Introduction

The application of concrete dates back to ancient times, embodying a profound historical legacy. Serving as the primary load-bearing material in extant structures, the exploration of concrete durability has persistently occupied a central position in academic research. Presently, the bulk of scholarly investigations predominantly center on the durability of terrestrial concrete structures. Conversely, the durability of subterranean concrete structures remains a relatively under-explored domain, with a paucity of in-depth studies dedicated to this critical aspect of civil engineering.

Owing to the intricate environment and diverse functions of underground concrete structures, stringent requirements are imposed on their bearing capacity and durability, rendering life assessment a challenging task. Ensuring their long-term usability, stability, and safety is of paramount importance1,2,3,4,5. Underground concrete structures enhance their mechanical performance by altering material properties, thereby achieving remarkable long-term effects and serviceability6,7. Quang Dieu Nguyen8 explored the exposure of various underground reinforced concrete pipelines to chloride and carbonation environments. The pipelines, with ages spanning from 43 to 70 years, were studied, and it was determined that, based on the minimum concrete protective layer of the steel bars, the service life of these pipelines under the exposure conditions could reach 100 years.

With the continuous transformation of urbanization and population growth, rail transit and underground space will continue to expand, inevitably affecting the operation of underground concrete structures9,10,11,12. Monitoring and evaluation of underground concrete structures in operation is essential to ensure their long-term availability, stability, and safety as a transportation network to monitor their long-term deterioration over time13,14,15,16.

The long-term performance of underground concrete structures is predominantly governed by the intricate interplay between the structure itself and its surrounding environment. This interaction is instigated by a multitude of factors, encompassing both internal and external, as well as natural and human-induced elements. Owing to the inherent long-term dynamic characteristics of the lining structure and the adjacent rock mass, the equilibrium conditions of underground concrete structures undergo continuous temporal changes17,18,19,20,21,22,23.

Due to the complexity of the underground environment, underground reinforced concrete structures will be affected by complex environments such as groundwater and soil layer loads, thereby affecting the durability of reinforced concrete structures24,25,26,27,28,29,30. Quang Dieu Nguyen8 studied the concrete cores from underground reinforced concrete pipes aged 43 to 70 years exposed to chloride salt and carbonation environments, and discussed the durability and service life of the pipes. Pengfei Li31 proposed using the durability concrete model-complex three-dimensional (DuCOM-COM3D) analysis software to conduct numerical simulations of underground culvert projects, investigated the complex mechanical and environmental loads affecting RC structures.

After extended operation, the microscopic characteristics and chemical composition of concrete within underground structures will undergo alterations, thereby influencing the macroscopic mechanical properties of the concrete. Chen Li32 investigated the chemical and mineralogical transformations in the concrete of underground tunnel structures constructed in China’s eastern coastal regions between 1980 and 1996. It was found that the durability of underground concrete structures in Chinese cities has changed following prolonged operation. China’s history of constructing underground structures is relatively short. The long-operating subway structure examined in this study is the earliest subway station in China, constructed in 1969. Having been in operation for 50 years, it has endured nearly half a century. The objective of this research is to explore the mechanical properties and microscopic changes of subway concrete stations after five decades of operation.

Currently, research on the concrete performance of subway station structures in long-term operation is scarce. This study endeavors to compare the material mechanical properties and microstructural characteristics of such long-operating subway structures. It does so by referencing the concrete design specifications for subway structure construction in similar years and the material characteristics of 90-day-old C40 concrete.

Background of concrete specimens

The buried depth of the operating subway station is 12 m, and the terrain is flat and stable. The soil layer is mainly divided into clay, sand, and pebbles. There is a groundwater layer within the scope of the subway station, and the chemical composition of the groundwater is not corrosive to the concrete structure. After the apparent state inspection of the station structure, after 50 years of operation of the subway station, there are a few micro-cracks on the surface of the structural concrete, and there are no large cracks in the whole structure. The asphalt waterproof layer is completely protected, and there is no leakage of water in the main structure. Because the subway station structure is less susceptible to environmental factors, this study predominantly focuses on the material properties of the concrete constituting the station structure.

Regarding the design of the underground concrete structure during the construction period, the standard axial compressive strength of the concrete is designed to be 35 MPa, the tensile strength is designed to be 3.0 MPa, and the elastic modulus is designed to be 35 GPa. The cementitious material is mainly ordinary Portland cement. Ribbed steel bars are used as the reinforcement, and the minimum tensile strength is designed to be 500 MPa.

In this study, the concrete of the side wall of the subway station was selected (Fig. 1). In this sampling, the method of combining a water drill and rope saw is used to reduce the disturbance to the sample properties and control the influence of sampling on the bearing capacity of the structure. The sampling process is shown in Fig. 2. After the large volume concrete blocks are processed into cylinders and concrete slabs, the specimens are processed in the specimen processing plant to reduce the test error.

Fig. 1
Fig. 1
Full size image

The surrounding and sampling location of the Beijing Metro Line 2 subway station.

Fig. 2
Fig. 2
Full size image

Sampling flow chart.

Test size of concrete specimens

In this study, by the Standard for Test Methods of Physical and Mechanical Properties of Concrete (GB/T 50081-2019), the macroscopic mechanical properties of concrete, such as compressive strength, tensile strength, elastic modulus, Poisson’s ratio, and the bond strength between concrete and steel bars, were tested. The test specifications strictly adhere to the requirements outlined in the aforementioned standards. For clarity, C100 denotes a cylindrical specimen with dimensions of 100 mm in diameter and 200 mm in height; C150 represents another cylindrical specimen, measuring 150 mm in diameter and 300 mm in height; and P150 refers to a prismatic specimen with dimensions of 150 mm × 150 mm × 300 mm. The bond strength specimens are cubic, with each side measuring 150 mm, and a 500-mm-long steel bar is centrally embedded through the cube. The sampling position and specifications of the microscopic test are shown in Fig. 3.

Fig. 3
Fig. 3
Full size image

Micro test sampling position.

Compressive strength test results of concrete specimens

Fig. 4
Fig. 4
Full size image

Compressive strength test results of C100, C150, and P150 samples.

It can be seen from Fig. 4 that according to the compressive strength test of three groups (three test specimens in each group) with different sizes, the axial compressive strength of C100 is 42.7 MPa, and the standard deviation is 3.6. The axial compressive strength of C150 is 38.8 MPa, and the standard deviation is 3.3. The axial compressive strength of P150 is 36.9 MPa, and the standard deviation is 1.9.

The test results show that the concrete used in the long-term operation station structure conforms to the mechanical characteristics of the compressive strength of ordinary concrete. The discreteness of the compressive strength of C100 and C150 is greater than that of P150.

According to the specification, the ratio of compressive strength of standard specimens of C100 and C150 is 0.95, and the ratio of compressive strength of standard specimens of C150 and P150 is 1. The ratio of the compressive strength of the test specimen C100 to C150 is 0.91, and the ratio of the compressive strength of the standard specimen C150 to P150 is 1.05.

Splitting tensile strength test results of concrete specimens

Fig. 5
Fig. 5
Full size image

Splitting tensile strength test results of C100 and C150 samples.

It can be seen from Fig. 5 that the splitting tensile strength of C100 is 3.2 MPa, and the standard deviation is 0.9 according to the splitting tensile strength test of two groups (three test specimens in each group) with different sizes. The splitting tensile strength of C150 is 2.8 MPa, and the standard deviation is 0.5.

The tensile strength of C100 is greater than that of C150, which is in line with the mechanical characteristics of the tensile strength of ordinary concrete, and the tensile strength is slightly smaller than the initial tensile strength of concrete. The tensile strength discreteness of C100 is greater than that of C150.

The test results show that the tensile strength of concrete used in the long-term operation station structure is greater than the initial tensile strength of concrete. The tensile strength of C100 is greater than that of C150, and the tensile strength dispersion of C100 is greater than that of C150. The tensile strength of concrete used in the long-term operation station structure is slightly smaller than the initial tensile strength of concrete.

Static compression elastic modulus test results of concrete specimens

Fig. 6
Fig. 6
Full size image

Static compression elastic modulus test results of C150 samples (EM-1, 2, 3).

It can be seen from Fig. 6 that, according to the measurement results of the three groups (six test specimens in each group) of samples, the static compressive elastic modulus obtained by the C150 cylinder sample is between 30.1 GPa and 31.7 GPa.

The elastic modulus is a measure of an object’s ability to resist elastic deformation, and the factors that affect the bonding strength of a material can influence the elastic modulus of that material.

The static elastic modulus of concrete reflects the strength of the bond between the cement paste and the aggregates in the concrete33,34,35. When compared with the initial elastic modulus of concrete, the elastic modulus of concrete in the structure of a long-term operational station is slightly lower than the initial elastic modulus of the concrete.

Poisson’s Ratio Test Results of Concrete Specimens.

Fig. 7
Fig. 7
Full size image

Poisson’s ratio test results of P150 samples (PR-1, 2, 3).

It can be seen from Fig. 7 that, according to the measurement results of the three groups of samples, the Poisson’s ratio value of the long-term operation station structure concrete measured by the P150 prism sample is about 0.2.

Poisson’s ratio, an elastic constant, characterizes the transverse deformation of materials and is expressed as a dimensionless quantity obtained through experimental measurements. Experimental evidence confirms that the elastic deformation properties of concrete utilized in long-term operational station structures align closely with those of ordinary concrete36,37,38,39.

Bond strength test results of concrete specimens

Fig. 8
Fig. 8
Full size image

The test results of bond strength samples (BS-1,2,3,4,5,6).

According to the test phenomenon and Fig. 8, it can be seen that the bonding strength between the concrete and steel bar of the long-term operation station structure is about 6.69 MPa.

According to the load-displacement curve of the bond strength, it can be seen that the failure mode of the pull-out test is splitting failure. Different from the pull-out failure of steel bars, the splitting failure is more sudden, the early deformation is slow, and the later deformation is faster40.

The steel bar used in the bonding test is a ribbed steel bar. Compared with a round steel bar, it has a larger bonding surface area and exhibits mechanical interlocking41. Owing to the complete hydration reaction of cement in the concrete, the strength of the concrete increases. As a result, the bonding force of the concrete becomes stronger than that in the early construction stage, and the interfacial bonding state between the steel bar and the concrete is more favorable42,43,44.

Comparison between material test results and design values of concrete

Material test

Design values

Test results

Compressive strength

35 MPa

38.8 MPa

Tensile strength

3.0 MPa

2.8 MPa

Elastic modulus

35 GPa

30.1–31.7 GPa

Poisson ‘s ratio

0.2

0.2

Test results of optical observation of minerals

The optical microscope was used to observe the concrete samples of the long-running station structure, and the microscopic photos were taken under the orthogonal microscope.

Through optical microscope imaging observation (Fig. 9), the composition and state of aggregate in concrete are as follows.

Fig. 9
Fig. 9
Full size image

(a) Optical microscope imaging (with rebar holes); (b) Optical microscope imaging (in cement).

The aggregate particle size is generally between 0.5 cm and 3 cm, and the particle size distribution is relatively uniform, mainly composed of broken gravel and pebbles. The aggregate composition is mainly two kinds, one is rock debris, and the other is single mineral debris. The mineral debris is mainly quartz and feldspar, which account for more than 95% of the total mineral debris, and a small amount of mica and hornblende. Between the aggregates are cement hydration products.

According to the observation, both rock debris and mineral debris are relatively fresh, and there is no obvious weathering phenomenon. In particular, mica and hornblende with weak weathering resistance have not yet been seen in traces of weathering.

Test results of SEM

Through the scanning electron microscope (SEM), the cement microstructure of the long-term operation station structure concrete and the 90-day-old C40 concrete were observed and compared to study the difference between the microstructure of the super-long-age concrete and the short-age concrete. C40 concrete is a type of cast-in-place concrete which is prepared by blending Portland cement, aggregate, fly ash, admixtures, and other raw materials in a specific ratio.

Fig. 10
Fig. 10
Full size image

(a) Microstructure of hydrated cement paste in the existing underground station structure; (b) Microstructure of cement in the 90-day-old C40 concrete.

According to the observation results (Fig. 10), it can be seen that the cement hydration products in the concrete of the existing underground station structure are dense, and the degree of hydration is almost complete. There are no obvious micro-cracks, and there are some holes formed in the hydration process. Due to incomplete hydration, the microstructure of cement in 90-day C40 concrete contains more microcracks than that of existing structural concrete.

The cement hydration products of the existing underground station structure concrete and the microstructure of the aggregate interface are compared as follows:

Fig. 11
Fig. 11
Full size image

Microstructure of the interface between hydrated cement paste and aggregate of the existing underground station structure concrete.

It can be seen from the above observation results that there are microcracks along the aggregate surface at the interface between the cement and the aggregate of the existing underground station structure concrete (Fig. 11). There exists an uncompacted area in the interfacial transition zone between the aggregate and the cement paste. In the early stage of the station construction, the concrete was mixed on-site.

During the mixing process, the non-uniform dispersion of cement particles and inadequate wetting of aggregates render the interface highly susceptible to segregation, as documented in references45,46. As cement hydration progresses over an extended period, the bonding capacity of the cement paste is enhanced. Nevertheless, the presence of the aggregate interface inevitably diminishes the interfacial bond strength, facilitating the initiation and propagation of microcracks47. These microcracks compromise the bond strength within the interfacial transition zone (ITZ) between aggregates and cement paste, thereby exerting a significant influence on the overall mechanical performance of concrete, as evidenced by studies46,48,49,50.

Test results of EDS

The EDS test used the geological glass processed by the existing underground station structure concrete sample. The ion distribution of hydrated cement paste in the concrete of the existing underground station structure was observed by EDS to study the element composition of the ultra-long-age concrete.

Fig. 12
Fig. 12
Full size image

(a) The concrete slice point 1 position; (b) Element composition of concrete slice point 1 position.

Fig. 13
Fig. 13
Full size image

(a) The concrete slice point 2 position; (b) Element composition of concrete slice point 2 position.

Fig. 14
Fig. 14
Full size image

(a) The concrete slice point 3 position; (b) Element composition of concrete slice point 3 position.

Based on the Energy Dispersive Spectroscopy (EDS) test results, the elemental components of the hydrated phases in the existing concrete predominantly consist of calcium, aluminum, silicon, oxygen, and iron. These components exhibit a strong resemblance to those of the hydrated phases in Portland cement, as illustrated in Figs. 12, 13 and 14. Since the cement hydration reaction is complete, according to the element ratio of cement hydrate, the hydration products are determined to be hydrated calcium silicate, hydrated calcium aluminate, and hydrated iron calcium aluminate.

Microscopic test results of XRD

The XRD test uses the samples processed by the existing underground station structure concrete. The mineral composition of hydrated cement paste in the existing underground station structure concrete was detected by XRD test.

Fig. 15
Fig. 15
Full size image

Composition of compounds in concrete sample 1.

Fig. 16
Fig. 16
Full size image

Composition of compounds in concrete sample 2.

Researchers studied the crystal phases and mineral contents of concrete samples under the conditions of 50 years of service using XRD (X-ray diffraction) analysis, with a special emphasis on the phases of calcite and silica51,52. Figures 15 and 16 display the X-ray diffraction (XRD) curves of Sample 1 and Sample 2. As the composition of the concrete is mainly aggregate, most of the diffraction peaks of the crystalline phases (both anhydrous and hydrated compounds) could not be identified. The diffraction peaks of quartz and the hydration product Ca(OH)2 in the sample composition are identifiable. No diffraction peaks associated with calcite were observed in the X-ray diffraction pattern. This observation may imply that under the conditions of 50 years, only a small quantity of Ca(OH)2 crystals were detected by XRD, indicating that the internal environment of the sample remains alkaline and no carbonation reaction has taken place53,54,55. The concrete inside the long-term operating station has not experienced carbonation56. Minerals such as C2S, C2S, C2A, and C4AF were not detected in the analysis, suggesting that the Portland cement has mostly hydrated after 50 years. This could be attributed to the relatively stable environment of the long-term subway station structure and the pozzolanic reactions47,57,58,59.

Conclusions

After 50 years of operation, an existing underground station was studied to evaluate its current concrete mechanical properties and microstructural characteristics, leading to the following conclusions:

  1. 1.

    The aggregate morphology of the concrete in the existing underground station structure remains fresh after 50 years of operation, with no significant signs of weathering. The reinforcement has not corroded, and the internal concrete structure remains undamaged.

  2. 2.

    The cement in the underground station concrete is well-hydrated with no carbonation. Yet, microcracks around aggregates, along with this enhanced cohesion, jointly impact concrete’s mechanical properties.

  3. 3.

    After 50 years of operation, the mechanical properties of the existing underground station concrete have changed. These changes are primarily attributed to the complete hydration of the cement and the presence of microcracks at the cement-aggregate interface.

  4. 4.

    The observed dispersion in specimen strength can be predominantly attributed to the stochastic distribution of microcrack locations, lengths, and densities within the concrete matrix.

  5. 5.

    This study provides material strength for numerical model research in subway engineering and lays a foundation for the subsequent study of the durability and toughness of existing subways.