Influence of Freeze-Thaw Cycles on Engineering Properties of Tonalite : Examples from China

0e deterioration of the physical and mechanical properties of tonalites subjected to freeze-thaw cycling under three different temperature ranges was explored using several experimental techniques. Uniaxial compression and three-point bending tests were conducted on untreated and treated tonalite specimens. Clear decreases in uniaxial compressive strength (UCS), Young’s modulus, and fracture toughness were observed in tonalite specimens with frost damage. Although Young’s modulus and fracture toughness did not show clear decreases as the minimum temperature of the freeze-thaw cycle decreased from −30°C to −50°C, the UCS decreased almost linearly. 0e macromechanical characteristics of the tonalites can be explained by changes in mineral content and microstructure. 0e intensity of X-ray diffraction (XRD) peaks of minerals in tonalites that had not been freeze-thaw cycled were approximately 10 to 20 times higher than the peaks for the specimens subjected to freeze-thaw cycling, implying that the internal structure of tonalite is less compact after frost damage. 0e microstructures of the tonalite specimens were also examined using scanning electronmicroscopy (SEM). Increased amounts of fragmentation and breaking of structural planes were observed as the minimum temperature of the freeze-thaw cycle decreased.


Introduction
Weathering, for example from frost damage, heat exposure, or chemical processes [1][2][3], is common in nature.e deterioration of rocks due to weathering has recently attracted increasing concern from both academic researchers and engineers.In China, granitic rock is an important construction material in civil engineering, mining engineering, tunnel construction, and railway engineering.Granitic rock masses occupy approximately 9% of the land area of China, or up to 800000 square kilometres.ere are many tunnels in northeastern and northwestern China, some of which are located in cold regions and experience different degrees of damage resulting from frost action [4].Concerns regarding the effect of cooling and heating on rocks has also arisen in many engineering projects, such as the storage of liquefied natural gas at the extremely low temperature of −160 °C [5].
An understanding of rock properties resulting from freeze-thaw cycling is valuable for engineers working on railway engineering, tunnelling, and underground construction in cold regions.When used as a construction material, tonalite is required to meet high-quality standards to ensure optimum behaviour under all conditions.Cold regions account for approximately 43% of the land area of China.Recently, weathering induced by the action of freezethaw cycles with different temperatures and cycle count has been extensively studied [3,6,7].Compared with other weathering processes, freeze-thaw cycling can lead to a high degree of rock deterioration [8].When porous brittle material such as rock falls below 0 °C, the liquid in micropores can expand by 7% to 9%, concentrating tensile stresses and generating cracks in areas adjacent to the micropores [8][9][10].When the rock mass returns to higher temperatures, the cracks become filled with liquid, which further exacerbates the deterioration of the rock mass [9,[11][12][13]].e effect of freeze-thaw cycling on the physical properties of rocks, including mass, density, P-wave velocity, porosity, and Young's Modulus, has previously been studied in [8,[14][15][16][17].Extensive investigations into the mechanical properties of various rocks subjected to freeze-thaw cycling have examined uniaxial and triaxial compressive strength, tensile strength, and deformation characteristics [18][19][20][21][22].However, the damage mechanism associated with freeze-thaw cycling in granitic rocks, i.e., ice crystallization cracking, is not fully understood, as it depends on a number of factors including temperature range, frequency of freeze-thaw cycles, applied stress, and water composition [23,24].
e physicomechanical behaviours of granitic rocks under macroscales and microscales are di erent but related.Mineralogy, texture, structure, and weathering decide the mechanical properties of granitic rock.However, even when possessing the same mineral composition, the mechanical properties of rocks may vary [25].Freeze-thaw cycles will a ect the microstructural characteristics [26], grain boundaries [27], mineral shape, and spatial arrangement [28] of granite.e e ects of weathering on rocks can be studied by examining mineral composition, phase transitions, and microstructures.X-ray CT imaging has been used to measure crack openings [29,30], but to date, there has been limited research into changes in the microstructure and mineral composition of rocks subjected to freeze-thaw cycling.Changes in mineral composition and di raction intensity of tonalite induced by freeze-thaw cycling can be identi ed using X-ray di raction (XRD). is is an e ective method to analyse and estimate the structural degradation of minerals at a microscale.e physical deterioration of minerals as a result of freeze-thaw cycling can be observed directly using scanning electron microscopy (SEM).In this study, SEM and XRD have been used to investigate rock properties on a mesoscale.
e main objective of this research is to investigate the deterioration of physical and mechanical properties of tonalite resulting from freeze-thaw cycling under di erent temperature ranges.To this end, uniaxial compression and three-point bending tests were conducted on tonalite specimens, and XRD and SEM were used to explore changes in mineral composition and microstructure resulting from freeze-thaw cycling.e ndings of this study regarding the e ects of mineral composition and microstructure on the mechanical properties of tonalite are summarized.

Geologic Background
e rock mass of Nan'an city is the product of late Mesozoic magmatism, and with an age of 135-136 Ma. Figure 1 shows the location of the outcrop where tonalite samples were taken.e geologic structure of the area surrounding Nan'an city is shown in Figure 2. e granitic rock in this region belongs to magmatic rock and has a ne-grained texture and massive structure.e primary minerals are quartz, potassium feldspar, plagioclase, and biotite.e quartz crystals have granular crystals.e majority of potassium feldspar has a hypidiomorphic tabular shape, and Carlsbad twinning can be observed under a microscope.Biotite exhibits a brown colour and eminent cleavage.

Experimental Testing
3.1.Sample Preparation.For uniaxial compressive strength (UCS) tests, the cylindrical specimens were cored perpendicular to the bedding plane from a depth of 2 m and cut to a height of 100 mm (Figure 3).e specimens were then processed to a diameter of 50 mm according to International Society for Rock Mechanics (ISRM) standards.Twenty specimens were prepared for UCS tests.
Single-edge-notched (SEN) specimens shown in Figure 4 were used for the three-point bending test, with a geometry shown in Figure 5. e length, height, and thickness of the SEN specimens were 250 mm, 50 mm, and 50 mm, respectively.In each of the specimens, a preprepared crack with a depth of 26.5 mm was introduced at a location midway between the clamps (Figure 5).Sixteen specimens were prepared for the three-point bending test.

Physical Properties.
e physical properties were determined using standard methods.e recommended ISRM procedures for sample preparation and measurement were used as follows: (1) Dry weight, dry density, and saturated density determined by saturation and calliper method (2) Pore volume values and porosity obtained by water saturation e results of these measurements are listed in Table 1.

Freeze-aw Tests.
e selection of freeze-thaw cycle parameters does not have a global standard.Most of the research has employed freezing and thawing times of 4 hours each, which we follow here.e tonalite specimens were placed in a WGD-501 freeze-thaw device and subjected to freeze-thaw cycling.e specimens were divided into three groups: in all cases, the identical maximum temperature (20 °C) was used, but minimum temperatures varied between groups (−30 °C, −40 °C and −50 °C).e temperature settings used during freeze-thaw cycling are shown in Figure 6.ere 2 Advances in Civil Engineering were four stages to each freeze-thaw cycle: (1) the temperature was decreased from 20 °C to the minimum temperature over 30 minutes; (2) the temperature was maintained at the minimum temperature for 4 hours; (3) the temperature was increased from the minimum temperature to 20 °C over 30 minutes; and (4) the temperature was maintained at 20 °C for 4 hours.Each freeze-thaw cycle therefore lasted 9 hours.is heating-cooling activity was repeated for 30 cycles.For all groups, the durations of stages (1) and (3) were 30 minutes, so the rate of temperature change varied depending on the minimum temperature of the cycle.e freeze-thaw cycle was set to 30 times.

XRD and SEM Test.
e mineral content before and after freeze-thaw cycling was determined using an X-ray di ractometer (Bruker/D8 Advance).Five specimens in each temperature group were prepared for XRD analysis.Advances in Civil Engineering 3 Sample microstructures were observed using the Tescan Vega 3 scanning electron microscope, and 5 specimens in each temperature group were prepared for SEM analysis.

Uniaxial Compression Tests. Uniaxial compression tests
were conducted using an electrohydraulic servo compression test machine manufactured by the Xi-an LETRY Testing

Advances in Civil Engineering
Machine Company, with a load capacity of 2000 kN.Five specimens in each temperature group were prepared for uniaxial compression testing.After freeze-thaw cycling, uniaxial compression tests were performed under load control at a loading rate of 0.5 kN/s until failure.

ree-Point Bending Tests.
ree-point bending tests were conducted using a SANS electrohydraulic servo compression test machine with a load capacity of 2000 kN.Four specimens in each temperature group were analysed.
ese tests were performed under load control at a loading rate of 50 N/s until failure.

Tonalite Mineral Composition.
e mineral composition of tonalite specimens without treatment was determined using XRD analysis (see Table 2).e results were analysed by the Microspectrum Technology Company and show that plagioclase, mica, and quartz are the three main mineral categories in the material studied.
Diffraction peak pattern results for the tonalite before and after freeze-thaw cycling are shown in Figure 7.It is shown that the main mineral composition (plagioclase, phlogopite, and quartz) does not change obviously before and after freeze-thaw cycles; however, peak intensities change.He [31] reported that the intensity of a peak is proportional to the crystallinity of the mineral and reflects the integrity of the mineral's internal structure.According to the data in Figure 6, the intensities prior to freeze-thaw cycling are approximately 10 to 20 times greater than those values recorded after freeze-thaw cycling, implying that the mineral components have less compact internal structure following freeze-thaw cycling.

Stress-Strain
Curves.Axial stress-strain curves obtained from tonalite specimens with and without exposure to freeze-thaw cycling are shown in Figure 8.
e curves corresponding to the samples that have been subjected to freeze-thaw cycling are similar, whereas the curve corresponding to the natural tonalite specimen is quite different.
e figure indicates that a significant decrease in the uniaxial compressive strength (UCS) occurs as a result of freeze-thaw cycling, and as the minimum temperature of the freeze-thaw cycle decreases, the stress-strain response moves further away from the reference curve.In addition, freeze-thaw cycling greatly reduces the strain at UCS, as the untreated tonalite specimens had much higher strain at failure compared to the samples that had been exposed to freeze-thaw cycling.However, the maximum strain of the freeze-thaw cycled specimens does not change significantly with the minimum temperature of the cycle.

Uniaxial Compressive Strength (UCS).
e uniaxial compressive strength (UCS) of tonalite with and without exposure to freeze-thaw cycling is plotted as a function of minimum cycle temperature in Figure 9.Although there is variability in the UCS measurements for each temperature, the overall trend indicates that the UCS declines almost linearly with a decrease in the minimum temperature of the freeze-thaw cycle.Frost damage in rock is primarily caused by the 7% to 9% volumetric expansion of water as it changes to ice [22,32].Considering only this mechanism, the UCS of specimens subjected to freeze-thaw cycling should be the same irrespective of the minimum cycle temperature.
erefore, it can be speculated that there must be an alternative explanation for the observed deterioration in properties.Figure 10 shows the relationship between the diffraction intensity of minerals and the UCS of tonalite specimens.e diffraction intensity of quartz and plagioclase, which has similar trend as the UCS of tonalite specimens, decreases with decreasing freeze-thaw minimum temperature.Compared to the initial state, the intensity of quartz peaks decreases by 94.3%, 97.5%, and 97.7% after exposure to −30 °C, −40 °C, and −50 °C freeze-thaw cycling, respectively.Compared to the initial state, the intensity of plagioclase peaks decreases by 94.6%, 98.3%, and 98.1% after exposure to −30 °C, −40 °C, and −50 °C freeze-thaw cycling, respectively.However, the intensity of phlogopite peaks does not exhibit any clear change.ese results indicate that the microstructure of minerals is closely connected to the mechanical properties of tonalite, especially that of quartz and plagioclase.Before and after freeze-thaw cycling, phlogopite may become the most relatively stable mineral.

Young's Modulus.
Young's modulus of a material is a measure of its stiffness and reflects the condition of its internal structure.In this study, Young's moduli were obtained from stress-strain relationships.Figure 11 shows the measured values of Young's modulus for the granite specimens with and without exposure to freeze-thaw cycling.e data show that there is a slight decrease in Young's modulus caused by the 30 freeze-thaw cycles, indicating a deterioration in the elastic properties of the granite.While the natural granite specimens had an average Young's modulus of approximately 21.92 GPa, those subjected to freeze-thaw cycling had an average Young's modulus of 18.98 GPa, corresponding to a reduction of 13.4%.However, as the minimum temperature of the freeze-thaw cycle decreased from −30 °C to −50 °C, Young's modulus remained constant.If it is assumed that the variation in Young's modulus is mainly related to porosity and microcracking, then it can be argued that the porosity and degree of microcracking are similar after freeze-thaw cycling with minimum temperatures between −30 °C and −50 °C.ree-point bending test was conducted to investigate the propagation of cracks under mode I loading, and the mode I fracture toughness, K IC , was selected to characterize the failure of the tonalite specimens tested.Yin [33] developed a mathematical formula for determining the fracture toughness of tonalite specimens with a single-edged planar crack in three-point bending tests using boundary collocation as described below.

Advances in Civil Engineering
According to Williams's stress function, expression (1) can be used to calculate the mode I fracture toughness, K IC : where K IC is the critical stress intensity factor; W, B, and S are the height, thickness, and length of the specimen, respectively; P is the external load applied to the middle of the specimen, and D 1 is a dimensionless parameter determined by the boundary conditions.In this study, the ratio of S to W is 5 and D 1 can be expressed in terms of a/W (where a is the depth of the single-edged planar crack).erefore, equation Advances in Civil Engineering where F(a/W) takes the form shown in the following equation (from the American Code SEM-E399): e fracture toughness, K IC , is therefore where P max is the peak load measured during testing.e mode I fracture toughness values of the tonalite specimens with and without exposure to freeze-thaw cycling are plotted in Figure 13.
In general, the fracture toughness values of the tonalite specimens subjected to freeze-thaw cycling were lower than those of the natural tonalite specimens.e average fracture toughness values of the −30 °C, −40 °C, and −50 °C minimum temperature freeze-thaw cycled specimens were 86.1%, 82.2%, and 81.8% of the average value for the natural tonalite specimens, respectively.us, as the minimum temperature decreases, the fracture toughness shows a slight decreasing trend.
4.6.Gri th Fracture Principal.According to Gri th's failure criterion [34], the critical fracture stress, σ cr , in threepoint bend testing can be described by where c is half the length of the preprepared crack; E is Young's modulus; and c is the surface energy density, which is related to the surface tension of the material.While [35] provides the surface energy densities for many minerals, that study excludes some granite minerals.
In the three-point bending tests, the crack length was xed.erefore, according to equation ( 5), if the mineral compositions of the tonalites were similar for all specimens, the value of σ cr should only be a ected by Young's modulus after freeze-thaw cycling.e results discussed previously support this hypothesis, as the variation in Young's modulus is similar to the variation in fracture toughness following freeze-thaw cycling.In other words, variation in the minimum temperature of the freeze-thaw cycle between −30  8 Advances in Civil Engineering image of a tonalite specimen without freeze-thaw treatment and Figures 14(b) and 14(c) show SEM images of tonalite specimens subjected to freeze-thaw cycling with minimum temperatures of −30 °C and −50 °C, respectively.Comparison of microstructures suggests that the tonalite specimen subjected to the minimum temperature of −50 °C is less compact and has more fragments and broken structural planes on the surface than the specimens subjected to the minimum temperature of −30 °C with no treatment.e mineral structure of the specimen without treatment is more compact than those exposed to freeze-thaw treatment, and microcracks are much less abundant.
A higher magni cation image of the microstructure of the tonalite specimen subjected to a minimum temperature of −40 °C is shown in Figure 15. is image displays micropores on the surface, some of which are lled with fragments.
Figure 16 shows a lower magni cation image of a phlogopite before and after freeze-thaw cycling with a minimum temperature of −50 °C.After freeze-thaw cycling, cleavage in the phlogopite can be observed.e schistose structure of phlogopite is highly susceptible to shear loading, and an applied force along the surface of schistose will easily cause fracturing.However, as the surface tension and surface energy density of the mineral are not a ected by freeze-thaw cycling, only the UCS of the tonalite is a ected.ere is no obvious degradation of the phlogopite observed after treatment.

Summary and Conclusions
Several techniques were used to examine the deterioration of tonalite resulting from freeze-thaw cycling.Physical and mechanical property data obtained from uniaxial compression Advances in Civil Engineering testing, three-point bend testing, and elastic wave velocity measurements were analysed and compared with the results of mineral composition and microstructure investigations obtained from XRD analysis and microscopy.It was found that there is relationship between the degradation of macromechanical properties and the changes in mineralogy and microstructure of tonalite associated with freeze-thaw cycling.e following conclusions can be drawn from this study: (1) Frost damage decreases the UCS, Young's modulus, and fracture toughness of tonalite specimens subjected to freeze-thaw cycling.e value of UCS decreases almost linearly with a decrease in the minimum temperature of the freeze-thaw cycle.However, Young's modulus and fracture toughness remain relatively stable for minimum cycle temperatures between −30 °C and −50 °C.(2) e minimum temperature of the freeze-thaw cycle affects the compactness of tonalite.Increasing amounts of fragmentation and breaking of structural planes can be observed with decreasing minimum temperatures.ese microstructural surfaces and microcracks generated by freeze-thaw cycling affect the mechanical properties of tonalite, especially its bearing capacity.
(3) ere is a close relationship between the diffraction intensity of minerals and the uniaxial compressive strength of tonalite specimens subjected to freezethaw cycles, especially for quartz and plagioclase.ere is no observable relationship between phlogopite diffraction intensity and the UCS of tonalite specimens.us, phlogopite may possess the highest relative stability during freeze-thaw cycling.

Figure 1 :
Figure 1: Location of outcrop from which tonalite samples were taken.

Figure 2 :
Figure 2: e geologic structure of the area around Nan'an city.
°Cand −50 °C has no discernible e ect on either Young's modulus or fracture toughness.

Figure 14 :
Figure 14: SEM images of tonalite (a) without treatment and after freeze-thaw cycling with (b) a minimum temperature of −30 °C and (c) a minimum temperature of −50 °C.

Figure 15 :
Figure 15: SEM image of tonalite after freeze-thaw cycling with a minimum temperature of −40 °C.

Figure 16 :
Figure 16: SEM images of phlogopite: (a) without treatment and (b) after freeze-thaw cycling with a minimum temperature of −50 °C.