Hot Recycling of Bituminous Mixtures

Abstract

This chapter first presents the results of a survey performed on the practices used in Europe. This survey is considered a first report on the subject and could be usefully completed by the states of the art prepared in the framework of the Re-road and Direct-Mat European projects. Then the chapter addresses the key issue of Reclaimed Asphalt (RA) manufacture in laboratory considered within the group as an important point to assess recyclability of bituminous mixtures. To answer this question, common research has been launched on the basis of an interlaboratory test programme to set a relevant laboratory ageing of bituminous mixtures protocol. The choices of the tested protocol, the setting of the interlaboratory test programme as well as the results obtained, are presented in this chapter. A conclusion in the form of a recommendation for laboratory ageing of bituminous mixtures is given at the end of the chapter.

Annex: Principle of Infrared Spectroscopy

7.1.1 Theory of Infrared Spectroscopy Technique

Infrared spectroscopy is an analytical technique that allows identification of functional groups of organic molecules and also certain mineral products. This technique is based on the excitation of matter molecules by an electromagnetic wave in the mid-infrared region, approximately 4,000–400 cm⁻¹(30–2.5 μm). When a molecule absorbs electromagnetic radiation there is an increase in its total energy. According to a quantum theory, only a quantum of radiation of a specific energy can be absorbed by a molecule, thereby raising its energy from the ground state to an excited state. The energy of transition, ΔE, is equal to the energy of the interacting photon (hν). The remaining energies of a molecule are due to vibrational, rotational and translational motion associated with the molecule. Infrared radiation is of sufficient energy to cause transitional, rotational and vibrational energy levels of a molecule. Each vibration mode corresponds to a wavelength or absorption energy. So the analysis of the light-absorption according to the wavelength corresponds to a specific spectrum of a molecule, an ion or a crystallised structure and can be considered as a digital print.

IR spectra are collected using Fourier Transform Infrared (FTIR) spectroscopy. Where radiation is varied by a monochromator, the IR radiation is guided through a Michelson interferometer (see Fig. 7.44). The purpose of the interferometer is to have a beam of IR radiation, split it into two beams, and make one of the beams travel a different (optical) distance than the other in order to create alternating interference fringes (Smith [29])

Fig. 7.44

The Michelson interferometer [15]

The FTIR spectra are usually presented as plots of intensity versus wavenumber (in cm-1). Wavenumber is the reciprocal of the wavelength. The intensity can be plotted as the percentage of light transmittance or absorbance at each wavenumber.

The infrared spectra can be obtained in different modes (transmission, Attenuated Total Reflectance (ATR) etc.). In this work, the transmission mode was used by seven laboratories and the ATR mode was used by one laboratory. In the transmission mode, IR radiation is passed through the sample. Some of the infrared radiation is absorbed by the sample and some of it passes through (transmitted). Attenuated total reflection infrared (ATR-IR) spectroscopy is a subset of Infrared spectroscopy which enables samples to be examined directly in the solid or liquid state without any preparation. ATR spectroscopy and multiple internal reflection spectroscopy are increasingly gaining popularity as significant tools to analyse coatings and opaque liquids. In the ATR mode, a beam of infrared light is passed through the ATR crystal which is covered on the top by the solid or the liquid sample such that the incident infrared light reflects internally after coming in contact with the crystal covered with the sample, forming an evanescent wave which extends into the sample, typically by a few microns. The angle of incidence must be greater than the critical angle for total internal reflectance of the infrared light. When the beam exits the crystal, it is collected by a detector and analysed and displayed in form of the ATR-IR spectra (Fig. 7.45). The ATR spectrum is simply the IR spectrum, so all the typical uses apply.

Fig. 7.45

Schematic a simple reflection ATR system

7.1.2 Quantification of Species

FTIR analysis allows the quantification of a given function by applying the Beer Lambert law:

$$ {\rm {A(\upsilon )}} = {{\rm {\varepsilon }}_{\rm{i}}}{\rm {(\upsilon )}} \times {\rm {l}} \times {{\rm {C}}_{{_{\rm{i}}}}} $$

(7.2)

with:

• A(ν): absorbance for the wavenumber ν

• ε_i(ν): molar absorptivity coefficient for the wavenumber ν

• l: optical path length

• C_i: studied species concentration

In the case of complex systems, like bitumens, the superimposition of absorption bands, due to the absorption of several molecules, is frequent. It is so possible overall to integrate several vibrations of the same type²(for example, the C=O ester, acid, ketone, lactone vibrations between 1,800 and 1,635 cm^–1), by considering an area comprised between two wavelengths.

On the basis of Beer Lamber law, the global absorbance can be then written as follows;

$$ \int_{{{{\rm{\nu }}_{{1}}}}}^{{{{\rm{\nu }}_{{2}}}}} {{\rm {A(\upsilon )d\upsilon }}} = \int_{{{{\rm{\nu }}_{{1}}}}}^{{{{\rm{\nu }}_{{2}}}}} {{{\rm {\varepsilon }}_{\rm{i}}}{\rm {(\upsilon )}} \times {\rm {l}} \times {{\rm {C}}_{{_{\rm{i}}}}}} {\rm {d\upsilon }} $$

(7.3)

with :

• A_(ν): absorbance for the wavenumber ν

• ε_i(ν): molar absorptivity coefficient for the wavenumber ν

• l: optical path length

• C_i: studied species concentration

But, for following the absorption area evolution or for the comparison of two samples, it is important to always consider the same band widths.

7.1.3 Application of FTIR to the Oxidation State of Bitumens

Apart from the fact that FTIR is useful for the detection and analysis of chemical oxidation bands (Doumenq [7, 8]; Pieri [22]), it can be also used for the classification of bitumen’s composition in terms of functionalities (aliphaticity, aromaticity, substitution etc.), that are quantified by indexes, expressed as ratios of characteristic spectral absorption bands (Pieri et al. [23]).

The assessment of oxidation level has been performed according to the test method of LPC no 69.

7.1.4 Preparation of Test Samples

Bitumen samples are prepared according to the norm NF EN 12594 ([2]). The binder is heated at a temperature high enough to decrease its viscosity before being stirred with a spatula. The quantification of oxygenated species is performed using the analytical technique of FTIR spectroscopy in transmission mode according to two possibilities of sample preparation:

• A hot binder droplet is laid on a thin plate transparent to the infrared radiations (KBr, NaCl or CsI) and then it is spread with a spatula in order to obtain a sample film transparent to the infrared radiation.

• A binder solution in the dichloromethane of “spectrosol” quality is prepared at 30 g/L concentration from 100 mg of sample intake (Pieri [22]) or at 170 g/L concentration from 500 mg of sample intake (Durrieu et al. [9]). These solutions are then directly analysed by FTIR spectroscopy according to the so-called dry-plate technique (Smith [29]), consisting in laying such a quantity of solution (25 μL) on an infrared thin plate (KBr, NaCl or CsI) and then to evaporate the solvent in order to obtain a homogeneous thin film of sample transparent to the infrared radiation. The total evaporation of dichloromethane is checked by the absence of infrared band at 1,265 cm^–1in the obtained FTIR spectra. In transmission mode, the spectra are recorded between 4,000 and 400 cm^–1with a scans number of 32 and a resolution of 2 cm^–1.

For these two techniques of sample preparation, it is important to verify that the obtained signal is in the concentration’s domain that permits to obtain a linear response of measured areas and, consequently, to lead to a constant value of indexes whatever bituminous binders are analysed.

An example of an obtained spectrum is shown on Fig. 7.46. The effects of bituminous binder’s ageing by oxidation are well-known in relation to the apparition of oxygenated chemical species. As a result, only the carbonyls functions (CO showing a characteristic absorption band around 1,700 cm^–1) and sulfoxides functions (SO showing a characteristic absorption band around 1,030 cm^–1) are selected.

Fig. 7.46

An example of FTIR spectrum of bitumen recorded between 4,000 and 400 cm^–1

7.1.5 Interpretation of FTIR Spectra of Bitumens and Calculation of Oxidation Indices

In order to take account of the majority of oxygenated species of bitumens, the calculations have been performed on an area of absorption bands. It is important to note that, in case of the follow-up of ageing evolution by oxidation, the formed oxygenated compounds that absorb out of selected boundary markers will be not taken into account.

Despite the sample preparation being done carefully, it is impossible to obtain a binder film with a perfectly homogeneous and reproducible thickness. So, it becomes difficult to obtain data in relation to molar absorptivity or extinction coefficients ε of each absorption band characteristic of various functional groups (these ε depend on the studied chemical group, the electronic environment of the group and the nature of the sample itself) and to determine the optical path length (thickness of the sample) as well.

In order to remove problems linked to the determination of the optical path length and molar absorptivity or extinction coefficients for complex chemical systems, the calculated areas are used in determining band area ratios characteristic of the relative proportion of the different chemical species (namely aromatic CH/aliphatic CH, etc.) or of the relative proportion within the sample (namely aliphatic CH/sum of integrated band areas, etc.). This methodology, called semiquantitative infrared analysis (Doumenq [7, 8]; Ganz and Kalkreuth [11]; Pieri [22]; Mouillet [17]), allows to disregard the thickness of the sample.

By considering two functional groups A and B, the ratio of the corresponding infrared band areas is defined, based on the Beer-Lambert law, as a function of ratio of the concentrations of the two studied species, increased in ratio of the two extinction coefficients.

$$ {\rm {Area }}\left( {\rm {A}} \right)/{\rm {Area }}\left( {\rm {B}} \right) = {\rm{\varepsilon }}\left( {\rm {A}} \right)/{\rm {\varepsilon }}\left( {\rm {B}} \right){ }.{\rm{ C}}\left( {\rm{A}} \right)/{\rm{C}}\left( {\rm{B}} \right) $$

(7.4)

Assuming that the changes of the ratio ε(A)/ε(B) is small within a same sample’s nature (for example, from one bituminous binder to another), the area ratios can therefore be compared relatively from one spectrum to another, in order to characterise the change of concentrations ratio (Ruau et al. [28]). Assuming that the ethylene groups (CH₂) at 1,460 cm^–1and methyl (CH₃) at 1,375 cm^–1are not significantly altered by oxidation of binders, the oxygenated functions quantities are determined using a semiquantitative approach based on the definition of the two following structural indexes:

• Carbonyl Index :

  $$ {\text{Ico}} = \frac{{{\text{Area}}\;{\text{around}}\;{1,700}\;{{\text{m}}^{{{ - 1}}}}}}{{{\text{Area}}\;{\text{around}}\;{1,460}\;{\text{c}}{{\text{m}}^{{ - {1}}}}{\text{ + Area}}\;{\text{around}}\;{1,375}\;{\text{c}}{{\text{m}}^{{ - {1}}}}}} $$

  (7.5)

• Sulfoxide Index:

  $$ {\text{Iso}} = \frac{{{\text{Area}}\;{\text{around}}\;{1,030}\;{\text{c}}{{\text{m}}^{{ - {1}}}}}}{{{\text{Area}}\;{\text{around}}\;{1,460}\;{\text{c}}{{\text{m}}^{{ - {1}}}}{\text{ + Area}}\;{\text{around}}\;{1,375}\;{\text{c}}{{\text{m}}^{{ - {1}}}}}} $$

  (7.6)