Low-temperature conditions

The temperature of the surface of pavements can vary significantly with the variation of climatic factors such as radiation, wind speed, atmospheric pressure, and low ambient temperature, among others. This surface temperature is mostly related to the surface characteristics of the pavement material as opposed to the heat transfer property of the pavement. According to Wang et al. (2014), the coefficients of thermal conductivity of asphalt mixtures can vary considerably because of the difference in aggregates. In their study on the factors influencing the temperature of the surface of the pavement, they observed that when all external elements were the same, the resulting calculated temperatures of a pavement surface were also the same, despite the difference in coefficients of thermal conductivity. This observation explains the importance the external factors have on determining the temperature of the pavement surface and therefore, the necessity to simulate them during testing conditions accurately.

More substantial variation in temperatures will be recorded at locations closer to the pavement surface, which in turn will lead to higher thermal stresses. Even though the magnitude of the temperature variation in winter is smaller than in summer, the pavement is subjected to freezing and thawing cycles. The most critical condition in the surface layer, in winter conditions, is the diurnal temperature, although both

seasonal and diurnal variations of temperature should be considered when designing a surface layer (Wang et al., 2014).

Low temperature cracking is one of the most common failure modes of flexible pavements in cold climate zones (Alataş and Yilmaz, 2017, Stimilli et al., 2017). These regions are characterised by daily temperature drop that can be remarkably rapid, very long cold seasons and the lowest temperatures experienced can be very low. All these factors will play a role in the continuous maintenance needed to keep the roads in acceptable service conditions, which can result in high direct and indirect costs.

Pavements executed in colder regions are subjected to significant variations in temperature. Not only when it comes to the level of temperature they must endure but also, to the pace of cooling. The variation of this cooling rate can create tensile stresses derived from the characteristic of asphalt mixtures to extend when heated and contract when cooled. These stresses occur because the asphalt layer is constrained in the pavement structure and is unable to relieve the thermal stresses by internal relaxation. As a result of the drop in temperature, thermal tensile stresses increase, and when they exceed the fracture strength of the mixture, cracks due to low temperature may appear (Pszczola and Szydlowski, 2018). Microcracks initially appear on the surface of the pavement when the stress of the pavement exceeds its tensile strength. If the low-temperature cycles continue, which is the usual scenario in colder regions, these microcracks propagate through the pavement leading to a more severe cracking problem in the mixture (Das et al., 2012).

In colder zones, when water fills these cracks during the winter months, it freezes, and ice lenses along with frost heave can develop. This condition results in the loss of fines and formation of voids throughout the pavement, leading to a load-bearing capacity reduction. Thus, it is of crucial importance to take into account all the variables that may affect the pavement in colder regions, during the designing stage. If these are not adequately considered, it can lead to the mentioned formation of cracks that in turn result in the reduced service life of the pavement, high maintenance costs and poor riding quality (Isacsson and Zeng, 1997). This condition, along with the change in the micro-structural stress mechanism, determines the fracture resistance behaviour of asphalt mixtures to temperature (Kim and Hussein, 1997). However, much other pavement distresses, such as rutting, moisture-induced damage, among others, may appear on asphalt pavements that have been in use for an extended period (Alataş and Yilmaz, 2017).

The low-temperature properties of asphalt mixtures can be, somewhat, indirectly derived from bitumen properties. The penetration depth is empirical, albeit only roughly, correlated with asphalt binder

performance (Pavement Interactive, 2019). Despite this, binder testing alone may not be sufficient to predict the effect of improvements, such as additives and modified bitumen, regarding the resistance of the mixture to thermal stresses (Pszczola and Szydlowski, 2018).

In order to accurately assess the factors that affect the asphalt mixtures in colder regions, several laboratory tests have been developed. One of these methods is the thermal stress restrained specimen test (TSRST), which determines the critical cracking temperature that results from a single drop in temperature to an extremely low value observed during severe winters. The test presents many advantages, although with some limitations. Because the failure temperatures obtained from the test depend on the established cooling rate, the results of thermal stress and temperature at fracture should be regarded as a comparative measure between asphalt mixtures (Pszczola and Szydlowski, 2018).

Other characteristics can be tested by different methods, such as tensile strength. One of the methods that describe the direct tensile strength of asphalt mixtures at low temperatures is the uniaxial tension stress test (UTST). Another is the indirect tensile test (IDT), which also measures creep compliance. The bending beam rheometer (BBR), can also be used to obtain the asphalt binder strength at low temperatures. Another method to describe the strength properties of asphalt mixtures at low temperatures is to use flexural strength (bending beam test) at low temperatures.

Fracture properties of asphalt pavements can be defined based on fracture mechanics theory and can be strictly related to laboratory test results. There are several test methods to assess fracture parameters, including the bending of single edge and notched beams (SENB), bending of semi-circular beams (SCB), and tension of disc-shaped specimens (DC-T). One of the most suitable and frequently used methods is the bending test of semi-circular specimens (SCB). For better cracking characterization, more parameters can be assessed, like fracture energy (pre-peak and post-peak), toughness index, and flexibility index, among others (Pszczola and Szydlowski, 2018).

Another indicator of the potential for crack formation is the relaxation potential of a mixture given by the relaxation modulus. It represents the capacity of a mixture to dissipate the thermal-induced stress. This capacity plays a vital role since the higher the relaxation potential, the lower the potential to reach the failure point and thus, to form cracks. Mixtures characterized by faster decay of relaxation modulus are less prone to cracking at low temperature and can be more suitable for use in cold climate regions (Stimilli et al., 2017).