Abstract
The tunnel drainage system in limestone areas is highly susceptible to calcium carbonate crystallization clogging, which abnormally increases pore water pressure and threatens structural safety. Although traditional passive dredging methods are widely used, there is a critical research gap regarding the dynamic evaluation of complex material interventions under active flowing karst water. To elucidate the internal clogging mechanism and propose an “active prevention” material-source control strategy, this study combined on-site sampling from the Hubeishan Tunnel of the Guangzhou-Shenzhen Expressway with large-scale dynamic physical model simulations under single-factor controlled conditions. The results demonstrate that: (1) The clogging material is predominantly calcite (>98% purity), originating from the continuous leaching of free calcium in highly alkaline shotcrete. (2) The accelerator exhibits a nonlinear regulatory effect; a 10% dosage serves as the optimal independent threshold to prevent severe calcium leaching caused by micro-defects at higher dosages. (3) For water reducers, a 5% dosage is identified as the optimal independent threshold, whereas increasing the dosage to 10% triggers a severe crystallization outbreak, indicating the highest clogging risk. (4) Fly ash demonstrates a robust linear inhibitory effect on crystallization accumulation, with a 20% dosage acting as its optimal independent threshold. Furthermore, aggressive highly mineralized water (e.g., NaHCO3-type) directly skips atmospheric CO2 diffusion, rapidly reacting with leached calcium and exponentially accelerating the clogging process. These findings shift the engineering paradigm from passive mitigation to active source interception, providing critical quantitative references for suppressing early-stage calcium leaching and laying a theoretical foundation for future multi-factor orthogonal mix design in deep-buried karst tunnels.
1 Introduction
In the process of advancing modern transportation infrastructure, high-speed rail and highway networks will inevitably extend to water-rich mountainous areas characterized by karst, limestone, and dolomite, and the hydrogeological problems encountered in tunnel engineering have become particularly complicated. Traditional mountain tunnels generally follow the principle of “giving priority to drainage and combining prevention and drainage”, and build groundwater drainage pipe network through circumferential, longitudinal blind pipe and central drainage ditch (; ; ; ; ; ). However, in the limestone soluble rock region, the groundwater that has experienced the action of water and rock for a long time is often rich in high-concentration fouling precursors such as carbonate and calcium ion () (; ; ; ; ; ). When this highly mineralized karst water penetrates the primary support shotcrete and finally converges into the drainage pipe network, it readily triggers intense and continuous calcium carbonate (CaCO3) crystallization. The blockage of tunnel drainage system will greatly reduce the over-current section and fundamentally reduce the hydraulic efficiency. After the failure of the drainage system, the huge pore water pressure will act on the secondary lining, causing structural cracking, leakage of construction joints and even instability, threatening the safety of driving. Traditional passive treatments, such as mechanical dredging and high-pressure water flushing, merely address the symptoms and are prone to short-term relapse (; ; ; ).
Analyzing the deep mechanism of crystallization blockage is an interdisciplinary problem, involving inorganic chemistry, computational fluid dynamics, building materials science and other fields. Since the 1970 s, international academic circles have laid a solid theoretical foundation for this field. In 1974, Berner and Morse conducted calcite dissolution and crystallization dynamics tests in simulated seawater, which confirmed that ion mass transfer rate and boundary layer diffusion thickness played a decisive role in speed limiting in the initial reaction stage (). Based on this, Plummer, Wigley and Parkhurst(1978) put forward the famous PWP kinetic model, pointing out that the dissolution and crystallization of calcite are composed of hydrogen ions, carbonated molecules, microscopic surface-controlled reactions driven by water molecules (). Simultaneously, Reddy et al. confirmed through crystallization tests containing trace phosphate ions that even in a micro-alkaline water environment (pH 8.8), specific anions can strongly inhibit CaCO3 nucleation and precipitation at the microscopic interface (). In the late 1980s, Larsen’s team further concluded from laboratory core tests that the growth rate of CaCO3 single crystals is strictly linearly related to the ion activity product and exhibits extreme sensitivity near thermodynamic saturation (). However, the basic physical chemistry theories of this period were mostly limited to ideal static aqueous systems and single-chemical interference tests, failing to reflect the non-linear disturbances caused by multi-phase fluid environments and long-term pressurized seepage of natural karst water in tunnel engineering, making them difficult to apply directly to predicting real macro-scale pipe clogging disasters.
With the maturation of basic theories, research in the 21st century began to directly address macro-scale tunnel defects and network hydrodynamic evolution. In 2004, Xu and Li confirmed that the intense degassing effect caused by tunnel excavation releases a large amount of CO2, leading to a sharp rise in pH and directly triggering crystallization outbreaks (). In 2014, Shin’s team quantified the mechanical deterioration mechanisms of double-layer lining instability and cracking after network blockage induced excess pore water pressure (). In 2019, Galan et al. comprehensively assessed the macro-durability of underground shotcrete support, explicitly identifying chemical erosion and irreversible leaching of free calcium as core pain points limiting the life cycle of underground projects (). Regarding network structure and fluid dynamic response, Chen et al. (2018) confirmed that the degree of crystallization is strictly controlled by the permeability coefficient of the surrounding rock and the tunnel geometry (). In 2021, Huang Zhouyi’s team innovatively introduced the Diffusion Boundary Layer (DBL) theory, quantifying the concentration gradient and mass transfer resistance characteristics of karst water within the pipe wall attachment layer (). In the same year, Xing’s team found that macro-flow velocity and shear stress within the pipe play a more core role in peeling off initial crystal attachments than the rise in pH (). Concerning disaster diagnosis and new pipe materials, Reinprecht and Pettauer et al. (2025) introduced continuous hydrochemical monitoring in the Austrian transportation drainage network, successfully achieving very early warning of chemically triggered clogging processes (). Similarly, Liu et al. recently developed an adaptive anti-crystallization drainage pipe based on flexible polymers (e.g., EVA, TPU), utilizing pipe wall micro-expansion deformation under dynamic flow to achieve extremely low crystal accumulation (). Despite significant progress in macro-defect investigation, hydrodynamic simulation, and pipe material innovation, most existing engineering prevention measures remain at the passive stage of network optimization or mechanical cleaning. There is a severe lack of targeted source tracking and chemical intervention for the “material source” of crystallization precursors (i.e., the primary support material itself), and the overall management approach remains limited by its “symptomatic” nature.
Based on the scientific consensus of “material source tracking,” blocking the crystallization material supply chain and micro-modifying shotcrete have become the most cutting-edge breakthroughs in the field. In 2020, Eichinger and Boch’s team confirmed through isotope tracing that the high-alkaline mother liquor (pH up to 11) triggering calcite outbreaks is purely driven by intense shotcrete leaching (). Later, Sakoparnig et al. (2021) confirmed through in situ tests that replacing part of clinker with mineral admixture can greatly reduce the generation of calcium hydroxide in concrete and significantly weaken the subsequent crystallization potential (). The Mittermayr team (2023) analyzed the pore characteristics of low clinker sprayed concrete and quick-setting slurry, and established the direct correspondence between the compactness of cement matrix and the long-term calcium loss potential (). Di et al. (2023) focused on the hardened cement paste under the coupling effect of calcium dissolution and dry-wet cycle, and found that this double degradation would obviously accelerate the expansion of internal micro cracks and the decline of mechanical properties (). Chen et al. (2024) found through the simulation of high-pressure water environment that the microenvironment changes caused by calcium dissolution will greatly promote the invasion and transmission of secondary corrosive ions (). In the aspect of dissolution dynamics model construction, An et al. (2025) confirmed the fundamental influence of different ion transport modes on calcium dissolution rate of cementing materials, and revised the traditional single diffusion theory (). On the induction mechanism of admixtures, Niu et al. (2024) deeply analyze the micro-effect of accelerator on calcium dissolution, and point out that non-alkali and low-alkali accelerator can promote early hydration at the same time, it has different effects on pore compactness and dissolution resistance (). Zhao et al. (2024) further investigated the coupled degradation mechanism of liquid accelerators under dual deterioration from sulfate attack and calcium leaching (). On the development of specific anti-leaching additives, Tong et al. (2024) conducted extensive comparative tests, confirming that introducing an optimal 14% dosage of anti-alkali agents and nano-silica can significantly inhibit calcium ion release in shotcrete (by up to 68.4%) (). In 2025, Jiang Yajun’s team also revealed the powerful pozzolanic effect of waste glass powder as a new admixture in reshaping concrete microstructure and enhancing calcium leaching resistance (). Theoretically, Bae’s team (2025) clarified the underlying constraint mechanism of early C-S-H gel evolution on calcium loss (), while Liu’s team (2025) pioneered the introduction of molecular dynamics simulations to precisely calculate the binding energy differences between different interface materials and CaCO3 molecules (). Although recent research has achieved single-point breakthroughs in microscopic material modification and source inhibition, most material tests are still limited to beaker-level static single-variable immersion. There is a severe lack of physical coupling verification involving highly aggressive natural karst water (e.g., high concentration NaHCO3) and extremely complex composite admixture systems within a real dynamic pipe network long-period flow field. Consequently, existing theories struggle to provide reliable optimal mix proportion safety thresholds for frontline engineering.
Synthesizing the literature evolution over the past half-century and the systemic limitations exposed at various stages, it is clear that thoroughly resolving the crystallization problem in limestone tunnels requires breaking the research barrier between “macro passive desilting” and “microscopic static material testing.” Accordingly, the core scientific question of this study is not merely to provide empirical engineering optimization, but to fundamentally reveal the governing mechanism of the calcium leaching-crystallization coupling process. Relying on deep tunnels such as the Hubeishan Tunnel located in a typical active limestone area, this study first conducted field in situ crystal phase collection and refined microscopic characterization for source tracking to identify the core precursor components of aggressive water. Building on this, we breakthroughly developed a high-fidelity “Tunnel Seepage-Crystallization Physical Model Dynamic System,” innovatively introducing a high-risk NaHCO3 aggressive simulant liquid. Under single-factor controlled conditions, long-term continuous hydraulic circulation dynamic disturbance tests were conducted on concrete blocks with varying independent gradients of accelerators, water reducers, and fly ash. By utilizing these material variables to intervene in the calcium leaching chain, the study aimed to precisely capture and quantify the evolution of pipe network crystallization mass extremes under the dynamic hydraulic stripping effects. Rather than proposing a definitive composite mix proportion, this study focuses on elucidating the independent influence weights and determining the optimal individual safety thresholds of each admixture to maximize the inhibition of early calcium leaching. Ultimately, this research not only fills the gap in dynamic crystallization time-series data under coupled multi-phase flow fields, but also realizes a technical strategic leap from “passive mechanical dredging” to “active material source blocking.” The mechanistic insights derived herein provide a solid theoretical foundation for future multi-factor orthogonal mix designs and the anti-crystallization design of high-karst deep tunnels.
2 Materials
The raw materials used in the Test mainly include three parts: first, the original crystals and karst water samples collected on the project site; Second, the concrete mixture and various admixtures used to simulate the initial branch of the tunnel; the third is the high purity chemical reagent used to synthesize specific erosive groundwater.
2.1 Field original crystal and water sample collection
Relying on the Hubeishan Tunnel of the Guangzhou-Shenzhen Expressway. During construction, large-scale white crystals appeared in the circumferential and longitudinal blind drainage pipes, causing severe reduction of the water flow cross-section in some sections. Multiple sampling points were set up at the Hubeishan Tunnel to extract solid crystals attached to the pipe walls and suspended particles in the water. For cross-geographical and hydrological comparison, supplementary sampling was conducted in the Yangpeng Tunnel of the Jingzhu North Expressway and relevant tunnels of the Kaichun Expressway (Sample groups A-D are drainage pipe crystals; E-F are Yangpeng and Kaichun Tunnel pipe crystals), as shown in Figure 1. The basic environmental parameters of each sampling point are shown in Table 1.
FIGURE 1
TABLE 1
| Location | Liquid samples | Crystal state | Temp (°C) | pH |
|---|---|---|---|---|
| Hubeishan tunnel | 3 | Flaky, slightly loose structure, pure white/yellowish | 20.3 | 10.3 |
| Jingzhu North | 2 | Blocky, dense, yellowish | 21.5 | 9.8 |
| Kaichun Hwy | 1 | Blocky, powdery, yellowish | 23.6 | 11.6 |
Overview of sampling sites and parameters.
2.2 Primary support simulating materials and admixtures
Wet-mix shotcrete used in tunnel construction is the engineering substrate in direct contact with seepage groundwater and is the largest potential soluble material source in crystallization clogging. To replicate the seepage extraction process in concrete pores, standard-sized concrete blocks were cast according to the actual field mix proportions. Given the critical role of admixtures in altering hydration rates, pore structures, and solution rheology, this study selected three essential tunnel construction admixtures as experimental variables:
Accelerator: Produced by Guangzhou Suli Construction Materials Technology Co., Ltd. Technical indicators are shown in Table 2.
TABLE 2
| 1day strength (MPa) | 28days strength (MPa) | Density (g/cm3) | Solid content (%) | pH | Total alkali (%) |
|---|---|---|---|---|---|
| 11.0 | 79 | 1.350 | 35.54 | 1.03 | 1.75 |
Accelerator performance indicators.
Fly Ash: Class I Fly ash (Shenzhen, Guangdong, China). Technical specifications are in Table 3.
TABLE 3
| 45 μm screening fineness (%) | Stability (mm) | Water demand ratio (%) | SO3 (%) | Moisture (%) | Ignition loss (%) | Free CaO (%) |
|---|---|---|---|---|---|---|
| 11.0 | 0.5 | 94 | 0.74 | 0.1 | 4.41 | 0.8 |
Fly ash technical indicators.
Water Reducer: A polycarboxylate-based superplasticizer (BASF, Germany) with a mortar water reduction efficiency of at least 14%.
2.3 Chemical reagents for external water quality simulation
To simulate natural karst and aggressive groundwater in a controlled laboratory environment, simulant liquids with specific ionic compositions must be prepared. While using field water is most authentic, the large volumes required and their susceptibility to degradation over time led to the choice of high-purity chemical reagents. Based on hydrogeological data and titration analysis, limestone groundwater is rich in bicarbonate (HCO3−), which triggers intense CaCO3 precipitation upon leaching concrete. Therefore, sodium bicarbonate (NaHCO3) and calcium chloride (CaCl2) were selected as the core reactive substances (Analytical Reagents from Sinopharm Chemical Reagent Co., Ltd.). Technical parameters are in Tables 4, 5. The reagents used are shown in Figure 2.
TABLE 4
| Appearance | Density (g/cm3) | Melting/Boiling point | Flash point (°C) | Solubility |
|---|---|---|---|---|
| White crystalline powder | 2.22 | Decomposes upon heating | 169.8 | Soluble in water; slightly in ethanol; insoluble in ether |
NaHCO3 technical parameters.
TABLE 5
| Grade | CaCl2 content (%) | Alkalinity as Ca(OH)2 (%) | Total alkali chlorides (%) | Water insoluble | pH | MgCl2 (%) | CaSO4 (%) |
|---|---|---|---|---|---|---|---|
| Spherical | 75.56 | 0.05 | 3.26 | 0.001 | 9.6 | 0.03 | 0.035 |
CaCl2 technical parameters.
FIGURE 2
3 Experimental program
3.1 Solid crystal and liquid water testing
Standardized tests were performed on field-collected samples to investigate their physicochemical properties. To avoid structural damage to hydrated minerals (e.g., ettringite) from heat, solid crystals were freeze-dried.
Micro-morphology and Elemental Analysis: Samples were gold-sputtered and observed using a Field Emission Scanning Electron Microscope (FE-SEM). EDS was used for elemental mapping (Ca, C, O, Mg) to infer compound composition. The electron microscope is shown in Figure 3.
FIGURE 3
Phase Identification (XRD): Ground samples (200 mesh) were scanned using an X-ray Diffractometer. Data were processed using Jade software to calculate phase fractions.
Water Chemistry Analysis: Filtered/diluted field water was analyzed via ICP-MS for ions. pH was measured using double-indicator acid-base titration. The test sample is shown in Figure 4. Sampling and testing equipment are listed in Table 6.
FIGURE 4
TABLE 6
| Equipment | Model | Application |
|---|---|---|
| Plastic tubes | — | Sample storage |
| Scraper | — | Sampling |
| pH tester | Sima PH848 | Field pH and temp measurement |
| XRD | D8 ADVANCE | Phase identification/Semi-quantitative analysis |
| FE-SEM | SU8220 | Microstructure and EDS analysis |
| ICP-MS | iCAP RQ | Groundwater ion analysis |
| Freeze Dryer | LABCONCO 4.5L | Sample drying |
Sampling and testing equipment.
3.2 Indoor simulation testing
To replicate the pipe network scaling process, a self-developed “Tunnel Seepage-Crystallization Physical Model System” was used, consisting of a water supply, frequency-controlled circulation pump, reaction zone, and test pipe modules (Figures 5, 6).
FIGURE 5
FIGURE 6
3.2.1 Specific experimental protocols
Initialization and Simulation: After cleaning with deionized water, NaHCO3 and CaCl2 were dissolved in separate tanks (flow rate 2300 L/h).
Admixture Disturbance: Concrete blocks with varying dosages of accelerators, water reducers, and fly ash were placed in the reaction zone. Simulant liquid circulated for 1–2 months. Pipes were dried (60 °C) and weighed periodically for crystal accumulation. Effluent was monitored for pH, residual Ca2+ (EDTA titration), and total alkalinity. The test matrix is shown in Table 7.
TABLE 7
| Test group | Experimental gradients (wt%) | Testing indicators |
|---|---|---|
| Accelerator | 6, 10, 20 | Crystal mass (dried pipe weighing); effluent pH; residual Ca2+ concentration (EDTA); total alkalinity (as CaCO3) |
| Water reducer | 6, 10, 20 | |
| Fly ash | 6, 10, 20 | |
| Water quality | Normal vs. NaHCO3-type |
Experimental matrix for admixture disturbance.
4 Results and discussion
4.1 Solid crystal and water sample analysis
4.1.1 Crystal characterization
Electron microscopy (SEM) was conducted on samples from three different tunnels, as illustrated in Figure 7.
FIGURE 7
Group A samples were taken from the tunnel section which was used as lining earlier, and their microscopic morphology (Figure 7a) was highly consistent with the calcite deposited in the natural karst cave, with dense crystal structure. These crystals directly grow in situ on the drain pipe wall without going through transportation or secondary deposition. With the continuous supply of Ca2+ from groundwater, calcite crystals constantly develop and cross-link, eventually forming a rigid blockage. In contrast, the SEM images of Group D (section with only primary support and waterproof boards), Group E, and Group F (freeze-dried suspended solids from water samples) (Figures 7b-d) exhibit consistent loose, granular characteristics. This shows that these crystals do not grow through surface attachment, but migrate and settle in the tube after precipitation in the bulk solution. Combined with EDS energy spectrum and SEM observation results (Figure 8), we can see that there are a large number of calcite deposits around acicular ettringite. According to the research, although the content of ettringite is relatively low, it has a considerable catalytic effect on the precipitation and accumulation of CaCO3 salts in the drainage system.
FIGURE 8
In general, there are two different crystallization differentiation mechanisms in the same tunnel drainage system: “attachment growth” and “migration deposition”. Therefore, the prevention and control measures should be treated differently: for the attached growth type crystals with strong adhesion, the conventional low pressure flushing basically has no effect; For the migrated deposition type crystals, emphasis should be placed on optimizing the structural hydraulic efficiency of the drainage system, to control particle settlement.
As shown in Figure 9, the main component of the crystallization product is high purity calcium carbonate with purity over 98%. Comparing the XRD pattern of the crystalline sample of the drain pipe with the standard pattern of calcite, we can find that the position and shape of the two characteristic peaks are highly consistent. This proves that the phase composition of crystals in the drainage system is mainly calcite.
FIGURE 9
4.1.2 Drainage water analysis
Water samples from the Hubeishan, Yangpeng, and Kaichun Tunnels were analyzed, with ionic concentrations detailed in Table 8.
TABLE 8
| No. | pH | Mg2+ | Ca2+ | Na+ | Cl− | SO42- | Ba2+ | CO32- | HCO3− |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 9.0 | 25.786 | 180.553 | 2.416 | 90.2 | 446.7 | 0.108 | 108.6 | 237.5 |
| 2 | 8.7 | 48.608 | 511.991 | 1.2 | 77.9 | 87.3 | 1.318 | 95 | 734 |
| 3 | 8.3 | 23.827 | 458.48 | 0 | 103.9 | 128.2 | 0.895 | 112.4 | 566.3 |
| 4 | 9.8 | 8.042 | 300.15 | 3.469 | 172.3 | — | 5.26 | 314.5 | 634.5 |
| 5 | 11.4 | 62.917 | 161.98 | — | 393.2 | 1317.9 | — | 192 | 115.9 |
| 6 | 11.3 | 48.216 | 103.192 | — | 306.6 | 1414.2 | — | 108 | 597 |
Ionic concentration.
The test results indicate that the average pH of the groundwater is 9.44, exhibiting strong alkalinity. The aggressive CO2 content is 3.75 mg/L, with a total hardness of 285.6 mg/L and a mineralization degree of 160.8 mg/L. The hydrochemical profile is predominantly of the Ca2+- Mg2+- HCO3− type, with relatively high concentrations of Cl− and SO42-.
Groundwater in these sections is typical corrosive karst water. During rainfall, surface water permeates along the rock fissures, and the metal cations (Ca2+, Mg2+) and CO2 are filtered into the underground water system. This process breaks the original ion balance and changes the pH environment, thus accelerating the precipitation and accumulation of crystals. In the long-term infiltration area of soluble rocks, karst development is strong and joint fissures are fully developed, providing a priority channel for groundwater circulation. This ensures that the fouling precursor is continuously supplied to the drainage system, eventually leading to system failure. Analysis proves that the high pH value of groundwater is one of the core factors causing rapid crystallization blockage.
4.2 Model simulation testing
4.2.1 Influence of accelerators
The model system has been running continuously for about 2 months. The first crystallization of the drain pipe is shown in Figure 10. Through experiments, the changes of pH value and total alkalinity under different accelerator dosage are shown in Figure 11, and the change curve of calcium ion concentration is shown in Figure 12.
FIGURE 10
FIGURE 11
FIGURE 12
As can be seen from Figure 11, the effluent pH of all accelerator dosage groups is maintained in the high range of 11.4∼12.6, and the overall trend is decreasing with time. The change rule of total alkalinity is consistent with pH value: it drops rapidly at the initial stage and then enters the slow attenuation stage. The initial alkalinity of 6% dosage group is the highest (12,650 mg CaCO3/L), and that of 0% dosage group is the lowest (5246 mg CaCO3/L). High pH and high alkalinity are due to the erosion of cement hydration products when groundwater permeates. At the initial stage of tunnel construction, cement hydration is incomplete, and seepage channels are also developing. Such alkaline substances as Ca(OH)2 salts are easy to be brought into the drainage system.
As can be seen from Figure 12, Under the condition of different dosage of accelerator, the concentration of calcium ion ()) in effluent shows a decline as a whole, among which the concentration fluctuation range of 0% and 6% dosage groups, it is larger than the 10% and 20% dosage groups. The performance of these fluctuations is to decrease rapidly first, then enter the stage of slow attenuation, and finally tend to be stable. The specific range of each concentration range is: 0% group 68–1,216 mg/L,6% Group 72–592 mg/L,10% group 76–188 mg/L,20% group 76–208 mg/L. During the whole monitoring process, we also noticed that the concentration values measured for the first time in all groups were lower than those measured for the second time. This phenomenon is attributed to the brief contact time and limited wetting area between the groundwater and the concrete during the initial seepage stage, resulting in a smaller total amount of leached and transported calcium. In order to further analyze the correlation between ion release and macroscopic crystallization, we convert the dissolved CaCO3 subunit mass into equivalent CaCO3 mass, and the specific data are recorded in Table 9 in detail.
TABLE 9
| Dosage | 0% | 6% | 10% | 20% |
|---|---|---|---|---|
| Total effluent calcium (g) | 212.336 | 150.134 | 98.629 | 107.368 |
| Pipe crystallization Mass (g) | 290.022 | 40.325 | 22.058 | 30.144 |
Leached calcium mass and cumulative crystallization under different accelerator dosages.
Based on the blank control (0% dosage) and composition analysis, Ca(OH)2 generated from cement hydration is the primary calcium source for the crystals. In the process of groundwater seepage, calcium-containing substances in concrete will be dissolved out, and CaCO3 effluent will be generated by reaction in high alkaline environment. Comparing the experimental results under different conditions, we can see that the order of total dissolved calcium amount and crystallization amount is 0% > 6% > 20% > 10%. 10% accelerator dosage has certain inhibitory effect on calcium loss, but this effect is not linear. Excessive dosage (e.g., 20%) causes uneven dispersion of the accelerator, forming water droplets encapsulated by hydration products that subsequently create internal pores. Furthermore, localized water shortages can easily induce dry shrinkage cracks. These microscopic defects are equivalent to providing some seepage channels for the dissolution of calcium ions, which will promote the occurrence of crystallization reaction to a certain extent. Therefore, simply increasing the dosage of accelerator is not an effective means to prevent and treat calcium loss. In addition, we also observed that the crystallization amount inside the pipeline will exceed the total calcium amount in the effluent without adding accelerator, but after adding accelerator, the dissolution of calcium mainly exists in the effluent in the form of ions, which shows that in the actual tunnel project under the condition of general water quality, the calcium loss of the initial support sprayed concrete, it is mainly manifested in the form of ion migration.
4.2.2 Influence of water reducers
This simulation test lasted for about a month, and data measurement was conducted every 3 days. The recorded contents included pH value of the solution, pipeline quality, , the concentration of CO32−and was used to analyze the effect of water reducer on crystallization kinetics. The changes of pH value, total alkalinity, concentration and crystallization amount with time are shown in Figures 13–16 respectively.
FIGURE 13
FIGURE 14
FIGURE 15
FIGURE 16
As can be seen from Figure 13, the change trend of the pH value of all the water reducing agent dosage groups is generally the same, showing a downward trend, and the numerical value fluctuates in the strong alkaline range of 9.4–10.6, among them, the decline in the first 24 days was the most obvious. The pH decline in the 0% dosage group is relatively gentle, whereas the 6% group exhibits a faster decrease with local rebounds on the 7th and 27th days. Interestingly, the change trajectories of the 10% and 20% dosage groups are highly synchronized, reaching the lowest value of 9.4 on the 24th day, and then slightly rising back, this is related to the situation that hydroxyl ions are consumed in large quantities at the initial stage of the crystallization process.
As shown in Figure 14, the total alkalinity exhibited an overall increasing trend, ranging from 750 to 1,600 mg/L. Notably, all experimental groups experienced a significant jump, peaking around the 20th day. The control group, that is, the 0% dosage group, tends to be stable after the decline in the first 4 days; The 6% and 10% dosage groups are in the slow growth stage in the first 13–19 days, after that, it rose rapidly and reached the maximum value on the 30th day; The 20% dosage group entered the fast rising channel after 16 days of fluctuation growth. The continuous accumulation of alkalinity shows that the high alkaline environment maintained by the system is related to the continuous dissolution of cement products that are not completely hydrated, this also provides a stable precursor source for subsequent crystallization reaction outbreaks.
The change of concentration shown in Figure 15 shows a relatively large fluctuation range and a downward trend as a whole. The value is mainly between 100 and 550 μg/L, compared with the total alkalinity, it is an order of magnitude lower. This difference reflects that calcium ions are more sensitive to the dynamic precipitation process of hydration products, especially in the 10% dosage group, whose fluctuation is the most prominent, the concentration fluctuates greatly between 125 and 550 μg/L, which indicates that 10% of the water reducing agent dosage, it has a significant interference effect on the precipitation kinetics of hydration products and the formation process of precursors, which will create a highly unstable ion environment and contribute to the rapid nucleation of crystals.
Under all experimental conditions, the order of magnitude of concentration is lower than the total alkalinity, which again reflects the high sensitivity of calcium ions to the dynamic precipitation process of cement hydration products. The experimental results further confirm that when the dosage of water reducing agent is 10%, the fluctuation and response characteristics of concentration are the most significant, this shows that this specific dosage has the greatest influence on the precipitation kinetics of hydration products and the formation of crystallization precursors.
Finally, take a look at the change of cumulative crystallization amount in Figure 16. The cumulative crystallization amount of all dosage groups is on the rise, with the total amount about 1 g. After the 24th day, it enters the stage of rapid growth, the obvious fluctuation of crystallization amount during this period reflects a dynamic change process of “deposition-stripping”. The reason for this is that the CaCO3 Crystal initially attached to the pipe wall will be continuously cut and washed by water flow, and the crystal with loose structure will be washed away, the crystal with dense structure will remain and continue to grow. Comparing the data of each group, we can find that the cumulative crystallization amount of the 10% dosage group is the largest, which means that the pipeline blocking risk is the highest under this dosage group, while the cumulative crystallization amount of the 6% dosage group is the smallest, this shows that the water reducing agent mainly causes pipeline blockage by regulating crystallization kinetics, rather than changing the crystal morphology.
After completing the experiment on the influence of different water reducing agent dosage on the crystallization of the pipeline, we collected the crystals generated from the experiment and compared the microscopic morphology with the samples taken from the actual tunnel engineering site, the relevant content is shown in Figure 17 to analyze the influence of water reducing agent on crystal morphology.
FIGURE 17
As can be seen from Figures 17, 18, the crystals generated by laboratory simulation are highly similar to the morphology and characteristics of samples collected on site. Although there are some subtle differences in macro performance due to the different groundwater occurrence environment, however, the crystals produced in the laboratory are mainly calcite, presenting rhombus or cube crystal form. After Microscopic Characterization of crystals with different dosage of water reducing agent, it is found that adding water reducing agent does not obviously change the geometric morphology and surface texture characteristics of crystals, indicating that the water reducer’s influence on the clogging mechanism is not realized by altering crystal growth habits, but by regulating crystallization kinetics (i.e., significantly changing the crystallization rate) thus causing blockage of drainage system.
FIGURE 18
4.2.3 Influence of fly ash
Due to the extremely low calcium ion concentrations in the fly ash experimental groups, which consistently remained below the sensitivity threshold of conventional titration methods, the evaluation of fly ash’s impact was primarily conducted through the analysis of macro-scale indicators, specifically total alkalinity and cumulative crystallization mass, as illustrated in Figures 19, 20. This analytical shift was necessary because the fly ash significantly altered the ionic release profile of the cementitious matrix, necessitating a focus on the long-term accumulation patterns rather than transient ionic fluctuations.
FIGURE 19
FIGURE 20
Regarding the variation of total alkalinity shown in Figure 19, the experimental results show that the total alkalinity of all fly ash dosage groups is on the rise continuously, and the numerical fluctuation range is generally between 1,100 and 2,400 mg/L. Although the alkalinity has been rising in the 30-day experimental period, the overall growth trend is more stable than that of the accelerator and water reducing agent experimental groups, this shows that the addition of fly ash plays a certain buffering role in the chemical environment of seepage water. The appearance of this stabilization effect is related to the gradual participation of fly ash in the secondary hydration reaction. It can slow down the dissolution of highly alkaline hydration products in the early stage, thus reducing the amount of water entering the drainage system. According to the change characteristics mentioned above, although the total alkalinity of the system has been accumulating all the time during the whole experiment, there are subtle differences between the periodic increment of alkalinity and mutation nodes under different conditions, this is mainly related to the difference of fly ash dosage and internal reaction kinetics.
The experimental results of cumulative crystallization amount in Figure 20 show that fly ash has a strong inhibitory effect on the formation of blockage. The initial crystallization accumulation speed of the 10% fly ash dosage group is relatively fast, but when the dosage increases to 20%, the cumulative crystallization amount becomes the lowest among all experimental groups. This linear inhibition relationship shows that fly ash has a good inhibition effect on the crystallization “explosion” stage that is easy to occur in the middle and later stages of drainage system operation.
The core mechanism of this inhibition is mainly related to the volcanic ash effect of fly ash and the densification of microstructure. From a chemical point of view, fly ash will react with calcium hydroxide, the main source of crystallization precursor, to generate additional calcium silicate hydrate (C-S-H) gel, which directly reduces the free calcium content that can be dissolved out; from a physical point of view, this secondary hydration reaction will refine the pore structure of sprayed concrete and effectively block the internal channel of calcium ion migration to the drainage pipe interface. Consequently, macroscopic observations (Figure 21) reveal that the pipe surface in the 20% fly ash group remains the most intact, with the least visible scaling. This demonstrates that fly ash effectively contributes to source control against drainage system failure, providing a valuable reference for material selection in similar engineering projects.
FIGURE 21
4.2.4 Influence of water quality
In order to simulate the influence of strong erosive karst water environment on tunnel drainage system, we added sodium bicarbonate (NaHCO3 dehydrogenase) with a concentration of 250 mg/L into circulating water. Macroscopic observation during the experiment showed that the group added with coagulant showed obvious white crystals in the first 24 h, and the growth rate was much faster than that of the common water quality group. This situation shows that the ion composition of groundwater is a key external factor, which can greatly shorten the induction period of crystallization process.
The core reason for this accelerated blockage is the direct chemical precipitation mechanism brought by high concentration of bicarbonate ions. In type water environment, the reaction is carried out according to the process of . Different from the crystallization process which is often limited by the atmospheric diffusion and slow dissolution rate of CO2 dioxide in common water quality, this direct reaction skips the obstruction of gas-liquid mass transfer. anion ions provided by underground water will react rapidly with Ca(OH)2 anion dissolved in concrete and anion ions in high alkaline environment to generate a large amount of calcite precipitation. In order to further study the physical and chemical evolution characteristics of this process, we dynamically monitored the pH value, ion concentration () and total alkalinity of the effluent, the relevant time variation rules are shown in detail in Figures 22, 23.
FIGURE 22
FIGURE 23
As can be seen from Figure 22, the pH of the solution in NaHCO3 aqueous group decreases faster than that in the ordinary water quality group, and the final stable value is also lower. This is because the erosive HCO3− anion ions will neutralize the anion ions rapidly, and the continuous movement of chemical balance will also promote the further dissolution of the hydration products of cement matrix. The time series monitoring results of Figure 23 further confirm that the concentration of effluent is far lower than the level under the condition of common water quality. This is not because the amount of dissolution decreases, but because a large number of glycosides will convert the dissolved calcium into solid CaCO3 glycocrystals almost immediately before discharge. Therefore, the total alkalinity in Figure 23 presents a specific rule of consumption-regeneration, reflecting that dissolved ions can be efficiently converted into solid phase deposition.
Table 10 summarizes the quantitative comparison data of substance balance under the condition of NaHCO3 water quality. Although the total quality of dissolved calcium detected by effluent is far lower than that of ordinary water quality group, the total quality of crystals retained in the pipeline is several times that of the latter. This seemingly contradictory result confirms an efficient conversion mechanism: in erosive karst water, most dissolved calcium will not be left in the solution in the form of free ions, but because metastable ions are sufficient, fast conversion to solid CaCO3 crystalline crystals. Therefore, the risk of sudden serious drainage failure is much higher for tunnels located in the region rich in NaHCO3 limestone, which requires more active source control and anti-crystallization design.
TABLE 10
| Dosage | 0% | 6% | 10% | 20% |
|---|---|---|---|---|
| Total effluent calcium (g) | 72.387 | 95.807 | 21.887 | 26.910 |
| Pipe crystallization Mass (g) | 560.125 | 424.983 | 86.952 | 102.412 |
The mass of calcium released by different dosages of setting agents.
Experimental data (Figure 24) show that the dosage of coagulant has a significant effect on the dissolution and deposition of calcium. Specifically, the total quality of calcium in effluent is 6% > 0%>20% > 10%, the order of crystallization quality in the pipeline is 0% > 6%>20% > 10%. Among all the experimental groups, the crystallization quality of the 10% dosage group is the lowest, which indicates that the appropriate dosage of coagulant can effectively inhibit the crystallization development. However, the crystallization quality of the 20% dosage group rose slightly, which also indicated that simply increasing the concentration of coagulant was not the best way to prevent crystallization. In addition, the quality of crystallization in the pipeline far exceeds the total quality of calcium detected in the effluent, which also shows that in the environment rich in bicarbonate, most of the calcium dissolved from the initial concrete will be intercepted in the pipeline in the form of solid crystals.
FIGURE 24
Compared with common water quality conditions, the crystallization phenomenon in NaHCO3 aqueous environment is much more serious, although the total quality of calcium detected in effluent is lower. The difference of the core mechanism between the two lies in the conversion efficiency: in the ordinary water quality, the Ca(OH)2 dehydrogenase dissolved by cement hydration can only be nucleated by atmospheric CO2 anion partial pressure or specific physical and chemical conditions, this limited conversion efficiency will cause a large amount of calcium substance to be washed away by water flow in the form of ions. However, in the environment rich in bicarbonate, phosphate ion can rapidly undergo chemical precipitation reaction with ca2+ phosphate under strong alkaline conditions and directly convert it into solid CaCO3 phosphate particles. This efficient conversion mechanism allows most dissolved calcium to deposit locally, thus greatly improving the blocking risk of the drainage system.
The influence degree of concrete composition and water quality on crystallization performance of tunnel drainage system is summarized in Table 11.
TABLE 11
| Component | Tested range | Recommended dosage/Strategy | Engineering significance |
|---|---|---|---|
| Accelerator | 0%–20% | 10% | 10% is the optimal balance point for suppressing early-stage calcium leaching |
| Water reducer | 5% | Mix designs should strictly avoid the high-risk dosage range to prevent sudden clogging | |
| Fly ash | 20% | Dosages > effectively densify the microstructure and reduce calcium migration | |
| Water quality | Karst environment | — | Areas with high mineralization require active drainage design and material source blocking |
The influence of concrete components and moisture on the crystallization performance of tunnel drainage system.
It is worth noting that in a complex shotcrete system, the interactions between components are inevitable. Specifically, the accelerator, water reducer, and fly ash exhibit intricate synergistic and antagonistic effects. For instance, a strong antagonistic effect exists between accelerators and water reducers: accelerators rapidly promote hydration (e.g., forming ettringite) and induce high alkalinity to achieve quick setting, which directly competes with the dispersion mechanism and adsorption efficiency of polycarboxylate water reducers designed to delay hydration and improve workability. If not perfectly balanced, this contradiction can lead to a porous and poorly compacted cement matrix, thereby exacerbating the leaching of calcium ions rather than preventing it. Furthermore, complex interactive effects occur between fly ash and accelerators. Fly ash relies on secondary pozzolanic reactions with Ca(OH)2 to densify the pore structure. However, the rapid hydration induced by the accelerator alters the generation rate and spatial distribution of early-stage Ca(OH)2. This rapid setting might encapsulate fly ash particles prematurely, hindering their long-term pozzolanic efficiency. Conversely, synergistic effects are prominent between fly ash and water reducers; the morphological effect of fly ash (spherical particles) works in tandem with the water reducer to improve the compactness of the shotcrete, physically blocking calcium migration channels and mitigating potential micro-defects (; ). Although this study primarily focuses on single-factor impacts, acknowledging and evaluating these potential synergistic and antagonistic effects is crucial for the transition from independent optimization to a scientifically rigorous composite design.
5 Conclusion
By developing a dynamic simulation system for tunnel seepage and crystallization, this study systematically investigated the effects of complex admixture systems and karst water environments on the crystallization evolution within drainage networks. The primary conclusions are summarized as follows:
Fundamental Mechanism of Crystallization Clogging: From a fundamental perspective, crystallization clogging in tunnel drainage systems is a typical multi-field coupling process involving “internal calcium leaching, external environmental intervention, and dynamic flow field deposition.” Its essence lies in the fracture and leaching of calcium chains from the hydration products of primary support shotcrete under a highly alkaline environment. Driven by the chemical potential gradient of the highly mineralized karst water environment, these ions migrate to the drainage network interface and undergo phase transformation. The microscopic defects within the material and the strong aggressiveness of the external water quality synergistically govern the evolution rate of this crystallization chain. Specifically, the clogging material is identified as high-purity calcite (>98%), and two distinct developmental modes were revealed: “in situ attachment growth” and “migration deposition.”
Influence of Material and Environmental Factors: Under single-factor controlled conditions, the intervention mechanisms of various factors differ significantly. A 10% accelerator dosage and a 5% water reducer dosage are identified as the optimal independent thresholds to suppress early-stage calcium leaching, whereas increasing the water reducer to 10% triggers a severe crystallization outbreak. In contrast, fly ash demonstrates a robust linear inhibitory characteristic by densifying the microstructure via the pozzolanic effect, with a 20% dosage acting as its optimal threshold. Furthermore, NaHCO3-type highly mineralized water acts as an external trigger that bypasses atmospheric CO2 diffusion limits, exponentially magnifying the deposition scale.
Engineering Recommendations for “Active Prevention”: To transition from passive dredging to active source control, we recommend a “dual-control” engineering strategy. First, strict calibration of the liquid accelerator dosage at the spray nozzle is mandatory; capping the dosage at 10% prevents the formation of drying shrinkage micro-cracks that serve as preferential pathways for leaching. Second, this control should be synergistically coupled with a 20% fly ash replacement to densify the existing pore structure. This approach effectively minimizes the formation of new leaching channels while simultaneously severing the calcium supply chain, maximizing the service life of the drainage system in active karst areas.
5.1 Outlook
Due to the long experimental cycle (1–2 months per test), this study primarily adopted a single-factor scheme. Future work will involve systematic orthogonal designs to evaluate the coupled interference and interactive effects among admixtures.
Statements
Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.
Author contributions
BS: Conceptualization, Methodology, Writing – original draft. CD: Formal Analysis, Writing – original draft. XL: Methodology, Software, Writing – original draft. WN: Software, Validation, Writing – original draft. HC: Validation, Writing – original draft. JL: Validation, Data curation, Writing – review and editing. BC: Writing – review and editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The financial support from National Natural Science Foundation of China (grant number: 52508490), Guangdong Basic and Applied Basic Research Foundation (grant number: 2024A1515110192).
Acknowledgments
Thanks to all those who contributed to the articles.
Conflict of interest
Authors BS and CD were employed by Poly Changda Engineering Co., Ltd., Guangzho. Fourth Branch Company. Authors XL and WN were employed by Guangzhou Xiaoning Road Engineering Technology Research Office Co., Ltd. Author HC was employed by Guangdong Highway Construction Co., Ltd. Author JL was employed by Jiangmen Yinzhouhu Expressway Co., Ltd.
The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AlijaS.TorrijoF. J.Quinta-FerreiraM. (2013). Geological engineering problems associated with tunnel construction in karst rock masses: the case of Gavarres tunnel (Spain). Eng. Geol.157, 103–111. 10.1016/j.enggeo.2013.02.010
2
AnS.JiangC.LiS.ShiC. (2025). Experimental study on the effect of ion transmission mode on the calcium leaching rate of cementitious materials. Case Stud. Constr. Mater.22, e04215. 10.1016/j.cscm.2025.e04215
3
BaeJ. H.JangD.KimN.ParkJ.ChoiS.KimS.et al (2025). Microstructural changes in carbonation-cured slag-blended calcium sulfoaluminate cements under electrically accelerated leaching. Dev. Built Environ.22, 100664. 10.1016/j.dibe.2025.100664
4
BernerR. A.MorseJ. W. (1974). Dissolution kinetics of calcium carbonate in sea water; IV, Theory of calcite dissolution. Am. J. Sci.274 (2), 108–134. 10.2475/ajs.274.2.108
5
ChenY.CuiY.Guimond-BarrettA.ChilleF.LassalleS. (2019). Investigation of calcite precipitation in the drainage system of railway tunnels. Tunn. Undergr. Space Technol.84, 45–55. 10.1016/j.tust.2018.10.021
6
ChenJ.LyuG.JinW.XiaJ. (2025). Influence of microenvironment evolution induced by calcium leaching in concrete under pressurized water conditions on chloride transport. Struct. Concr.26 (3), 2505–2525. 10.1002/suco.202401385
7
DiH.GuanX.LiuS. (2023). Changes in mechanical performance and pore structure of hardened cement paste under the combined effect of calcium leaching and dry/wet cycles. Constr. Build. Mater.407, 133519. 10.1016/j.conbuildmat.2023.133519
8
EichingerS.BochR.LeisA.KoraimannG.GrenggC.DombergerG.et al (2020). Scale deposits in tunnel drainage systems: a study on fabrics and formation mechanisms. Sci. Total Environ.718, 137140. 10.1016/j.scitotenv.2020.137140
9
FanH.ZhuZ.SongY.ZhangS.ZhuY.GaoX.et al (2021). Water pressure evolution and structural failure characteristics of tunnel lining under hydrodynamic pressure. Eng. Fail. Anal.130, 105747. 10.1016/j.engfailanal.2021.105747
10
GalanI.BaldermannA.KusterleW.DietzelM.MittermayrF. (2019). Durability of shotcrete for underground support–review and update. Constr. Build. Mater.202, 465–493. 10.1016/j.conbuildmat.2018.12.151
11
GarbaM. J.ZhaoX.YuanQ.YuC.ShalabyY.ChenL.et al (2024). Exploring the impacts of liquid accelerators on sulfate resistance and calcium leaching behavior of wet-mixed shotcrete. Constr. Build. Mater.453, 139069. 10.1016/j.conbuildmat.2024.139069
12
HeB. G.ZhangY.ZhangZ. Q.FengX. T.SunZ. J. (2021). Model test on the behavior of tunnel linings under earth pressure conditions and external water pressure. Transp. Geotech.26, 100457. 10.1016/j.trgeo.2020.100457
13
HouS.LiF.TangH.WenT.ChenZ.GaoH. (2022). Investigations on the performance of shotcrete using artificial lightweight shale ceramsite as coarse aggregate. Materials15 (10), 3528. 10.3390/ma15103528
14
HuangZ. (2021). Research on the crystallization law of karst water in tunnel drainage pipeline based on DBL theory. Chin. J. Undergr. Space Eng.17 (5), 1546–1553.
15
JiangY.ZhengY.YuL.LiaoL.SuY. (2025). Mechanism and effectiveness of glass powder effect on calcium leaching resistance of primary support concrete in tunnels. Constr. Build. Mater.474, 141190. 10.1016/j.conbuildmat.2025.141190
16
LarsenT. (1980). Calcium carbonate crystal growth and scale mechanisms based on theory of crystal growth dynamics. J. Cryst. Growth49 (2), 394–398.
17
LiuD. G.YangY.MaoC. J.WuJ. F.WuJ. C. (2022). A comparative Study on hydrodynamics and hydrochemistry coupled simulations of drainage pipe crystallization blockage in Karst tunnels. J. Earth Sci.33, 1179–1189. 10.1007/s12583-022-1720-3
18
LiuZ.ZhangX.LiuS. (2025a). Adaptive drainage pipe for crystallization prevention: mechanism and experimental Study. Tunn. Undergr. Space Technol.143, 105499.
19
LiuS.ZhangX.ZhangB. (2025b). Crystallization blockage in highway tunnel drainage system based on molecular dynamics. AIP Adv.15 (3), 035201. 10.1063/5.0256234
20
PlummerL. N.WigleyT. M. L.ParkhurstD. L. (1978). The kinetics of calcite dissolution in CO2-water systems at 5 degrees to 60 degrees C and 0.0 to 1.0 atm CO2. Am. J. Sci.278 (2), 179–216. 10.2475/ajs.278.2.179
21
ReddyM. M. (1977). Crystallization of calcium carbonate in the presence of trace concentrations of phosphorous containing anions. I. Inhibition by phosphate and glycerophosphate ions at pH 8.8 and 25°C. J. Cryst. Growth41 (2), 287–295. 10.1016/0022-0248(77)90053-1
22
ReinprechtV.PettauerM. (2025). Clogging of landslide drainages: a hydrochemical approach for early detection. Sci. Total Environ.906, 167448.
23
SakoparnigM.GalanI.SteindlF. R.KusterleW.JuhartJ.GrenggC.et al (2021). Durability of clinker reduced shotcrete: Ca2+ leaching, sintering, carbonation and chloride penetration. Mater. Struct.54, 78. 10.1617/s11527-021-01644-7
24
ShiH.SongL.ZhangH.ChenW.LinH.LiD.et al (2022). Experimental and numerical studies on progressive debonding of grouted rock bolts. Int. J. Min. Sci. Technol.32 (1), 63–74. 10.1016/j.ijmst.2021.10.002
25
ShiH.ChenW.ZhangH.SongL.LiM.WangM.et al (2023). Dynamic strength characteristics of fractured rock mass. Eng. Fract. Mech.292, 109678. 10.1016/j.engfracmech.2023.109678
26
ShiH.ChenW.WuJ.RongC.WangZ.SongL.et al (2025). Unified rock strength theory incorporating holistic consideration of macro-meso-micro defect coupling effects. Eng. Fract. Mech.328, 111534. 10.1016/j.engfracmech.2025.111534
27
ShinJ. H.LeeI. K.JooE. J. (2014). Behavior of double lining due to long-term hydraulic deterioration of drainage system. Struct. Eng. Mech.52 (6), 1257–1271. 10.12989/SEM.2014.52.6.1257
28
SteindlF. R.MittermayrF.SakoparnigM.JuhartJ.BriendlL.LindlarB.et al (2023). On the porosity of low-clinker shotcrete and accelerated pastes. Constr. Build. Mater.368, 130461. 10.1016/j.conbuildmat.2023.130461
29
TongY.YeF.TianC.ZhangJ.JiangY.HanX.et al (2024). Optimizing the development and field application of calcium stabilizing agents for preventing calcium leaching in tunnels. J. Clean. Prod.449, 141757. 10.1016/j.jclepro.2024.141757
30
WuJ.YangS.WilliamsonM.WongH. S.BhudiaT.PuH.et al (2025a). Microscopic mechanism of cellulose nanofibers modified cemented gangue backfill materials. Adv. Compos Hybrid. Mater8, 177. 10.1007/s42114-025-01270-9
31
WuJ.ZhangW.WangY.JuF.PuH.RiabokonE.et al (2025b). Effect of composite alkali activator proportion on macroscopic and microscopic properties of gangue cemented rockfill: experiments and molecular dynamic modelling. Int. J. Min. Metall. Mater32, 1813–1825. 10.1007/s12613-025-3140-8
32
XuJ.LiJ. (2004). Influence of CO2 release on process of precipitation of calcium carbonate. J. Tongji Univ. Nat. Sci.32 (4), 516–520.
33
XuZ.LinP.XingH.PanD.HuanX. (2021). Hydro-mechanical coupling response behaviors in tunnel subjected to a water-filled karst Cave. Rock Mech. Rock Eng.54 (8), 3737–3756. 10.1007/s00603-021-02423-0
34
XuC.ChenY.YangY.LiP.WangS.LiL. (2022). Experimental investigation of crystal blocking in drainage pipes for tunnels in the Karst Region. Appl. Sci.12 (21), 10928. 10.3390/app122110928
35
XuQ.ZhangS.LiP.LiuC.BaoT. (2023). Lining failure performance of highway tunnels induced by the drainage system deterioration. Eng. Fail. Anal.149, 107236. 10.1016/j.engfailanal.2023.107236
36
ZengX.WuQ.LiangX.LanL.LiangJ.WuJ. (2026). Effect of fly ash content on the leaching resistance of shotcrete. Front. Mater.13, 1753634. 10.3389/fmats.2026.1753634
37
ZhangH. Q.ZhaoZ. Y.TangC. A.SongL. (2006). Numerical study of shear behavior ofintermittent rock joints with different geometrical parameters. Int. J. Rock Mech. Min. Sci.43 (5), 802–816. 10.1016/j.ijrmms.2005.12.006
38
ZhangH. Q.HeY. N.TangC. A.AhmadB.HanL. J. (2009). Application of an improved flow-stress-damage model to the criticality assessment of water inrush in a mine: a case study. Rock Mech. Rock Eng.42, 911–930. 10.1007/s00603-008-0004-2
39
ZhangZ.NiuM.LiuZ.LiG. (2024). The mechanism of accelerator types on calcium leaching in shotcrete. Constr. Build. Mater.457, 139353. 10.1016/j.conbuildmat.2024.139353
40
ZhangH.ShiH.ZhaoH.SongL. (2025). Characterization of transverse vibration response of resin-anchored bolt under axial tension and life cycle health assessment of anchorage debonding. Measurement256, 118273. 10.1016/j.measurement.2025.118273
Summary
Keywords
calcium carbonate crystallization, calcium leaching, clogging mechanism, shotcrete, tunnel drainage system
Citation
Shu B, Du C, Li X, Nie W, Cui H, Li J and Chen B (2026) Experimental study on the mechanism and active prevention and control of crystalline clogging in tunnel drainage systems in limestone areas. Front. Mater. 13:1858512. doi: 10.3389/fmats.2026.1858512
Received
17 April 2026
Revised
25 May 2026
Accepted
29 May 2026
Published
07 July 2026
Volume
13 - 2026
Edited by
Hao Shi, Anhui University of Science and Technology, China
Reviewed by
Liang Jinxi, China University of Mining and Technology, China
Longlong Chen, Chang’an University, China
Updates
Copyright
© 2026 Shu, Du, Li, Nie, Cui, Li and Chen.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Bo Chen, 984166060@qq.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.