Effect of thermal effect on lattice transformation and physical properties of white marble

1 Introduction

White marble, a kind of carbonatite, exhibits excellent resistance to efflorescence and demonstrates stability under environmental fluctuations. It is predominantly used in numerous stone cultural artifacts that embody significant historical and cultural values [1]. Crafted from white marble, these artifacts represent the ingenuity of ancient artisans; they also serve as crucial testimonies to human civilization. Typically positioned in open-air settings, these cultural relics are exposed to natural efflorescent processes driven by atmospheric conditions such as temperature, humidity, wind, and solar exposure, as well as erosion caused by rainfall. Consequently, most of these artifacts exhibit varying degrees of surface degradation [2–5]. The efflorescence on these surfaces can obscure inscriptions and designs, thereby diminishing their cultural, artistic, and historical significance. Additionally, this process increases the porosity of the artifact surfaces and reduces the cohesive force among the mineral grains. Over time, this degradation can alter the physical and chemical properties of the artifacts, leading to structural weaknesses and potentially causing their deterioration or collapse [6].

The Wanning Bridge, constructed in 1285, features components such as railings and balusters made of white marble. It is one of the 15 heritage elements along Beijing’s central axis and stands as the oldest bridge on this axis. Additionally, it is the sole remaining Yuan Dynasty bridge still serving the city’s transportation needs. However, its preservation condition is poor. The primary cause of these defects lies in environmental factors such as temperature, humidity, and light that promote the weathering of white marble. This process weakens the cohesion of marble particles and reduces their mechanical properties.

To restore the cohesion between particles and the mechanical properties of weathered stone, osmotic reinforcement is commonly employed [3,7]. Prior to application on stone cultural relics, it is imperative that infiltration reinforcement materials undergo thorough testing and evaluation to assess their effectiveness. The ideal evaluation of these materials should be conducted on weathered samples. However, naturally weathered stone samples are often not available in adequate quantities or with consistent characteristics. Therefore, it becomes essential to develop methods for artificially weathering stone to produce specimens that exhibit uniform and reproducible weathering characteristics.

Currently, common methods for inducing stone efflorescence include acid corrosion, natural weathering, ultraviolet exposure, and thermal effects. Acid corrosion efflorescence, notably through immersion in sulfuric acid, nitric acid, or other inorganic acids, is employed to simulate stone weathering processes. Gibeaux et al. exposed limestone to environments containing mixed acid 1 (HNO₃ and H₂SO₃), mixed acid 2 (HNO₃ and H₂SO₄), and simulated rain, subsequently assessing macro- and micro-morphological changes over time. Evaluations included alterations in color, weight, porosity, salt content, and dissolved calcium content, exploring the impacts of nitrogen and sulfur compounds on the stone [8]. Additionally, Gibeaux et al. created weathered stone samples in the laboratory by simulating acid gas exposure and acidified stormwater runoff for scientific analyses [9]. Although this method shows minimal effects on stones primarily composed of SiO₂, it induces the formation of a brittle powder on the surface of calcium carbonate-based stones, such as white marble. In such cases, strong acids may react vigorously with the marble, compromising the structural integrity of the white marble matrix.

Natural and ultraviolet efflorescence primarily involve exposing stone to natural elements such as wind, sunlight, and rain. This process is inherently slow, requiring extended periods to observe significant changes. Lisci et al. [10] conducted comprehensive assessments by measuring gloss and color parameters, analyzing microscopic textures using scanning electron microscopy (SEM), and determining microscopic roughness with a digital roughness meter. Their findings revealed that under conditions of 60°C ultraviolet light at a wavelength of 340 nm with an irradiance of 1.55 W·m⁻², and subsequent water spray of 7 L·min⁻¹ MilliQ-water for 15 min followed by condensation at 50°C for 3 h and 45 min over a 14-day period, the artificial weathering achieved levels comparable to 6 months of natural exposure. Such accelerated artificial weathering methods significantly reduce the duration of testing cycles and expedite experimental progress, though natural weathering still presents challenges due to its slower pace.

Thermal efflorescence primarily involves using heat treatments to elevate the temperature of stone, thereby altering its porosity and other physical properties to simulate weathered conditions for further experiments. Frequent thermal fluctuations are acknowledged as a principal cause of natural stone deterioration [11–13]. Thermal expansion and contraction within the stone can exacerbate existing microfractures and initiate new fractures, resulting in a marked increase in porosity. Consequently, variations in temperature may significantly affect the mechanical properties of natural stone, dependent on the rock type, and the extent and duration of the temperature exposure. Ruffolo et al. subjected Campanian igneous rock specimens to heating at 45°C to prepare weathered samples for subsequent evaluations of reinforcement materials [14]. Roveri’s team produced simulated weathered stone through a combination of heating (65 ± 5°C), ultraviolet irradiation, and rain erosion, investigating the protective capabilities of two silane-based photocatalytic nanocomposites on stone [15].

In contrast to acid weathering, thermal efflorescence does not produce comparable matrix damage and operates at a faster rate than both ultraviolet and natural weathering, making it an exceptionally rapid and effective method. However, research on the optimal thermal weathering temperatures for white marble is currently lacking. This study primarily investigates the effects of heating white marble to temperatures of 150, 200, 250, 300, and 350°C, examining the alterations in parameters such as chroma, gloss, Leeb hardness, mass, porosity, compressive strength, and flexural strength before and after exposure. The physical properties of stone relics in Wanning Bridge were compared with those in Wanning Bridge site to find the heating temperature that is most suitable for laboratory penetration reinforcement test samples. The results indicate that increasing temperatures significantly impact the physical properties of white marble samples: the chromatic disparity enlarges, gloss generally trends upward, and there is a notable increase in mass loss and porosity. Ultrasonic velocity decreases, and Leeb hardness diminishes. Compressive and flexural strengths decline below 300°C; however, these strengths enhance after exposure to 350°C. SEM and X-ray diffraction (XRD) analyses reveal that the crystal lattice structure of the samples changes post-heating at 300°C, influencing their physical properties. At the same time, it is found that a number of data of 300°C heated white marble samples are similar to stone relics at the site of Wanning Bridge, which can be used as test samples in laboratory experiments to verify the reinforcement effect of penetration reinforcement materials. It is expected to provide reference for the preparation of weathered white marble samples in the restoration and protection of white marble relics in Beijing area.

2 Materials and methods

2.1 Materials

In this experiment, two sizes of white marble from the Fangshan area in Beijing were utilized, measuring 40 × 40 × 40 mm³ and 40 × 40 × 100 mm³, respectively. XRD results presented in Figure 1 identify dolomite as the primary mineral component, which consists of magnesium calcium carbonate (MgCa(CO₃)₂). The X-ray fluorescence (XRF) spectrometer analysis of the white marble, as shown in Table 1, indicates that CaO and MgO constitute 61.16 and 36.48% of the composition, respectively. Additionally, there are minor quantities of SiO₂ and trace elements including S, Fe, K, Mn, and Sr.

Figure 1

XRD results of white marble.

Table 1

XRF results of white marble

Element	CaO	MgO	SiO2	SO3	Fe2O3	K2O	MnO	SrO
Percentage	61.16	36.48	1.36	0.48	0.41	0.07	0.03	0.01

The powder sample was put into a porcelain crucible, weighed, and placed in the Muffle furnace, set the temperature at 950°C, burned for 1 h, and then the porcelain crucible was taken out after it cooled to room temperature naturally. The analytical balance was used to weigh and calculate the firing loss. The results of firing loss are shown in Table 2, and the average firing loss of five groups of samples is 45.57%. During calcining, MgCa(CO)₂, the main component of the white limestone, disintegrated to produce CO₂, and the final product was CaO and MgO. According to the molecular weight calculation, the final product was 52.29% and the calcining loss was 47.71%, indicating that there was about 2.14% impurities in the sample which did not decompose at high temperature, which was consistent with the XRF result.

Table 2

Loss on ignition (LOI) results of white marble

Sample	1	2	3	4	5	Average
LOI (%)	45.54	45.63	45.53	45.58	45.57	45.57

3 Results and discussion

Figure 2 illustrates some changes of a white marble sample after undergoing heating at 150, 200, 250, 300, and 350°C. Figure 2(a) indicates that chromatic aberration increases progressively with temperature at 150°C, the aberration is measured at 1.49, while at 350°C, it reaches 3.86, both values lying within the acceptable range for chromatic aberration [16]. Figure 2(b) presents the gloss differences observed after three cycles of heating, displaying a generally upward trend that fluctuates slightly, with values rising from 0.08 to approximately 0.17. These variations are minor and fall within acceptable limits for changes in the glossiness of cultural relics [17]. The alterations in chroma and gloss primarily stem from the slight yellowing of white marble when heated, which impacts both chroma and gloss levels.

Figure 2

Influence of heating temperature on the color: (a) chromatic aberration, (b) gloss difference, (c) mass loss rate, and (d) poriness and porosity change rate (sample quantity n = 25; cycle heating for three times).

Figure 2(c) displays the mass loss rate of the sample after undergoing three heating cycles at various temperatures, demonstrating a correlation between increased temperature and mass loss. Specifically, after three heating cycles at 150°C, the average mass loss rate of the sample is 0.03%. This rate increases to 0.08% after heating at 200°C. Subsequently, the mass loss rate accelerates significantly: it reaches 0.27% at 250°C, 0.38% at 300°C, and peaks at 0.66% at 350°C, highlighting that the most substantial mass loss occurs at 350°C. Despite this loss, the overall rate remains minimal, and there are no significant defects or signs of shedding observed in the sample. Thus, these results indicate that the temperature exerts a minor impact on the structural integrity of white marble samples.

Figure 2(d) depicts the porosity and its rate of change in the white marble sample. Initially, the fresh white marble sample exhibits an average porosity of 0.45%. After heating to 150°C, the porosity increases to 0.52%, corresponding to a change rate of 12.18%. When the temperature reaches 350°C, the porosity further escalates to 0.77%, with the change rate soaring to 65.60%, approximately 5.39 times higher than that observed at 150°C. These data suggest that temperature significantly influences porosity, with higher temperatures resulting in increased mass loss and porosity. This trend may be attributed to the presence of small intrinsic cracks within the white marble, coupled with the material’s inherent anisotropy. Heat exposure causes these cracks to expand outward, thereby increasing porosity and slightly reducing mass. However, the minimal decline in mass suggests that the increase in porosity is not primarily due to particle loss. Instead, upon heating, the original large grains fracture into smaller grains due to significant differences in thermal expansion coefficients and uneven heating. This fragmentation, along with the expansion of cracks and pores, leads to some small grains detaching, contributing to the observed decline in mass.

The intrinsic correlation between porosity evolution and mass loss can be further interpreted through micro-damage mechanisms. As shown in Figure 2(d), the steep rise in porosity exhibits significant synchronicity with the abrupt increase in mass loss rate in Figure 2(c), indicating that both are governed by the same dominant mechanism. Combined with SEM observations (Figure 3(b)), the rapid increase in porosity at this stage is primarily attributed to the bulk detachment of ceramic particles from the matrix interface: when the applied stress reaches a critical threshold, interfacial cracks propagate to the particle–matrix bonding sites, leading to particle detachment and the formation of macroscopic detachment pits, a process directly accompanied by mass loss. Additionally, the connectivity effect of the crack network contributes to porosity increase by propagating existing micro-cracks to form interconnected pores, although this mechanism does not directly contribute to mass loss.

Figure 3

Thermal performance of white marble samples at elevated temperatures: (a) ultrasonic wave velocity and its changing rate, (b) Leeb hardness and change rate, and (c) compressive strength and flexural strength (sample quantity n = 25; cycle heating for three times).

The ultrasonic wave transmission velocity in white marble samples is notably sensitive to variations in water content, micro-cracks, and mineral composition [18,19]. As the heating temperature increases, the rate of change in ultrasonic wave velocity progressively decreases, demonstrating a reduction in the velocity (V ₁) post-heating. For example, at 150°C, the wave velocity (V ₁) is 2.33 cm·μs⁻¹ with a corresponding rate of change (σ) of 91.92%; at 350°C, V ₁ decreases to 1.64 cm·μs⁻¹, and σ drops to 64.92%. This indicates a significant impact of temperature on the integrity of white marble, primarily due to the intensification of crack generation and expansion. The fracturing process essentially involves the breaking of bonds, which requires overcoming an energy barrier. Elevated temperatures provide the necessary energy to surpass this barrier, thereby facilitating crack formation, which leads to the disintegration of larger grains into smaller ones. Additionally, the disparity in the coefficient of thermal expansion between directions along and perpendicular to the fissure causes the crack to spread outward as temperature rises, thereby enlarging the internal voids and increasing porosity within the marble. Since ultrasonic waves propagate more rapidly through solid media than through gaseous media, the presence of increased porosity and cracks necessitates a longer path for wave transmission, culminating in a reduction in ultrasonic wave speed. Given that the initial ultrasonic wave velocity (V ₀) of the white marble sample is within the range of 2.54 ± 0.15 cm·μs⁻¹, and V ₁ consistently decreases while V ₀ remains essentially unchanged, the ratio of ultrasonic wave velocity (σ) continues to decline.

As shown in Figure 3(b), when the sample is heated to 150°C, the Leeb hardness change rate (τ) is observed to be 5.25%. At 350°C, τ escalates to 30.66%, demonstrating a gradual decrease in H ₁ and an increase in τ, which indicates a significant influence of temperature on the Leeb hardness of the sample. The primary cause of this phenomenon is the alteration in the internal structure of white marble upon heating, characterized by an increase in pores and cracks. Heating accelerates the formation of new cracks and the extension of existing ones. As these cracks proliferate, porosity within the sample increases, consequently leading to a reduction in Leeb hardness.

Figure 3(c) illustrates the changes in compressive and flexural strength of white marble samples after being subjected to different heating temperatures. For samples heated from 150 to 300°C, compressive strength progressively declined from 125.66 to 97.56 MPa. However, for the sample heated at 350°C, an increase in compressive strength was observed. Similarly, flexural strength decreased with rising temperature, reaching a minimum of 6.25 MPa at 300°C, but exhibited an increase to 8.63 MPa at 350°C, mirroring the trend seen in compressive strength. The primary cause of these variations is the differential thermal expansion coefficients of the grain in each direction within the sample. As the heating temperature increases, temperature-induced extrusion and breakage of the white marble crystals intensify. Furthermore, the tensile and compressive stresses generated by existing and temperature-induced cracks, both along and perpendicular to the crack directions, enhance the likelihood of extrusion and breakage. This process contributes to the failure of interparticle cementation and a decrease in the local bearing capacity within the extrusion and breakage zones, leading to reduced compressive and flexural strengths [20]. However, as the temperature increases, recrystallization within the white marble crystals occur [21,22], mitigating some of the residual stresses. Concurrently, the fragmentation-induced small particles may fill existing cracks in the white marble, thus potentially increasing both compressive and flexural strengths to some extent.

Figure 4(a) illustrates that after heating at 150°C, the surface of the white marble sample develops lamellar and blocky features resembling “steps.” As depicted in Figure 4(b), these step-like structures gradually diminish when the temperature is increased to 200°C. At 250°C, significant grain shedding becomes evident, as shown in Figure 4(c). Upon heating to 300°C, the step-like platforms disappear, cracks become pronounced, and fine particles emerge on the surface of the larger grains, as illustrated in Figure 4(d) through (f). These observations suggest the initiation and expansion of cracks, with larger grains splitting into several smaller ones. At 350°C, the surface appears smoother than at 300°C, as depicted in Figure 4(g). Despite the presence of finer particles, the surface retains more flat areas, likely due to recrystallization and new lattice surface formation. Additionally, Figure 4(h) shows some pores and cracks being filled by the fragmented smaller particles. Thus, up to 300℃, the compressive and flexural strength of the samples gradually decline with the increase in temperature. However, upon reaching 350°C, both the compressive and flexural strength exhibit a slight increase.

Figure 4

The internal morphology of the sample after heating at different temperatures: (a) 150°C (×500), (b) 200°C (×500), (c) 250°C (×500), (d) 300°C (×300), (e) 300°C (×500), (f) 300°C (×1,000), (g) 350°C (×500), and (h) 350°C (×2,000).

At 300°C, SEM observations highlight pronounced intergranular cracking (Figure 3(a)) and dispersed fine debris on the surface, originating from micro-fracture along weak mineral interfaces. This morphology signifies the initiation of stress-induced damage, where cracks act as stress concentrators, diminishing the marble’s effective load-bearing cross-section. Figure 3(c) shows a 28% reduction in compressive strength at this stage, directly correlated with increased crack density. The fine debris, likely from particle detachment, further weakens the intergranular bonding. At 350°C, recrystallization is evident, characterized by small, equiaxed calcite grains (Figure 3(b)). This process partially repairs micro-cracks generated at lower temperatures and strengthens grain boundaries via reduced defect density. Consequently, compressive strength recovers by 15% compared to the 300°C state (Figure 3(c)), highlighting recrystallization’s role in mitigating thermal damage. This behavior aligns with the “dynamic recovery” theory in polycrystalline materials.

Figure 5 displays the XRD patterns of white marble samples heated at 250, 300, and 350°C. The samples heated to 250 and 350°C exhibit diffraction peaks at 2θ values of 22.2°, 24.4°, and 26.8°. The peaks corresponding to the 350°C heating are notably weaker at 2θ = 26.8°. These three diffraction peaks align with the crystallographic planes (101), (012), and (002) of MgCa(CO₃)₂. However, the sample heated to 300°C shows peaks only at 2θ = 22.2° and 24.4°, with diminished intensity. This pattern suggests an absence of the (002) lattice, likely due to thermal disruption. Thermal loads on marble induce various physical effects at the mineral particle scale [5,23]. One plausible explanation for the observed changes is that white marble crystal particles expand differently along their main axes due to varying thermal expansion coefficients. This anisotropy leads to significant tensile or compressive stresses at the particle boundaries during heating and cooling cycles, weakening interparticle binding forces and promoting crack formation and growth. Moreover, elevated temperatures provide the energy needed to overcome the bond energy barrier, thereby damaging the lattice surface of (002). The expansion of existing cracks and the induced crystal phase changes may explain why the (002) plane does not appear in the sample heated to 300°C.

Figure 5

XRD patterns of samples heated at 250°C, 300°C, and 350°C.

At 350°C, a weak diffraction peak at 2θ = 26.8° suggests recrystallization, likely forming smaller grains of the (002) crystal plane. While recrystallization initiates, it remains kinetically constrained, yielding small-sized calcite nuclei with random crystallographic orientations. In contrast, the intense (002) peak at 250°C corresponds to a dominant (002) preferred orientation in larger grains, originating from the original marble’s sedimentary fabric. These newly formed particles could fill existing cracks and pores, potentially restoring some of the (002) lattice structure. Consequently, this temperature may enhance the compressive and flexural strength of the white marble samples.

As shown in Figure 6, at 300°C, the main component of white jade is MgCa(CO₃)₂, which will decompose under the action of thermal effect to produce calcium carbonate, magnesium oxide, and carbon dioxide. The reaction equation is as follows:

Figure 6

Main physical and chemical reaction degrees of white marble at 300°C.

In addition, as a metamorphic rock, white jade itself will carry a small amount of organic matter during the formation process, at 300°C, organic matter reacts, become water and carbon dioxide, leaving white jade stone. Therefore, at 300°C, there is not only a lattice transformation, but also a physico-chemical reaction, and the two work together to reduce the physical properties of white marble stone.

As shown in Table 3, the Leeway hardness of the artificially weathered sample heated at 300°C is similar to that of the Wanning Bridge fence plate. The average Leeway hardness of the white marble fence plate of Wanning Bridge is 354 HL, while that of the white marble sample heated at 300°C is 360 HL. The hardness value is similar, indicating that weathering caused by heating reduces the surface hardness of the sample, which is similar to the hardness of the white marble fence board preserved in the actual environment. In addition, as shown in Table 3, the ultrasonic wave velocity of the two is close. The ultrasonic wave velocity of the white jade fence measured in the field is 1.84 cm·μs⁻¹, and that of the white jade heated artificially at 300°C is 1.82 cm·μs⁻¹. The results show that the Leeb hardness of white marble samples heated at 300°C is similar to the ultrasonic wave speed of Wanning bridge fence plate, which can be used as a simulation material for testing the effect of penetration reinforcement. The value of ultrasonic wave velocity is similar, which indicates that the volume of pore or non-dense area in the sample is similar to that of white marble fence plate in Wanning Bridge site. Therefore, the white marble sample heated at 300°C can be used as a simulation material for laboratory testing the effect of penetration reinforcement.

4 Conclusion

This study demonstrates that heating serves as an effective and reproducible method for producing artificial weathering effects on white marble samples, facilitating the preparation of simulated weathered specimens for heritage conservation research. Notably, heating does not induce significant defects or disintegration. It leads to a decrease in several parameters such as chroma, gloss, porosity, Leeb hardness, and mechanical properties. Specifically, for the white marble from the Fangshan area of Beijing, heating at 300°C results in the most pronounced decline in these parameters, thereby causing the most severe weathering effects. At this temperature, the Leeb hardness and ultrasonic wave velocity are similar to that of Wanning bridge railings, which can be used to simulate weathering samples in stone cultural relics protection and restoration. It provides a method for the preparation of simulated weathering samples for the restoration of white marble cultural relics in Beijing area, and also provides a reference for the restoration and protection of stone cultural relics in the world to a certain extent. Additionally, the controlled nature of heating allows for the adjustment of conditions to achieve the desired level of weathering in the samples.

The impact of temperature on the physical and mechanical properties of white marble was investigated through heating experiments on fresh dolomite samples. The results indicate that the critical temperature for lattice transition in white marble is 300°C. Below this critical temperature, the primary physical changes in the rock samples are predominantly due to the uneven expansion and contraction of the crystal particles upon heating. When the temperature exceeds 300°C, recrystallization may occur. The observations from SEM corroborate the XRD findings. This thermal process enhances the compressive and flexural strengths of the marble, with significant alterations observed from the microscale to the macroscale in the properties of the material. In sum, the thermal aging process of white marble shows a temperature threshold effect: at 150–250°C, the physical damage is main, and the porosity and color difference increase slowly. Near 300°C: the physical–chemical synergies are significant, and the crystal form damage and strength drop sharply. The temperature continued to rise to 350°C: recrystallization and pore filling partially offset the damage, and the strength recovered slightly.

Although this study revealed the law of the influence of thermal effect on the lattice transformation and physical properties of white marble, it still had the following shortcomings: the experiment only simulated a single thermal effect, did not consider the synergistic effect of environmental factors such as humidity, acid rain, and ultraviolet light, and the number of heating cycles was limited, which failed to fully simulate the long-term stability of natural weathering. The mineralogical mechanism analysis relies on XRD and SEM characterization, and lacks the dynamic tracking of phase transition kinetics and crack propagation by in situ high-temperature XRD and molecular dynamics simulation. This study systematically investigates the thermal effect on lattice transformation and physical properties of dolomitic marble from Fangshan. The research object is limited to the dolomitic marble of Fangshan, and the generalizability of the conclusions to other types of marble needs to be further verified. Future research can focus on the following directions: to carry out multi-field coupling experiments of temperature, humidity, and pollutants, and optimize artificial weathering conditions combined with on-site microenvironment data of Wanning Bridge. In situ high temperature XRD, synchrotron radiation CT, and nanoindentation techniques were introduced to analyze the crystal phase evolution and the mechanism of grain binding force. Using the samples heated at 300°C as the carrier, the suitability of reinforcement materials such as silane and nano-calcium carbonate was systematically evaluated, and the composite repair process of “thermal pretreatment + penetration reinforcement” was explored. The standardized process of artificial thermal weathering has been Eq.ted, and a multi-scale model of “mineral composition – structural evolution – performance degradation” has been constructed in conjunction with multidisciplinary forces to provide theoretical support and technical standards for the protection of stone cultural relics.