materials
Materials 2016, 9, 932
2 of 15
composition and consist of lubricating and extender oils containing a high proportion of maltene constituents [3]. To overcome the poor penetration ability of oily rejuvenator through the surface of bitumen, microencapsulated rejuvenator in situ use in bitumen may be an alternative approach [4]. In previous study, a method has been reported to fabricate microcapsules containing rejuvenator utilizing methanol modified melamine-formaldehyde (MMF) resin as shell material [5–10]. This product is believed to be an environmental-friendly powder encapsulating suitable size rejuvenator for chemical engineering and construction engineering [4]. During the aging process of bitumen, microcapsules could be punctured by microcracks and leaked the oily-liquid rejuvenator into microcracks [8]. The mechanism is that a crack repair in an asphalt pavement system attributes to the wetting and inter-diffusion of material between the two surfaces of a microcrack to achieve properties of the original material [10]. With the help of capillarity, the rejuvenator filled the cracks with a movement speed mainly determined by the volume of microcapsules in bitumen [8]. A diffusion phenomenon was also observed by using a fluorescence microscope [8–10]. During the diffusion behaviors, the small molecules will insert into the macromolecules of bitumen, and then the microstructure will be changed. It was found that the diffusion behaviors made a role of determining the microstructure of bitumen [9]. Rejuvenating products are designed to restore the original characteristics to oxidised (aged) bitumen binders in order to soften the aged binder and create a broad-spectrum rejuvenation that replenishes the volatiles and dispersing oils while promoting adhesion. Therefore, diffusion is an important concept considering processes in bituminous binders such as oxidative ageing, stripping and rejuvenator in asphalt recycling [11]. A literature review shows that diffusion in bitumen has attracted much research. A number of issues related to diffusion in bitumen have been investigated by Karlsson et al. [12]. For example, consequences of Fick’s law governing the diffusion process were demonstrated, which, among other things, indicated that the time needed for a diffusion process to occur was proportional to the square of the binder layer thickness [13]. Factors influencing diffusion, such as the temperature, ageing and refining of bitumen (bitumen obtained from one and the same crude oil from various degrees of distillation) have also been studied [13]. Normally, the diffusing capability of rejuvenator into aged bitumen can be enhanced with the increasing of temperature and time, however, the diffusing of rejuvenator into aged bitumen is restricted due to the volatilization of light component and aging of rejuvenator under high temperature [11–13]. At the same time, a few methods have been reported to measure the diffusion rules in bitumen. For example, atomic force microscope (AFM) and gel permeation chromatography (GPC) were used to analyze the diffusion mechanism of different kinds of rejuvenator [14]. The result showed the rejuvenator influence gradually weakened when the diffusion depth grew. Adsorbed rejuvenator in age asphalt hindered the diffusion of large molecular modifying component by dispersing the asphaltenes. The diffusion level of the modifying rejuvenator closed to that of low viscosity regeneration. These research results provide technical support to the regeneration road performance. In our previous work [9,10], capillarity and diffusion behaviors of rejuvenator in aged bitumen were observed by a fluorescence microscope using light characters such as reflection, diffraction and refraction. The scale of fluorescence microscope was applied to measure the size of capillarity and diffusion. As bitumen is a temperature-sensitive material, the observation is in an environment of 0 ◦ C temperature. The defect of this testing method is its low precision limited by observer vision. Fourier transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR) was used to monitor the diffusion of selected well-defined substances [12]. The penetration depth of rejuvenator seal materials on hot asphalt mixtures was studied by means of rheology test and FTIR test. Karlsson [13] proved that the FTIR-ATR was a reliable method to measure the rejuvenator diffusion in bitumen. The advantages of FTIR-ATA method include the in situ measurement, non-destructive measurement and real-time measurement. Although several methods have been used to measure diffusion behaviors of rejuvenator into bitumen, little knowledge can be applied to monitor the internal diffusion principle of microencapsulated rejuvenator in bitumen. Figure 1 illustrates the diffusion process of microencapsulated rejuvenator into bitumen. The diffusion dynamics is the concentration difference
Materials 2016, 2016, 9, 9, 932 932 Materials
3 of 15
microencapsulated rejuvenator into bitumen. The diffusion dynamics is the concentration difference of of rejuvenator rejuvenator material material in in bitumen. bitumen. One One reason reason is is that that it it is is difficult difficult to to measure measure the the internal internal diffusion diffusion occurring in microcracks with a micrometer size. Another reason is that it is hard to measure occurring in microcracks with a micrometer size. Another reason is that it is hard to measure the the diffusion behaviors because of less mass of rejuvenator in microcapsules. Furthermore, the diffusion behaviors because of less mass of rejuvenator in microcapsules. Furthermore, the microstructure rejuvenator microstructure of of microcapsules microcapsules (shell (shell thickness thickness and and mean mean size) size) will will greatly greatly influence influence the the rejuvenator movements [6]. It has been hypothesized that the temperature and microstructure of bitumen may be movements [6]. It has been hypothesized that the temperature and microstructure of bitumen may critical for for thethe diffusion process, as diffusion is aismolecule’s movement, butbut it is yetyet clear how it be critical diffusion process, as diffusion a molecule’s movement, it not is not clear how may affect it. it may affect it.
Figure 1. Illustration of the diffusion process of microencapsulated rejuvenator into bitumen. Figure 1. Illustration of the diffusion process of microencapsulated rejuvenator into bitumen.
Based on the above analysis, the objective of this research is to evaluate diffusion behaviors of Based on the above analysis, objective of this is tomethod. evaluate Various diffusioncontents behaviors microencapsulated rejuvenator in the aged bitumen by aresearch FTIR-ATR of of microencapsulated rejuvenator in aged bitumen by a FTIR-ATR method. Various contents of microcapsules were mixed in bitumen to form thin film samples. Liquid nitrogen was used to trigger microcapsules were mixedsamples. in bitumen to form thin film samples. Liquid wasdiffusion used to a microcrack in bitumen Fluorescence microscope was used to nitrogen observe the trigger a microcrack in bitumen samples. Fluorescence microscope was used to observe the diffusion behaviors. FTIR-ATR was applied to measure the diffusion rates affected by time, temperature, behaviors. FTIR-ATR applied to of measure theAdiffusion rates affectedmodel by time, microcapsules content, was and age degree bitumen. preliminary diffusion was temperature, given for the microcapsules content, and age degree of bitumen. A preliminary diffusion model was given the microcapsules/bitumen samples considering the parameters of microstructure, timeforand microcapsules/bitumen samples considering the parameters of microstructure, time and temperature. temperature. The diffusion principles are expected to guide the optimization of this intelligent The diffusion principles are expected to guide the optimization of this intelligent material systems material systems and structures. and structures. 2. Experimental Method 2. Experimental Method 2.1. Materials Materials 2.1. The shell shell material material of of microcapsules microcapsules was was commercial commercial prepolymer prepolymer of of melamine-formaldehyde melamine-formaldehyde The modified by methanol (solid content was 78.0%) purchased from Aonisite Chemical modified by methanol (solid content was 78.0%) purchased from Aonisite Chemical Trade Trade Co., Co., Ltd. Ltd. ® (Tianjin, China). Styrene maleic anhydride (SMA) copolymer (Scripset 520, Hercules, CA, USA) was (Tianjin, China). Styrene maleic anhydride (SMA) copolymer (Scripset® 520, Hercules, CA, USA) applied as dispersant [4]. Diphenylsilane (DPS) (Tianjin Chem., Tianjin, China) was used as was applied as dispersant [4]. Diphenylsilane (DPS) (Tianjin Chem., Tianjin, China) was used as rejuvenator, which was a low viscosity and colorless liquid. The bitumen used in this study obtained rejuvenator, which was a low viscosity and colorless liquid. The bitumen used in this study obtained from Qilu Petrochemical (Linzi, China). The aged bitumen samples with penetration grade of 70.5, from Qilu Petrochemical (Linzi, China). The aged bitumen samples with penetration grade of 70.5, 55.4, 47.6, 36.8 and 29.7 were artificially produced by a thin film oven test [10]. 55.4, 47.6, 36.8 and 29.7 were artificially produced by a thin film oven test [10]. 2.2. Microcapsules Microcapsules Fabrication Fabrication Processes Processes 2.2. The method method of of fabrication fabrication microcapsules microcapsules containing containing rejuvenator rejuvenator by by coacervation coacervation process process has has The been reported in previous works [6–10]. In details, it can be divvied into three steps: (a) SMA was been reported in previous works [6–10]. In details, it can be divvied into three steps: (a) SMA was added to and allowed mixmix for for 2 h.2Then a solution of NaOH (10%) was added added to 100 100 mL mLwater wateratat5050°C◦ C and allowed h. Then a solution of NaOH (10%) was dropwise adjusting its pH value tovalue 10. The surfactant solution and rejuvenator were emulsified added dropwise adjusting its pH toabove 10. The above surfactant solution and rejuvenator were mechanically under a vigorous rate for 10rate minfor using a high-speed dispersedisperse machine;machine; (b) The emulsified mechanically under astirring vigorous stirring 10 min using a high-speed encapsulation was carried out in a 500 mL three-neck round-bottomed flask equipped with a
Materials 2016, 9, 932
4 of 15
(b) The encapsulation was carried out in a 500 mL three-neck round-bottomed flask equipped with a condensator and a tetrafluoroethylene mechanical stirrer. The above emulsion was transferred in the bottle, which was dipped in a steady temperature flume (room temperature). MMF prepolymer was added dropwise with a stirring speed of 500 r·min−1 . The temperature was increased to 80 ◦ C with a rate of 2 ◦ C·min−1 ; (c) After polymerization for 1 h, the temperature was decreased slowly to ambient temperature. At last, the resultant microcapsules were filtered and washed with pure water and dried in a vacuum oven. 2.3. Preparation of Microcapsules/Bitumen Samples The 50 pen grade aged bitumen was blended with different microcapsules using a propeller mixer for 30 min at 160 ◦ C with a constant speed of 200 r·min−1 . Hot microcapsules/bitumen mixture was poured on a glass sheet to form thin a film with a thickness less than 200 µm. The samples were keeping under 0 ◦ C waiting for tests. 2.4. Microcrack Generation Figure 2a illustrates the microcrack generation process through a self-designed installation. Hot bitumen/microcapsules mixture (100 ◦ C) was poured into the middle of two aluminium plates forming a thin bitumen layer with a thickness of 2 mm. Rectangular notches (length 5 mm) were carved at each end of the bitumen layer. The prepared thin bitumen/microcapsules sample carefully poured with drops of liquid nitrogen (N2 ) at one end [8]. Microcrack was quickly generated in this sample because of the low temperature brittleness. Materials 2016, 9, 932
5 of 15
Figure 2. The illustration of FTIR-ATR method testing the diffusion behavior of rejuvenator in Figure 2. The illustration of FTIR-ATR method testing the diffusion behavior of rejuvenator in bitumen, bitumen, (a) the aluminum plate containing bitumen with notch ends, microcrack generated by liquid (a) the aluminum plate containing bitumen with notch ends, microcrack generated by liquid N2 ; (b) the N2; (b) the FTIR-ATR testing prism. FTIR-ATR testing prism.
3. FTIR-ATR Method and Theoretical Framework Normally, FTIR-ATR is used to exploit the attenuation of light reflected internally in a nonabsorbing prism, due to energy absorption of an analyte in contact with the reflecting surface. To further enhance the attenuation, and consequently, the absorption spectra, the prism usually has an oblong and trapezoidal shape to allow multiple internal reflections. By using this method, the diffusion through a thin film is detected by quantifying the change in absorbance at wave numbers characteristic for the diffusing substances, thus obtaining the diffusion coefficient (D). In these tests, the diffusing substances are flushed over the film while keeping the concentration constant at the
Materials 2016, 9, 932
5 of 15
2.5. Morphologies Observation The dried microcapsules were observed by using an Environmental Scan Electron Microscopy (ESEM, Philips XL30, Philips, Amsterdam, The Netherlands) at an accelerated voltage of 20 kV. Self-healing behaviors of bitumen were observed by a fluorescence microscope (CKX41-F32FL, OLYMPUS, Tokyo, Japan). As bitumen is a temperature-sensitive material, the observation is in an environment of 0 ◦ C temperature. About 2 g MMF-shell microcapsules was mixed in 5 g epoxy resin. After the composite was dried at room temperature, it was carefully cut to obtain the cross-section. The thickness of shells can be measured from the SEM images of cross-section of microcapsules [4]. At least 20 shells of each microcapsule sample were measured and the average data were calculated. 2.6. FTIR-ATR Tests A modified FTIR-ATR (Vector 22, Bruker, Billerica, MA, USA) method [12] was used to continuously monitor rejuvenator diffusion into bitumen. ATR exploits the total internal reflectance of infrared light in a non-absorbing prism. Any absorbing substances in contact with the prism surface will attenuate the internally reflected light and, as a consequence, an infrared absorbance spectrum is obtained, corresponding to a spectrum recorded as if the light passed through the surface layer of the material studied. To determine diffusion rates of a rejuvenator penetrating a bitumen sample, a thin bitumen layer with a microcrack was applied on top of a zinc selenide (ZnSe) ATR prism as shown in Figure 2b. The application was accomplished by gluing brass frames on top of the prism. The bottom layer (the bitumen) was allowed to settle in the heated sample holder at 100 ◦ C for 30 min to obtain absolute contact between the sample and prism, and to avoid initial problems with air bubbles and other types of distress at the beginning of the diffusion test. All thin bitumen layers had a thickness of 2 mm. The temperature was set and infrared absorbance recorded. 3. FTIR-ATR Method and Theoretical Framework Normally, FTIR-ATR is used to exploit the attenuation of light reflected internally in a non-absorbing prism, due to energy absorption of an analyte in contact with the reflecting surface. To further enhance the attenuation, and consequently, the absorption spectra, the prism usually has an oblong and trapezoidal shape to allow multiple internal reflections. By using this method, the diffusion through a thin film is detected by quantifying the change in absorbance at wave numbers characteristic for the diffusing substances, thus obtaining the diffusion coefficient (D). In these tests, the diffusing substances are flushed over the film while keeping the concentration constant at the surface of the film. A modified FTIR-ATR method [12,13] was applied to determine diffusion rates of rejuvenator penetrating bitumen. Thin layer of bitumen was applied on top of a zinc selenide (ZnSe) ATR prism. The test was started by switching on the temperature control unit, after which the interferograms immediately began to be recorded. The time before reaching the test temperature was monitored to allow corrections for the initial heating period. An insulating cap was placed on top of the ATR sample holder to obtain a more uniform temperature in the sample. The absorbance at certain wave numbers was calculated using the integrated peak area as Figure 3a shown. In this study, a fairly specific absorption band for DPS at 843 cm−1 was selected as a marker band to calculate the diffusion coefficient [12]. The calculated absorbance values figured out using a logarithmic time scale (Figure 3b), where the location of the sloping line in the horizontal direction was determined by the D.
Materials 2016, 9, 932 Materials 2016, 9, 932
6 of 15 6 of 15
Figure 3. Calculating method of diffusion coefficient, (a) the fairly specific absorption band for DPS Figure 3. Calculating method of diffusion coefficient, (a) the fairly specific absorption band for DPS at −1 selected as a marker band to calculate the diffusion coefficient, the integrated peak area at 843 com 843 com−1 selected as a marker band to calculate the diffusion coefficient, the integrated peak area determined as absorbance; (b) a logarithmic scale curve of time-absorbance, the sloping line in the determined as absorbance; (b) a logarithmic scale curve of time-absorbance, the sloping line in the horizontal direction determined by the diffusion coefficient. horizontal direction determined by the diffusion coefficient.
Diffusion is defined as transport of matter due to random molecular movements termed Diffusion is defined transport of dependent. matter due to random molecular movements termed Brownian Brownian motions, and isastemperature For bituminous materials, diffusion takes place at motions, and is temperature dependent. For bituminous materials, diffusion takes place at the the micro-level and enables the recycled binder to become homogeneous [1,12]. Compatibility micro-level and enables the recycled binder to become homogeneous [1,12]. Compatibility between between binders is a requirement for creating a homogeneous binder, and is mainly dependent on binders is aand requirement for creating a homogeneousassociations. binder, and isThe mainly dependent on the nature and the nature distribution of the intermolecular concept of the diffusion process distribution of the intermolecular associations. The concept of the diffusion process can be physically can be physically verified using a stage extraction method, in which the inner and outer layers of the verified using a stage extraction method, in whichItthe and outer layersthe of binder the recycled binder film recycled binder film were extracted separately. wasinner reported that how stiffness (higher were extracted separately. It was reported that how the binder stiffness (higher penetration indicates penetration indicates lower stiffness) of the reclaimed material originally varied (the outer layer was lower than stiffness) of thelayer). reclaimed originally varied (themathematical outer layer was stiffer than the inner stiffer the inner Fick’smaterial law is generally used for the description of diffusion, layer). law is generally the mathematical description and is a simplification and is aFick’s simplification of the used morefor general theory of the influence of of diffusion, chemical potential on diffusion of the more general theory of the influence of chemical potential on diffusion as Equation (1) shown, as Equation (1) shown, ∂cc ∂2 c2 =DD· 2 c ∂t ∂x 2
t
x
(1) (1)
where c is concentration, t is time, x is position and D is the diffusion coefficient. From Equation (1), where c is concentration, t is time, x is position and D is the diffusion coefficient. From Equation (1), it can be seen that diffusion is dependent on the concentration of the diffusion and the diffusion it can be seen that diffusion is dependent on the concentration of the diffusion and the diffusion coefficients. At the same time, the diffusion coefficients are different as linear or nonlinear with regards coefficients. At the same time, the diffusion coefficients are different as linear or nonlinear with to various models. The Arrhenius equation is a formula for the temperature dependence of reaction regards to various models. The Arrhenius equation is a formula for the temperature dependence rates. The relationship between D and temperature can be derived using the heat energy activation of reaction rates. The relationship between D and temperature can be derived using the heat energy approach resulting in Equations (2) and (3), activation approach resulting in Equations (2) and (3), k2
D = k1 · e Tk 2 T
D k1 e
D( T ) = D0 e(d1 /T )+d2
(2) (2)
(3) D(T ) D0e (3) where k1 and k2 are constants, d1 and d2 are constants, and T is absolute temperature (unit K). It has been proved Equation (2) dis1 useful the diffusion ratetemperature in bitumen under the Itsame k2 are constants, and d2 to arecharacterize constants, and T is absolute (unit K). has where k1 and that temperature and Equation (3) can be applied to evaluate the D values under different temperatures [12]. been proved that Equation (2) is useful to characterize diffusion rate in bitumen under the same Based on the logarithmic scale ofapplied time-absorbance, curve wasunder performed graphically under temperature and Equation (3)curve can be to evaluate the fitting D values different temperatures different temperature with variousscale microstructure of microcapsules/bitumen (meanwas size,performed thickness, [12]. Based on the logarithmic curve of time-absorbance, curve fitting content). A number of measures were undertaken to increase the reliability modeling. graphically under different temperature with various microstructure ofofmicrocapsules/bitumen (mean size, thickness, content). A number of measures were undertaken to increase the reliability of modeling. ( d1 / T ) d 2
Materials 2016, 9, 932 Materials 2016, 9, 932
4. Discussion and Results
7 of 15 7 of 15
4. Discussion and Results
4.1. Microcapsules/Bitumen Samples 4.1. Microcapsules/Bitumen Usually, microcapsule Samples product has three main parameters: morphology, mean size and shell
thickness. Figuremicrocapsule 4a,b shows product the optical of microcapsules in emulsion. canshell be seen Usually, hasmorphologies three main parameters: morphology, mean sizeItand thatthickness. the MMFFigure prepolymer have polymerized on the exterior of the oil droplets forming shells. 4a,b shows the optical morphologies of microcapsules in emulsion. It can be seenThis that the MMFisprepolymer have of polymerized onSMA the exterior of the oil droplets formingatshells. This polymerization due to the help hydrolyzed molecules, which are adsorbed the interfaces polymerization is due to the help of hydrolyzed molecules, which are adsorbed thehydrophobic interfaces of oil droplets; these molecules readily undergoSMA directional arrangement such thatatthe of oilare droplets; these in molecules readily undergo directional arrangement such that groups positioned the oil droplets while the hydrophilic groups extend outthe of hydrophobic the droplets [4]. groups are positioned in the oil droplets while the hydrophilic groups extend out of the [4].have Figure 4c shows ESEM morphologies of dried microcapsules with a mean size about 20 droplets µm. They Figure 4c shows ESEM morphologies of dried microcapsules with a mean size about 20 μm. They a regular globe shape and smooth surfaces. The mean size values can be controlled by the method have a regular globe shape and smooth surfaces. The mean size values can be controlled by the of regulating the emulsion speed during polymerization. Figure 4d shows a typical fluorescence method of regulating the emulsion speed during polymerization. Figure◦ 4d shows a typical fluorescence microscope morphology of microcapsules /bitumen composite at 25 C. The microcapsules retain their microscope morphology of microcapsules /bitumen composite at 25 °C. The microcapsules retain compact shell structure during mixing with hot bitumen, maintaining their original shape and smooth their compact shell structure during mixing with hot bitumen, maintaining their original shape and surface. It is evident that the thermal dideffect not break thebreak shells. phenomenon is consisted smooth surface. It is evident that theeffect thermal did not theThis shells. This phenomenon is ◦ without to previously reported result that microcapsules could resist a temperature of about 180 consisted to previously reported result that microcapsules could resist a temperature of aboutC 180 °C being broken. without being broken.
Figure 4. Morphologies of microcapsules and microcapsules/bitumen composite, (a,b) optical
Figure 4. Morphologies of microcapsules and microcapsules/bitumen composite, (a,b) optical morphologies of microcapsules; (c) ESEM morphologies of microcapsules; (d) fluorescence morphologies of microcapsules; (c) ESEM morphologies of microcapsules; (d) fluorescence microscope microscope morphology of microcapsules/bitumen composite at 25 °C. morphology of microcapsules/bitumen composite at 25 ◦ C.
The shell thickness values were measured by observing the cross-section ESEM morphology of The shell thickness values werethickness measured by observing theamount cross-section morphology microcapsules. Normally, the shell is dominated by the of shell ESEM material during the of coacervation Normally, process, a lower core/shell ratiois(weight ratioby of the core/shell) to amaterial higher shell microcapsules. the shell thickness dominated amountleads of shell during valueprocess, [6]. To simplify complexity, microcapsules the same core/shell ratio of the thickness coacervation a lowerthe core/shell ratiothe (weight ratio of have core/shell) leads to a higher shell 2:1. Table 1 lists six microcapsule samples, coded as M-1, M-2, M-3, M-4, M-5 and M-6, with various thickness value [6]. To simplify the complexity, the microcapsules have the same core/shell ratio shell thickness mean size values. It must be mentioned that shell thickness andM-6, mean size of of 2:1. Table 1 listsand six microcapsule samples, coded as M-1, M-2, M-3, M-4, M-5 and with various microcapsules both have size a normal distribution to the in situshell polymerization. Eachmean data was shell thickness and mean values. It mustaccording be mentioned that thickness and size of an average value of five testing results. The samples have the shell thickness in a range of about microcapsules both have a normal distribution according to the in situ polymerization. Each data 1.2 μm–2.7 μm. The thickness of asphalt binder between aggregates is less than 50 μm, the size of was an average value of five testing results. The samples have the shell thickness in a range of about microcapsules containing rejuvenators must be smaller than 50 μm to avoid being squeezed or 1.2 µm–2.7 µm. The thickness of asphalt binder between aggregates is less than 50 µm, the size pulverized during asphalt forming [4]. Therefore, the samples have the mean size values about 10 of microcapsules containing rejuvenators must be smaller than 50 µm to avoid being squeezed or and 20 μm.
pulverized during asphalt forming [4]. Therefore, the samples have the mean size values about 10 and 20 µm.
Materials 2016, 9, 932 Materials 2016, 9, 932
8 of 15 8 of 15
Table Table 1. Microcapsule samples with various shell thickness and mean size.
Microcapsule Samples Shell Thickness (μm) Shell Thickness (µm) M-1 1.2 ± 0.6 M-1 1.2 ± 0.6 M-2 2.6 ± 0.5 M-2 2.6 ± 0.5 M-3 1.50.6 ± 0.6 M-3 1.5 ± M-4 2.7 ± M-4 2.70.7 ± 0.7 M-5 1.6 ± 0.7 M-5 1.6 ± 0.7 M-6 2.7 ± 0.6 M-6 2.7 ± 0.6
Microcapsule Samples
Mean Size (μm) Mean Size (µm) 10 ± 0.45 10 ± 0.45 10 ± 0.40 10 ± 0.40 20 ± 1.53 20 ± 1.53 20 ± 1.53 20 ± 1.53 30 ± 3.52 30 ± 3.52 30 ± 3.08 30 ± 3.08
4.2. Observation Observation of of the the Microcrack Microcrack Generation Generation and and Diffusion Diffusion Behaviors Behaviors 4.2. Before investigating investigating internal internal diffusion diffusionbehaviors, behaviors,it itmust mustbebe confirmed that healing Before confirmed that thethe healing of of microcapsule/bitumen composites is coming from the microcapsules. In other words, the microcapsule/bitumen composites is coming from the microcapsules. In other words, the microcapsules can can be be broken broken and and the the encapsulated encapsulated rejuvenator rejuvenator can can be be released released and and diffuse diffuse into into microcapsules bitumen. For this purpose, a microcrack therefore was generated by liquid N with a width less than 2 bitumen. For this purpose, a microcrack therefore was generated by liquid N2 with a width less than 10 μm µm to self-healing process. process. Figure Figure 5a,b 5a,b shows shows the the microcrack microcrack propagated propagated successfully successfully 10 to stimulate stimulate aa self-healing and pierced pierced microcapsules microcapsules along along the the development development track track and and the the shells shells of of microcapsules microcapsules were were broken broken and by a microcrack. Especially in Figure 5b, the diffusion behavior can be observed around the broken by a microcrack. Especially in Figure 5b, the diffusion behavior can be observed around the broken microcapsules. In In previous previous work work [9,10], [9,10], itit has has been been reported reported that that the the liquid liquid rejuvenator rejuvenator can can leak leak out out microcapsules. from microcapsules and flow into the microcrack, and then rejuvenator spreads along the whole from microcapsules and flow into the microcrack, and then rejuvenator spreads along the whole microcrack through through capillarity. capillarity. It It can can also also be be demonstrated demonstrated in in Figure Figure 55 that that aa microcrack microcrack can can pierce pierce microcrack several microcapsules thethe propagation trace.trace. Thus,Thus, enough rejuvenator can be supplied diffusing several microcapsulesalong along propagation enough rejuvenator can be supplied into the aged bitumen. diffusing into the aged bitumen.
Figure Figure 5. 5. ESEM ESEM morphologies morphologies of of microcracks microcracks in in bitumen, bitumen, (a) (a) aa microcrack microcrack propagation propagation direction; direction; (b) a microcrack punctured several microcapsules. (b) a microcrack punctured several microcapsules.
Figure 6a–c shows the fluorescence microscope morphologies of microencapsulated rejuvenator Figure 6a–c shows the fluorescence microscope morphologies of microencapsulated rejuvenator diffusing into bitumen under 25 °C at a different time. When the microcapsule was broken, the diffusing into bitumen under 25 ◦ C at a different time. When the microcapsule was broken, rejuvenator could quickly disperse around the shell. Mass transfer by molecular diffusion is one of the rejuvenator could quickly disperse around the shell. Mass transfer by molecular diffusion is one of the basic mechanisms in many branches of science. Molecular diffusion is a transport property which the basic mechanisms in many branches of science. Molecular diffusion is a transport property which controls the rate of mass transfer of species in a medium. However, the diffusion behaviour of controls the rate of mass transfer of species in a medium. However, the diffusion behaviour of bitumen bitumen has a significant relationship with temperature, time, and viscosity of the diffusion medium, has a significant relationship with temperature, time, and viscosity of the diffusion medium, diffusant diffusant size and polarity [15]. To simplify this diffusion process, the diffusion behaviour was size and polarity [15]. To simplify this diffusion process, the diffusion behaviour was identified by identified by measuring the diffusion diameters (S1, S2 and, S3) of rejuvenator from one microcapsule measuring the diffusion diameters (S , S and, S3 ) of rejuvenator from one microcapsule at different at different time of 30, 60 and 90 min.1 By2 comparing the diameter values of diffusion areas of these time of 30, 60 and 90 min. By comparing the diameter values of diffusion areas of these microcapsules, microcapsules, the diffusion direction can be pointed out and the diffusion rate can be roughly the diffusion direction can be pointed out and the diffusion rate can be roughly estimated. With the estimated. With the time increasing, the diffusion area gradually becomes larger from 10 μm to 100 time increasing, the diffusion area gradually becomes larger from 10 µm to 100 µm. Furthermore, the μm. Furthermore, the diffusion kinetic behavior, which is determined by the rate of mass transfer between the rejuvenator and aged bitumen under given conditions, can be calculated based on repeated measurement. At the same time, the diffusion coefficient between liquid-solid phases can
Materials 2016, 9, 932
9 of 15
diffusion kinetic behavior, which is determined by the rate of mass transfer between the rejuvenator Materials 2016, 9, 932 9 of 15 and aged bitumen under given conditions, can be calculated based on repeated measurement. At the same time, thebased diffusion coefficient can be deduced basedtemperature on the factorsand of be deduced on the factors between of liquidliquid-solid molecularphases structure, solid structure, liquid molecular structure, solid structure, temperature and pressure. pressure.
Figure 6. Fluorescence microscope morphologies of microencapsulated rejuvenator diffusing into Figure 6. Fluorescence microscope morphologies of microencapsulated rejuvenator diffusing into bitumen under under 25 25 ◦°C at different different time, time, the the diffusion diffusion diameters diameters of bitumen C at of rejuvenator rejuvenator from from one one microcapsule: microcapsule: (a) S 1 at 30 min; (b) S2 at 60 min, and (c) S3 at 90 min. (a) S1 at 30 min; (b) S2 at 60 min, and (c) S3 at 90 min.
4.3. Microcapsules Contents Dependency of Diffusion 4.3. Microcapsules Contents Dependency of Diffusion It must be noted that the diffusion coefficients of the asphalt binder in this study are not the true It must be noted that the diffusion coefficients of the asphalt binder in this study are not the true values because the measurement of diffusion rates of selected substance is only a marker. By carefully values because the measurement of diffusion rates of selected substance is only a marker. By carefully comparing the diffusion rates varying parameters, such as microcapsules content (Wm), comparing the diffusion rates varying parameters, such as microcapsules content (Wm ), microcapsules microcapsules mean size (δsize) and shell thickness (δthick), the results can be applied to analyze the mean size (δsize ) and shell thickness (δthick ), the results can be applied to analyze the microstructure microstructure of asphalt binders. It is necessary to be aware the fact that the diffusion molecules of asphalt binders. It is necessary to be aware the fact that the diffusion molecules (DPS) will affect (DPS) will affect the diffusion media and thereby change the diffusion rates [12]. Figure 7 shows the the diffusion media and thereby change the diffusion rates [12]. Figure 7 shows the absorbance-time absorbance-time curves of DPS at 843 cm−1 band under 30 °C during 1000 min in bitumen samples curves of DPS at 843 cm−1 band under 30 ◦ C during 1000 min in bitumen samples with the Wm values with the Wm values of 2.0, 4.0 and 6.0 wt %. All the additive microcapsules were the sample M-1. As of 2.0, 4.0 and 6.0 wt %. All the additive microcapsules were the sample M-1. As the microcapsules the microcapsules have the same core/shell ratio of 2:1, DPS weight content in microcapsules/bitumen have the same core/shell ratio of 2:1, DPS weight content in microcapsules/bitumen composite can be composite can be calculated as Equation (4), calculated as Equation (4), 22 WDPS (4) m W DPS= 3 ×WW (4) m
3
where WDPS is the DPS weight content, Wm is the microcapsule weight content. The logarithmic is the DPS content, Wm isfit the microcapsule weightwell, content. Theislogarithmic time where WDPSreveals time scale that weight the diffusion curves the theoretical curves which consisted with scale revealsresults that the diffusion curves fitpossible the theoretical curves well, which consisted with the the reported [13]. In this way, it is to indicate by comparing theisslopes of curves, as reported results [13]. In this way, it is possible to indicate by comparing the slopes of curves, as expected, that higher concentration of DPS increases the diffusion rates. It means that the Wm had expected, that higher concentration of DPS increases the diffusion rates. It means that the Wm had affected the diffusion coefficient. It has been well known that one important capability of rejuvenator is to soften the age bituminous binders and thereby accelerate the diffusion process [8].
Materials 2016, 9, 932
10 of 15
affected the diffusion coefficient. It has been well known that one important capability of rejuvenator is to soften age bituminous binders and thereby accelerate the diffusion process [8]. Materials 2016, 9,9,932 10 Materials 2016,the 932 10of of15 15
◦°C Figure 7.7.Absorbance-time Absorbance-time curves of 843 cm Figure7. Absorbance-timecurves curvesof of843 843cm cm−−1−11 band band at at 30 30 °C with the the microcapsules microcapsules contents contents of of 2.0, 2.0, Figure band at 30 C with with the microcapsules contents of 2.0, 4.0 and 6.0 wt %. 4.0 and 6.0 wt %. 4.0 and 6.0 wt %.
Another factor can enhance the rate isis the which will promote the Anotherfactor factor can enhance the diffusion diffusion the temperature, temperature, which willthe promote the Another can enhance the diffusion rate israte the temperature, which will promote molecules molecules movement speed [16]. Figure 8 shows the diffusion coefficient values of DPS in bitumen molecules movement Figurethe 8 shows the coefficient diffusion coefficient values DPS in samples bitumen movement speed [16]. speed Figure[16]. 8 shows diffusion values of DPS in of bitumen samples mixing with microcapsules (sample of M-1) with contents of 2%, 4% and 6% under samples mixing with microcapsules (sample of M-1) with contents of 2%, 4% and 6% under aa mixing with microcapsules (sample of M-1) with contents of 2%, 4% and 6% under a temperature range temperature range of 10 °C–60 °C. With the temperature increasing, the diffusion coefficient values temperature range of 10 °C–60 °C. With the temperature increasing, the diffusion coefficient values of 10 ◦ C–60 ◦ C. With the temperature increasing, the diffusion coefficient values also liner increased also also liner liner increased increased sharply. sharply. Moreover, Moreover, more more microcapsules microcapsules mixed mixed in in bitumen bitumen had had enhanced enhanced the the sharply. Moreover, more microcapsules mixed in bitumen had enhanced the diffusion coefficient diffusion diffusioncoefficient coefficientvalues valuesin inaccord accordwith withthe theresults resultsin inFigure Figure7. 7. values in accord with the results in Figure 7.
Figure 8.8.Diffusion Diffusion coefficient values microcapsules (sample in with contents of Figure8. Diffusioncoefficient coefficientvalues values of microcapsules (sample of M-1) M-1) in bitumen bitumen with contents of Figure ofof microcapsules (sample of of M-1) in bitumen with contents of 2%, ◦ C–60 ◦ C. °C. 2%, 4% 6% aatemperature range 10 2%,and 4%and andunder 6%under under temperature range of 10°C–60 °C–60 °C. 4% 6% a temperature range of 10of
4.4. 4.4.Mean MeanSize Sizeand andShell ShellThickness ThicknessDependency DependencyofofDiffusion Diffusion self-healing process of aged bitumen based on the microcapsules containing rejuvenator Theself-healing self-healingprocess processof ofaged agedbitumen bitumenbased basedon onthe themicrocapsules microcapsules containing rejuvenator The containing rejuvenator is is ais a series of movement of rejuvenator, including the releasing from broken microcapsules, the capillary a series movement rejuvenator, includingthe thereleasing releasingfrom frombroken brokenmicrocapsules, microcapsules,the the capillary series ofof movement ofof rejuvenator, including movement movementin inmicrocracks microcracksand andthe thediffusion diffusionmovement movementin inbitumen bitumen[8]. [8].This Thiscomplex complexmechanism mechanismhas has been been reported reported [9] [9] in in previous previous works works by by analyzing analyzing the the microstructure microstructure of of microcapsules/bitumen microcapsules/bitumen samples. samples. Therefore, Therefore, the the microstructure, microstructure, such such as as mean mean size size (δ (δsize size)) and and shell shell thickness thickness (δ (δthick thick)) of of microcapsules microcapsuleswill willgreatly greatlyinfluence influencethe theshell shellstrength strength[5]. [5].Shell Shellstrength strengthdetermines determinesthe thedamage damagerate rate of ofmicrocapsules microcapsulesby bythe themicrocracks microcracks[9]. [9].Higher Highershell shellstrength strengthwill willresist resistthe thetip tipstress stressof ofaamicrocrack microcrack
Materials 2016, 9, 932
11 of 15
movement in microcracks and the diffusion movement in bitumen [8]. This complex mechanism has been reported [9] in previous works by analyzing the microstructure of microcapsules/bitumen samples. Therefore, the microstructure, such as mean size (δsize ) and shell thickness (δthick ) of microcapsules will greatly influence the shell strength [5]. Shell strength determines the damage Materials 2016, 9, 932 11 of 15 rate of microcapsules by the microcracks [9]. Higher shell strength will resist the tip stress of a microcrack the integrality a microcapsule. result less rejuvenator can be and released keeping thekeeping integrality a microcapsule. The result The means lessmeans rejuvenator can be released then and then soften the bitumen. soften the bitumen. Figure the diffusion coefficient values of DPS inofbitumen withsamples 6% microcapsules Figure9 9shows shows the diffusion coefficient values DPS insamples bitumen with 6% under a temperature of 10 ◦ C–60 These microcapsule samples (M-2, M-4 and microcapsules under arange temperature range◦ C. of 10 °C–60 °C. These microcapsule samples (M-2,M-6) M-4had and different mean sizes. It can be seen that the diffusion coefficient values have a linear growth with M-6) had different mean sizes. It can be seen that the diffusion coefficient values have a linear growth the of temperature. It isItconsistent with thethe above withincreasing the increasing of temperature. is consistent with aboveconclusions conclusionsasasshown shownin in Figure Figure 8. On On another another hand, hand, itit is is found found that that larger larger microcapsules microcapsules lead lead to to aa higher higher diffusion diffusion rate rate under under the the same same ◦ C, bitumen samples mixing with M-2, temperature for the various samples. Under temperature of 10 temperature for the various samples. Under temperature of 10 °C, bitumen samples mixing with MM-4 andand M-6M-6 have a nearly samesame diffusion coefficient value.value. However, the diffusion coefficient value 2, M-4 have a nearly diffusion coefficient However, the diffusion coefficient of M-6/bitumen sample is nearly 130% of M-2/bitumen. In previous work, it has been proved that value of M-6/bitumen sample is nearly 130% of M-2/bitumen. In previous work, it has been proved larger microcapsules can can enhance the the encapsulation ratio of of core materials. that larger microcapsules enhance encapsulation ratio core materials.The Theabove aboveconclusion conclusion may to the thereason reasonthat thatlarger largermicrocapsules microcapsulesown own more rejuvenator. Figure 10 shows may be attributed to more rejuvenator. Figure 10 shows the the diffusion coefficient values of bitumen samples mixing with 2% microcapsules of M-1, M-3 and diffusion coefficient values of bitumen samples mixing with 2% microcapsules of M-1, M-3 and M-5 ◦ C. M-1, M-3 and M-5 have the different shell thickness M-5 under a temperature 10 ◦ C–60 under a temperature rangerange of 10 of °C–60 °C. M-1, M-3 and M-5 have the different shell thickness values. values. With temperature increasing, the diffusion coefficient also has a growth. linear growth. However, With temperature increasing, the diffusion coefficient also has a linear However, fitting fitting straight lines for various microcapsules/bitumen samples are close to coincidence. It indicates straight lines for various microcapsules/bitumen samples are close to coincidence. It indicates that that the shell thickness values do not greatly influence diffusion coefficient. This conclusion the shell thickness values do not greatly influence thethe diffusion coefficient. This conclusion cancan be be drawn that the microcapsules samples with different shell thickness values have a nearly same drawn that the microcapsules samples with different shell thickness values have nearly same amount amount of of rejuvenator. rejuvenator. ItIt also also can can be be deduced deduced that that the the three three microcapsules microcapsules in in bitumen bitumen may may have have the the same same damage damage ratio ratio by by microcracks. microcracks. The microcapsules were fabricated in this study study by by an an in in situ situ polymerization, polymerization, which which lead lead the the shell shell thickness thickness had had aa size size with with aanormal normaldistribution. distribution. The The shell shell of of microcapsules microcapsules is is aatiny tinypolymeric polymericmembrane. membrane.Moreover, Moreover,all allmicrocapsules microcapsuleshave havethe thesame samecore/shell core/shell ratio The normal distribution maymay weaken the influence of shellofthickness on the average ratioininthis thisstudy. study. The normal distribution weaken the influence shell thickness on the data of shell In other words, this normal weakens the of average datathickness of shell [17]. thickness [17]. In other words,distribution this normalalso distribution alsodiscrepancy weakens the damage ratioof fordamage different microcapsules bitumen by microcracks. discrepancy ratio for differentinmicrocapsules in bitumen by microcracks.
Figure 9. 9. Diffusion Diffusion coefficient coefficient values values of of microcapsules microcapsules(content (contentof of6%) 6%) in in bitumen bitumen with with different different mean mean Figure size values (samples of M-2, M-4 and M-6) under a temperature range of 10 °C–60 °C. ◦ ◦ size values (samples of M-2, M-4 and M-6) under a temperature range of 10 C–60 C.
Materials 2016, 9, 932
12 of 15
Materials 2016, 9, 932
12 of 15
Materials 2016, 9, 932
12 of 15
Figure 10. Diffusion coefficient values of microcapsules (content of 2%) in bitumen with different shell Figure 10. Diffusion coefficient values of microcapsules (content of 2%) in bitumen with different shell thickness values (samples of M-1, M-3 and M-5) under a temperature range of 10 °C–60 °C. thickness values (samples of M-1, M-3 and M-5) under a temperature range of 10 ◦ C–60 ◦ C. Figure 10. Diffusion coefficient values of microcapsules (content of 2%) in bitumen with different shell
thickness values (samplesofofDiffusion M-1, M-3 and M-5) under a temperature range of 10 °C–60 °C. 4.5. Aged Degree Dependency
4.5. Aged Degree Dependency of Diffusion
Figure 11 shows the diffusion coefficient values of DPS in bitumen with 6% microcapsules (M-1) 4.5. Aged Degree Dependency of Diffusion Figurea 11 shows therange diffusion DPS in samples bitumenwere withapplied 6% microcapsules (M-1) under temperature of 10coefficient °C–60 °C. values Variousofbitumen with different ◦ C–60 ◦ C. Various Figure 11 shows the diffusion coefficient values of DPS in bitumen with 6% microcapsules (M-1) under a temperature range of 10 bitumen samples were applied with different penetration grade values of 29.7, 36.8, 47.6, 55.4 and 70.5. It can be seen that the temperature enhances under a grade temperature range ofbitumen 10 °C–60 °C. bitumen samples werethe applied different penetration values ofsame 29.7, 36.8, 47.6, 55.4Various and 70.5. It can sample. be seen Under that enhances the diffusion rate for the microcapsules/bitumen thetemperature samewith temperature, penetration grade values of 29.7, 36.8, 47.6, 55.4 and 70.5. It can be seen that the temperature enhances the diffusion ratecoefficient for the same bitumen sample. Undervalues the same temperature, the diffusion values have amicrocapsules/bitumen linear growth. The diffusion coefficient do not have a the diffusion rate for the same bitumen microcapsules/bitumen Under same temperature, significant growth under temperature of 10 growth. °C. However, the sample. fittingcoefficient linear atthe 60values °C has thenot largest the diffusion coefficient values have a linear The diffusion do have a the diffusion coefficient values have a linear growth. The diffusion coefficient values do not have a ◦ ◦ slope, which indicates that DPS has a highest diffusion rate. This conclusion is consistent with significant growth under temperature of 10 C. However, the fitting linear at 60 C has the largestthe slope, significant growth under temperature of 10 °C. However, the fitting linear at 60 °C has the largest reported results has a higher rate in is higher penetration grade of which indicates thatthat DPSthe hasrejuvenator a highest diffusion rate. diffusion This conclusion consistent with the reported slope, which indicates that DPSsmall has amolecules highest diffusion rate. This conclusion consistent bitumen with the bitumen [12].rejuvenator The reason ishas that of rate rejuvenator canpenetration easily moveisthrough results that the a higher diffusion in higher grade ofthe bitumen [12]. reported results loose that structure the rejuvenator has a higher diffusion rate material in higher penetration grade of a relatively [18]. Bitumen iscan oneeasily kind of colloid the large molecules The with reason is that small molecules of rejuvenator move through theand bitumen with a relatively bitumen [12]. The reason is that small molecules of rejuvenator can easily move through the bitumen together by intermolecular force and form a network structure. The network structure might be the loose structure [18]. Bitumen is one kind of colloid material and the large molecules together by with a relatively structure [18].of Bitumen is one molecules. kind of colloid material the the large molecules principal barrier loose for the diffusion rejuvenator At the sameand time, addition of intermolecular force and form a network The network might bemight the principal together by can intermolecular force and formstructure. a network structure. The structure network structure be the rejuvenator break the network structure and accelerate the molecular diffusion. The content of barrier for the diffusion of rejuvenator molecules. At the same time, the addition of rejuvenator principal barrierofforage thebitumen diffusion of large, rejuvenator molecules. At the time,diffusion the addition ofcan large molecules was therefore, rejuvenator hadsame a higher rate in break the network structure and accelerate the molecular diffusion. The content of large molecules rejuvenator break thelarger network structuregrade. and accelerate the molecular diffusion. The content of of bitumen withcan a relatively penetration age bitumen was large, therefore, rejuvenator a higher diffusion had rate ainhigher bitumen with a rate relatively large molecules of age bitumen was large, had therefore, rejuvenator diffusion in bitumen with a grade. relatively larger penetration grade. larger penetration
Figure 11. Diffusion coefficient values of microcapsules (content of 6%) in bitumen with different penetration grade values (29.7, 36.8, 47.6, 55.4 and 70.5) under a temperature range of 10 °C–60 °C. Figure Diffusion coefficientvalues valuesofofmicrocapsules microcapsules (content with different Figure 11. 11. Diffusion coefficient (contentofof6%) 6%)ininbitumen bitumen with different ◦ penetration grade values (29.7, 36.8, 47.6, 55.4 and 70.5) under a temperature range of 10 °C–60 °C.◦ C. penetration grade values (29.7, 36.8, 47.6, 55.4 and 70.5) under a temperature range of 10 C–60
Materials 2016, 9, 932
13 of 15
4.6. Preliminary Model of Diffusion Coefficient All the above results indicate that the microstructure of microcapsules greatly influent the diffusion behaviors based on the concentration of released rejuvenator. Therefore, it is meaningful research to supply a semi-experimental model of diffusion coefficient. In previous works, the self-healing efficiency has been evaluated by mechanical tests such as modified beam on elastic foundation method [9] and a repetitive direct tension method [19]. Direct methods involve the compositional analysis of mechanical properties of microcapsules/bitumen before and after a designed self-healing process. Basically, it was deduced that the mechanism of self-healing based on microcapsules technology was a rejuvenator diffusion process [9,10]. In other words, a model of diffusion coefficient can be applied as a guide to indirectly predict the self-healing efficiency. It must be noted that the diffusion coefficient concept in this study is different to the normal rejuvenator diffusion coefficient, because the microcapsules have great effects on the diffusion process. The model is only a semi-experimental model, which can expose some microstructure information about asphalt. In the microcapsules/bitumen material system, diffusion coefficient (D) can be considered as a function as Equation (5), D = f (δsize , δthick , Rc/s , Dr , Wm , Ad , T ) (5) where Rc/s is the core/shell material ratio, Dr is the damage rate of microcapsules in bitumen, and Ad is the age degree of bitumen. The Ad is defined as the percentage of penetration degree between aged bitumen at time (t) and the original bitumen. Rc/s value is a constant in this study. The influence of δthick is ignored in this model because it does not play a decisive role for diffusion behaviors as the results in Figure 10. In contrast, the Dr is a main factor affecting the amount of rejuvenator diffusing into aged bitumen. It has been proved in Figure 7 that the diffusion curves fit the theoretical curves well with a logarithmic time scale. Based on the above consideration, the Equation (2) can be written as Equation (6), ln D ( T ) = λ1 + λ2 /T (6) where λ1 and λ2 are constants. Introducing the relation of the above parameters into this formula, ln D = λ1 Ad ϕ1 + λ2 (δsize δthick Rc/s ) ϕ2 ( Dr Wm ) ϕ3 Ad − ϕ4 /T
(7)
where ϕ1 , ϕ2 , ϕ3 and ϕ4 are constants. ϕ2 is determined by the microstructure of microcapsules, ϕ3 is determined by the released amount of oily rejuvenator in diffusion process, and ϕ1 and ϕ4 is determined by the aged degree of bitumen. The above three parameters will be given in future work by mathematical fitting with a large number of experimental data. Based on the above experimental results, this model will be a guide to the construction and application of self-healing bitumen using microcapsules. 5. Conclusions In this study, diffusion behaviors of microencapsulated rejuvenator in aged bitumen were evaluated by a FTIR-ATR method. The core material of microcapsules used as rejuvenator was diphenylsilane (DPS), its fairly specific absorption band was selected as a marker band to calculate the diffusion coefficient. From the mentioned preliminary results, the following conclusions can be concluded. 1.
2.
The microstructure affected the D values of DPS in aged bitumen. A higher mean shell thickness decreased the D values because of the decrement of damage probability of microcapsules under the same content. With the same microcapsule sample in bitumen, the D values presented a trend of linear increase when the content of microcapsules was increased. All these results indicated that the microstructure greatly influent the diffusion behaviors based on the concentration of released rejuvenator.
Materials 2016, 9, 932
3.
14 of 15
A preliminary model of diffusion behaviors of microencapsulated rejuvenator in bitumen was given based on the Arrhenius equation. This model considered the influence factors of microstructure, the amount of released rejuvenator and the age degree of bitumen. It is a guide to the construction and application of self-healing bitumen using microcapsules.
Acknowledgments: The authors acknowledge financial support by Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering (Project GDDCE15-07) and National Natural Science Foundation of China (U1633201 and 51573135). Author Contributions: S.H. and Y.W. conceived and designed the experiments; S.H. performed the experiments; J.S., P.Y. and N.H. analyzed the data; L.W. contributed reagents/materials/analysis tools; J.S. wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.
Nomenclature Diffusion coefficient Concentration Temperature Coordiante direction Time Diphenylsilane Microcapsule weight content DPS weight content Microcapsules mean size Microcapsules shell thickness DPS weight content Core/shell material ratio Damage rate of microcapsules in bitumen
D (m2 /s) C (L/m3 ) C (L/m3 ) x (m) t (s) DPS Wm (wt %) WDPS (wt %) δsize (m) δthick (m) WDPS (wt %) Rc/s Dr (%)
References 1. 2. 3. 4. 5.
6. 7.
8. 9.
Wang, C.; Zhang, H.; Castorena, C.; Zhang, J.X.; Kim, Y.R. Identifying fatigue failure in asphalt binder time sweep tests. Constr. Build. Mater. 2016, 121, 535–546. [CrossRef] Qiu, J. Self-Healing of Asphalt Mixtures. Ph.D. Thesis, Delft University of Technology, Delft, The Netherlands, 2012. García, A.; Schlangen, E.; Van de Ven, M. Properties of capsules containing rejuvenators for their use in asphalt concrete. Fuel 2011, 90, 583–591. [CrossRef] Su, J.F.; Schlangen, E. Synthesis and physicochemical properties of novel high compact microcapsules containing rejuvenator applied in asphalt. Chem. Eng. J. 2012, 198, 289–300. [CrossRef] Su, J.F.; Dong, H.; Wang, X.Y. Influence of temperature on the deformation behaviors of melamine-formaldehyde microcapsules containing phase change material. Mate. Lett. 2012, 84, 158–161. [CrossRef] Su, J.F.; Schlangen, E.; Qiu, J. Design and construction of methylation melamine-formaldehyde microcapsules containing rejuvenator for asphalt. Powder Technol. 2013, 235, 563–571. [CrossRef] Su, J.F.; Qiu, J.; Schlangen, E.; Wang, Y.Y. Investigation the possibility of a new approach of using microcapsules containing waste cooking oil: In situ rejuvenation for aged bitumen. Constr. Build. Mater. 2015, 74, 83–92. [CrossRef] Su, J.F.; Qiu, J.; Schlangen, E.; Wang, Y.Y. Investigation the self-healing mechanism of aged bitumen using microcapsules containing rejuvenator. Constr. Build. Mater. 2015, 85, 49–56. [CrossRef] Su, J.F.; Qiu, J.; Schlangen, E.; Wang, Y.Y. Experimental investigation of self-healing behavior of bitumen/microcapsule composites by a modified beam on elastic foundation method. Mater. Struct. 2015, 48, 4067–4076. [CrossRef]
Materials 2016, 9, 932
10.
11. 12. 13. 14. 15. 16. 17. 18. 19.
15 of 15
Su, J.F.; Wang, Y.Y.; Han, N.X.; Yang, P.; Han, S. Experimental investigation and mechanism analysis of novel multi-self-healing behaviors of bitumen using microcapsules containing rejuvenator. Constr. Build. Mater. 2016, 106, 317–329. [CrossRef] Ma, T.; Huang, X.; Zhao, Y.; Zhang, Y. Evaluation of the diffusion and distribution of the rejuvenator for hot asphalt recycling. Constr. Build. Mater. 2015, 98, 530–536. [CrossRef] Karlsson, R.; Isacsson, U. Laboratory studies of diffusion in bitumen using markers. J. Mater. Sci. 2003, 38, 2835–2844. [CrossRef] Karlsson, R.; Isacsson, U. Application of FTIR-ATR to characterization of bitumen rejuvenator diffusion. J. Mater. Civil. Eng. 2003, 15, 157–165. [CrossRef] Putman, B.J.; Amirkhanian, S.N. Characterization of the interaction effect of crumb rubber modified binders using HP-GPC. J. Mater. Civil Eng. 2010, 22, 153–159. [CrossRef] Dong, Y.; Huang, B.; Zhang, X.Y.; Woods, M.E. Use of molecular dynamics to investigate diffusion between vigin and aged asphalt binders. Fule 2016, 174, 267–273. [CrossRef] Bhasin, A.; Palvadi, S.; Little, D.N. Influence of aging and temperature on intrinsic healing of asphalt binders. Transport. Sci. Technol. 2011, 2207, 70–78. [CrossRef] Su, J.F.; Wang, X.Y.; Dong, H. Micromechanical properties of melamine formaldehyde microcapsules by nanoindentation: Effect of size and shell thickness. Mater. Lett. 2012, 89, 1–4. [CrossRef] Ma, T.; Huang, X.; Zhao, Y.; Zhang, Y.; Wang, H. Influences of preheating temperature of RAP on properties of hot-mix recycled asphalt mixture. J. Test. Eval. 2016, 44, 762–769. [CrossRef] Su, J.F.; Wang, Y.Y.; Han, S.; Han, N.X.; Li, W. Investigation of the self-healing behaviors of bitumen/microcapsules composites by a repetitive direct tension test. Materials 2016. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
