Experimental Study on the Bearing Capacity of Reinforced Concrete Pipes with Corrosion-Thinning Defects Repaired by UHP-ECC Mortar Spraying

Abstract

The in situ spraying method is widely used because of its advantages as a trenchless pipeline repair technology, including a fast construction speed and close bonding between the repair lining layer and the reinforced concrete pipe. However, current research on high-performance spray repair materials, the bearing capacity of pipelines before and after repair, and the failure modes between the two interfaces after repair is insufficient. Through laboratory tests designed with multiple sets of control tests, this paper outlines the bearing capacity of reinforced concrete pipelines with corrosion thinning defects repaired with ultra-high-performance concrete. The variation law of the residual bearing capacity of reinforced concrete pipes and the influence of different corrosion degrees, repair thicknesses, and interface forms on the bearing capacity of reinforced concrete pipes were studied following UHP-ECC for pipe repair. The results showed that the bearing capacity of the structure decreased with an increase in the corrosion thickness of the pipeline. After repair with ultra-high-performance concrete, the bearing capacity of corroded pipelines greatly improved. When the corrosion and repair thicknesses were the same, the bearing capacity of the repaired pipeline with different interface forms was very different. After the interface was implanted with nails, spray repair was carried out and the bearing capacity of the pipeline improved the most, followed by the naturally bonded interface. When plastic film was pasted on the repair interface, the bearing capacity of the pipeline improved the least.

1. Introduction

A high sulfate concentration and low pH value of wastewater are suitable for the growth of sulfur-oxidizing bacteria, which produce sulfuric acid, making the reinforced concrete drainage pipes within a wastewater environment susceptible to long-term acid corrosion [1]. Reinforced concrete pipes then crack and spall, causing a gradual reduction in their wall thickness due to corrosion, which decreases the bearing capacity of the pipes and can potentially cause structural fail; therefore, timely repair of reinforced concrete pipes is necessary [2].

The spray construction method, originally used as a repair technology for steel pipes via manual application, aims to stop further corrosion of internal pipe walls. In 1933, American company Centriline introduced the cement mortar spray construction restoration technique, which utilizes machinery for mortar spraying and is commonly referred to as the Centriline technique. In 1934, the city of Newark, NJ, USA, employed the spray construction method to repair an 8.4 km section of a steel pipe, solving the problem of leakage and improving the flow capacity of the original pipeline [3]. In 2003, during the renovation of the Shanghai Yang Shupu water plant outlet pipe, it was determined that the pipe has been initially repaired with cement mortar during its construction [4].

Scholars around the world have researched the coordinated deformation relationship between the lining and existing pipelines. Zhao analyzed the concrete pipe–liner–mortar layer relationship and concluded that the three are completely bonded; that is, the composite structure, and the three as a whole, bear the load together [5]. Zhang determined that when the coordinated deformation of the existing pipeline lining depends on the relationship between the load on the interface and the interface resistance, and if the resistance is greater than the load, the two form a superimposed composite structure. An indoor test of the composite structure model indicated that the bearing capacity of the upper and lower beams could not be fully exerted due to the gap caused by the shrinkage of the mortar layer, which was approximately 0.6–0.9 times the theoretical value [6]. WRC studied the influence of the modulus of the mortar layer on the bearing capacity of the repaired pipeline. It was found that the deformation of the lining pipe was 0.16% and 0.08% at 8 and 12 MPa, respectively, while deformation without lining was 0.56%. The contribution of the mortar layer to the bearing capacity of the repaired pipeline was higher than that of the pipeline lining [7]. Shi et al. studied the influence of the lining thickness, deformation degree, and reinforcement ratio on the bearing capacity of existing pipelines based on a trilateral load test of mortar-repaired reinforced concrete pipelines [8]. Most existing theories regard the lining layer and the existing pipeline as primary force structures, though the interface between the lining layer and the pipeline may be detached under an external load. However, there is a lack of research on the overall mechanical properties of reinforced concrete composite members under corrosion conditions [9].

In this paper, the residual bearing capacity of reinforced concrete pipes with different degrees of corrosion was simulated, and the variation law of the residual bearing capacity of reinforced concrete pipes was studied using a ring stiffness instrument loading test. Ultra-high-performance concrete materials were used to repair the corroded and thinned reinforced concrete pipes, and three parameters were set up, including different corrosion and repair thicknesses and different interface forms. The control variable method was used to study the influence of each parameter on the bearing capacity of reinforced concrete pipes, and the applicability of ultra-high-performance concrete material as a spray repair material was analyzed. The bearing capacity of the entire structure, with the corrosion-thinning reinforced concrete pipe repair lining layer using this material, was studied. This revealed its mechanical characteristics under an external load, providing a basis for designing ultra-high-performance concrete repair lining.

2. Test Program

2.1. Specimen Design

In this experiment, DN200 reinforced concrete pipe sections with a wall thickness of 30 mm and a length of 300 mm were made, and different wall and repair thicknesses were set. DN400 reinforced concrete pipe sections with a wall thickness of 40 mm and a length of 300 mm were also used, and different interface forms were set. The specific parameters are shown in

Table 1. Design parameters of the DN200 pipe sections.

Group	Degree of Corrosion	Thickness of Repair
F-1	1% (2 mm)	0 mm
F-2		10 mm
F-3		20 mm
F-4	2.5% (5 mm)	0 mm
F-5		10 mm
F-6		20 mm
F-7	5% (10 mm)	0 mm
F-8		10 mm
F-9		20 mm

and

Table 2. Design parameters of the DN400 pipe sections.

Group	Interface Form	Thickness of Repair
S-1	Natural bonding	250 mm
S-2	Interfacial nail implantation	250 mm
S-3	Pasted plastic film	250 mm

.

The design of the DN200 pipe reinforcement is shown in

Table 3. The DN200 pipe’s reinforcement parameters.

Rebar Index	Design Value
Circumferential tendons	Diameter/mm	3
	Ring inner diameter/mm	221
	Ring count	18.5
	Pitch/mm	54.1
Longitudinal tendons	Diameter/mm	5
	Number of roots	6

.

Silicate cement with a strength grade of 52.5 was selected for the preparation of UHP-ECC repair materials. UHP-ECC is a new type of ultra-high-performance engineering cement-based composite material with high mechanical properties, composed of cement, fine sand, other admixtures, and other materials. The addition of ethylene-vinyl acetate copolymer (EVA) enhances the dispersion of the material, while polyvinyl alcohol fiber (PVA) is randomly distributed within it. It has high mechanical properties and meets the requirements of fast hardening, early strength, and corrosion resistance [10,11,12]. The raw test material mix is shown in

Table 4. Raw material mix ratio (unit: g).

Cement	Fly Ash	Silica Fume	Quartz Sand	PVA Fiber	VAE Glue Powder	Water Reducer
780	120	100	1000	10	20	20

.

2.2. Measurement Point Arrangement and Loading Scheme

Resistive strain gauges were attached at the top, bottom, and arch lines inside and outside the pipe sections to measure the strain at each critical point. The top, bottom, and arch lines of the outer wall of the pipes were marked as OC, OI, OL, and OR, respectively. Meanwhile, the top, bottom, and arch lines of the inner wall of the pipes were marked as IC, II, IL, and IR, respectively. The locations of strain measurement points are shown in Figure 1.

A YWC-30 displacement meter was used to measure the vertical displacement of the top and bottom of the inner wall of the pipe sections; the position of the displacement measurement point is shown in Figure 2, in which the displacement of the top and bottom of the inner pipe are recorded as VD1 and VD2, respectively [13].

The loading test was conducted according to the external compression load test method GB/T 16752-2016 “Test Methods for Concrete and Reinforced Concrete Drainage Pipe” [12]. The upper support beam was made of steel blocks and the lower support beam was made of two battens with a width of 80 mm and a thickness of 40 mm. Regarding the DN200 pipe, the spacing between the two battens was set to 25 mm; for the DN400 pipe sections, the spacing between the two battens was set to 40 mm. The test loading speed was set to 5 mm/min and the loading displacement to 10% of the pipes’ diameter [14,15].

3. Test Results and Analyses

3.1. The Effect of Corrosion Thickness on the Bearing Performance of the Pipes

3.1.1. Comparative Analysis of the Load-Bearing Performance of Corroded Pipes

The plate load test results of specimens F-1, F-4, and F-9 are shown in

Table 5. Flat plate load test results of the corroded pipe sections.

Group	Crack Load Dc/kN	Ultimate Load Du/kN
F-1	6.19	11.83
F-4	5.36	10.12
F-7	4.51	7.69

, and the load–displacement curves of each specimen are shown in Figure 3. From Table 5, it can be seen that the cracking load of specimens F-4 and F-9 was reduced by 13.41% and 27.14% compared to specimen F-1, and the ultimate load was reduced by 14.45% and 35.00%, respectively. The test results show that with an increase in the corrosion wall thickness, the cracking load and ultimate load of the pipeline gradually decreased, and the greater the degree of corrosion was, the greater the reduction in the bearing capacity was. This is because, when the corrosion thickness is small, the thickness of the pipe sections is the thickness of the protective layer, which has little effect on the bearing capacity of the pipeline.

From Figure 3, it can be seen that the load change rules of the three specimens (F-1, F-4, and F-7) are basically the same, and that there are three stages: (1) The elastic stage, which encompasses the initial loading to the crack load. The load increases rapidly at this stage, the deformation of the specimen is small, and cracks are generated. (2) The crack development stage, which extends from the cracking of the specimen until the ultimate load is reached. At this stage, the original cracks continue expanding, and the load continues to increase to the ultimate load under the support of the steel bar. (3) The third stage is the yielding stage. After reaching the ultimate load, the entire specimen no longer works. Reinforcement reaches the yielding stage, the load decreases slowly, and the deformation of the specimen keeps increasing, but the specimen is not fractured [16,17].

The ultimate load of specimen F-7 was lower and the rate of the load decrease at the yielding stage was higher. This is because the protective layer of the specimen was almost completely lost, and the corrosion thickness in some areas reached an effective protective layer thickness. During the late stage of loading, the reinforcements at the top, bottom, and arching lines of the pipes peeled off, and the supporting effect of the reinforcements was weakened, while the rate of the load value increase during the crack development stage was reduced substantially. With the increase in corrosion thickness, the specimens exhibited faster cracking and reached a smaller crack load displacement.

3.1.2. Comparative Analysis of the Pipes’ Load-Bearing Performance after Rehabilitation

The load–displacement curve of pipe sections with the same repair thickness at different corrosion depths is shown in Figure 4. From the figure, it can be seen that when repairing pipe sections with different corrosion thicknesses, the load–displacement curves of the whole structure formed using a repair lining layer of the same thickness are basically the same. The load–displacement variation was approximately linear before the cracking load was reached during the early stage of loading, and when the pipe sections cracked, the slope of the load curve gradually decreased as the load increased and the crack continued to expand. However, the greater the corrosion thickness, the smaller the slope of the load curve. As shown in

Table 6. Plate load test results of pipe sections with the same corrosion thickness.

Group	Ultimate Load Du/kN
F-2	24.87
F-5	21.73
F-8	14.96
F-3	35.64
F-6	31.08
F-9	27.09

, with the initial increase in the corrosion thickness of the pipe sections, the ultimate load of the overall structure decreased. Considering the initial thinning of the pipe sections’ thickness, to reduce the load-bearing capacity and increase the lining layer between the structural stiffness gap of the pipe sections themselves, a certain thickness of the lining layer was unable to effectively inhibit the overall structure of the deformation. Combined with the damage of specimen F-8, when the corrosion depth reached a thickness that impacted the load-bearing capacity, the existing pipe sections underwent structural deformation, resulting in a significant weakening of the interfacial bonding. The laminated structure then quickly transformed into a composite structure, and the lining layer of the bearing role could not carry out its full role, resulting in a decline of the performance of the repaired pipe sections.

The results show that when the corrosion depth did not reach the thickness of the protective layer, the spraying repair of the pipeline effectively reduced the occurrence of a local rupture, leakage, and other defects caused by structural deformation of the pipeline. Simultaneously, timely repairs were conducted on the corroded pipeline with a relatively small spraying thickness, thereby ensuring the flow capacity of the existing pipeline.

3.2. Influence of the Repair Thickness on the Load-Bearing Performance of Repaired Pipes

The load–displacement curves of specimens F-1 to F-3 are shown in Figure 5. The improvement of the bearing capacity of F-2 and F-3 compared to F-1 in their initial state is shown in

Table 7. Comparison of the flat plate load test results of specimens F-1 to F3.

Group	Crack Load Dc/kN	Ultimate Load Du/kN
F-1	6.19	11.83
F-2	9.8	24.87
F-3	15.41	35.64

. Here, it can be seen that the change trend of the load–displacement curve of each specimen after repair is basically the same: The larger the repair thickness, the greater the rate of the load increase, while the greater the crack load of the specimen, the greater the ultimate load of the specimen. As seen from Table 7, compared to the initial state of specimen F-1, the crack load of specimens F-2 and F-3 increased by 58.32% and 148.95%, respectively, while the ultimate load increased by 110.23% and 201.28%, respectively.

The load–displacement curves of specimens F-4 to F-6 are shown in Figure 6. The improvement of the bearing capacity of F-5 and F-6 compared to F-4 in the initial state is shown in

Table 8. Comparison of the flat plate load test results of specimens F-4 to F-6.

Group	Crack Load Dc/kN	Ultimate Load Du/kN
F-4	5.36	10.12
F-5	16.85	21.73
F-6	22.83	31.08

, where it can be seen that the load change trend of each specimen at this stage is basically the same from the initial stage of loading to the ultimate bearing capacity. As can be seen from the table, compared to the initial state of specimen F-4, the crack load of the specimens F-5 and F-6 increased by 214.37% and 325.93%, respectively, while the ultimate load increased by 114.72% and 207.11%, respectively.

The load–displacement curves of specimens F-7 to F-9 are shown in Figure 7. The improvement of the bearing capacity of the initial defect pipeline F-7 by specimens F-8 and F-9 is shown in

Table 9. Comparison of the flat plate load test results of specimens F-7 to F-9.

Group	Crack Load Dc/kN	Ultimate Load Du/kN
F-7	4.51	7.69
F-8	9.97	14.96
F-9	13.37	27.09

. From Figure 7, it can be seen that the load–displacement curves of specimens F-7 to F-9 are approximately the same as those of the above specimens. As can be seen from Table 9, compared to the initial state of specimen F-7, the crack load of the specimens F-8 and F-9 increased by 121.06% and 196.45%, respectively, while the ultimate load increased by 94.54% and 252.28%, respectively.

The above test results show that ultra-high-performance concrete material can effectively improve the load-bearing performance of pipes by repairing reinforced concrete pipes. As the repair thickness increased, the inhibiting effect of the lining layer on crack expansion increased, and the longer the whole pipeline lining in the state of a superimposed structure [18]. The crack and ultimate loads of the original pipes increased significantly, the bearing effect of the lining layer was given full play, and the load-bearing performance of the existing pipes greatly improved.

3.3. Influence of the Interface Form on the Load-Bearing Performance of Rehabilitated Pipes

By repairing the DN400 pipe sections with three interface forms, including the naturally bonded interface, interfacial nail implantation, and plastic film adhesion, and setting the repair thickness to 25 mm, the pipe was loaded using a ring stiffness testing machine to obtain the damage form and load–displacement curve of the pipe sections [19,20].

3.3.1. Damage Form

The damage form of specimen S-1 (naturally bonded interface) is shown in Figure 8, and the crack width is shown in Figure 9. At the initial stage of loading, there was no damage to the specimen. When the displacement reached 2.30 mm, penetrating cracks occurred at the top and bottom of the pipes. The crack at the top of the tube is shown in Figure 9a, and the crack width was 0.29 mm. Subsequently, cracks appeared above the arching line on both sides of the pipe sections and the load was 11.40 kN. When the displacement reached 2.95 mm, the load increased to an ultimate load of 11.70 kN. With the increasing load, a crack developed along the interface. The interface peeled off when the displacement reached 3.24 mm, as shown in Figure 9b. At this time, the crack width was 0.60 mm and the load value was 11.31 kN. During the later stage of loading, the specimen no longer produced new cracks; the cracks at the top of the pipes expanded rapidly and the specimen failed. The cracks at the top and bottom of the tube are shown in Figure 9c,d, and the crack widths were 4.96 and 1.25 mm, respectively.

The damage form of specimen S-2 (interfacial pegging) is shown in Figure 10, and the crack width is shown in Figure 11. From Figure 10, it can be seen that when the displacement reached 2.60 mm, penetrating cracks occurred at the top and bottom of the pipes. As shown in Figure 11a,b, the crack widths were 0.35 and 0.21 mm, respectively. Moreover, two center-symmetric cracks appeared near the arch line of the pipe sections, and the load at this time was 20.83 kN. When the displacement reached 2.95 mm, the load increased to 21.96 kN. With further increases in the load, the cracks continued to expand. However, unlike those pipe sections with a naturally bonded interface, there was no interfacial peeling between the pipe and the lining within the pipe sections where the interface was pinned. The nailed pipe sections at the interface only led to failure of the specimen due to excessive cracks. The cracks at the top and bottom of the pipes at the time of damage are shown in Figure 11c,d, with crack widths of 2.73 and 2.33 mm, respectively.

The failure mode of specimen S-3 (bonded plastic film) is shown in Figure 12, and the crack width is shown in Figure 13. In Figure 12, it can be seen that when the displacement reached 1.32 mm, the load reached 10.71 kN. At this time, cracks appeared on top and the left arch line of the pipe, while two cracks appeared approximately 135° clockwise from the top of the pipe. At the same time, the left and right interfaces peeled off and the cracks were measured, as shown in Figure 13a,b. The crack widths were 0.61 and 0.38 mm, respectively, and the interfacial bonding effect failed. At this time, the ultimate load was reached. As shown in Figure 13a–d, the crack widths on both sides of the pipe sections were 0.88 and 0.55 mm, respectively, when the ultimate load was reached. Because the interface peeled off at this time, the crack propagation speed was slow and its width was small. Due to the failure of the interfacial bonding, the lining layer could not play a bearing role, resulting in a lack of further improvement of the bearing capacity during the later stage of loading, and the cracks continued to expand. At this time, the left interface of the pipe sections completely peeled off, the right interface partially peeled off, new cracks appeared along the interface to the outside, and the bearing capacity was maintained at a low level.

3.3.2. Comparison of the Ultimate Load and Load Capacity Improvement

The ultimate load of each pipe section with different interface forms is shown in

Table 10. Ultimate loads of the pipe sections of each boundary form.

Group	Ultimate Load Du/kN
S-0	6.32
S-1	11.70
S-2	21.96
S-3	10.71

, and the ultimate load of pipe section S-0 in the initial state was 6.32 kN. The improvement of the bearing capacity of three interface forms of the initial pipe section after repair was 69.46%, 85.13%, and 247.47%, respectively; these interface forms were a gluing plastic film, a naturally bonded interface, and an interfacial nail interface. The results show that in the case of the same corrosion and repair thicknesses, the bearing capacity of the repaired pipe section was greatly improved by using the interfacial nail treatment method, which was approximately 3.56 and 2.91 times that of the bonded plastic film and the naturally bonded interface, respectively. This is due to the adhesion of the plastic film reducing the interfacial bonding force between the lining layer and the existing pipeline. The pipeline was in an approximate composite structure state, and the interface between the two layers peeled off during the loading process, thereby hindering any further improvement in the overall structure’s bearing capacity. The interfacial nail implantation strengthened the interfacial bonding between the pipeline and the lining, so that the pipeline was in a composite structure state under the action of an external load. The coordinated deformation relationship between the lining layer and the existing pipeline was enhanced, so that the bearing capacity of the lining layer was fully utilized and the bearing capacity of the overall structure was greatly improved. This also shows that when repairing existing reinforced concrete pipes, the bearing capacity of the pipe section can be improved by planting nails on the interface before repairing it to enhance the interfacial bonding effect. The appropriate design of the lining structure and wall thickness can effectively reduce the loss of the flow section [21,22].

3.3.3. Load–Displacement Curves

Load–displacement curves of each interface form were obtained, as shown in Figure 14. Observing the load–displacement curves of each specimen, it is clear that the curve trends are basically the same. However, when considering the crack development trend, the pipe sections with a plastic film pasted on it exhibited rapid cracking and interfacial peeling. The naturally bonded pipe sections also showed interfacial peeling during the loading process, while the crack development of the pipe sections with interfacial nailing was relatively slow and there was no interfacial peeling. Additionally, based on the trend of load changes after cracking, the bearing capacity of the pipe sections with a pinned interface no longer increased during the continued loading process due to serious interfacial peeling and remained at a certain level, while the bearing capacity of the pinned and naturally bonded interfacial pipe sections underwent a certain increase after the pipe sections cracked. This illustrates that as the interfacial bonding increased, the longer the pipe in the laminated structure, the more effectively the liner layer functions, resulting in an increased bearing capacity of the after the repair was carried out [23,24].

4. Conclusions

In this paper, corrosion-thinning reinforced concrete pipe sections were simulated by manually grinding the concrete protective layer. Ultra-high-performance concrete (UHP-ECC) was used to repair the corroded pipe sections. The influence of different corrosion thicknesses, repair thicknesses, and interface forms on the bearing capacity of the repaired pipe sections was studied, employing the control variable method. By analyzing the crack development law, failure mode, and load–displacement curve of each pipe section, the following conclusions were drawn:

(a)

Following pipeline thinning due to corrosion, the bearing capacity is reduced, the corrosion thickness increases, and the degree of structure deformation damage increases. When the corrosion depth reaches the thickness of the protective layer, the residual bearing capacity of the pipeline is reduced by 35%.

(b)

Regarding certain other parameters, with the increase in corrosion thickness, the degree of structural deformation of the existing pipeline increases, the coordinated deformation relationship between the lining layer and the existing pipeline becomes weaker, and the ultimate bearing capacity of the pipeline decreases. The pipeline should be repaired in a timely manner when the degree of corrosion is low, which can reduce the repair thickness to achieve a larger increase in the bearing capacity, thereby reducing the loss of the overflow section.

(c)

For the same corrosion and repair thickness, the bearing capacity of a pipeline with a plastic film after repair is lower than that of a pipeline with a naturally bonded interface. The bearing capacity of a pipe with interfacial nail implantation is higher than that of the naturally bonded interface and the improvement is more significant.

(d)

A bonding plastic film weakens the interfacial bonding between the lining layer and the existing pipeline. Under the action of an external load, the two cannot fully coordinate the deformation and the interface is more likely to peel off, resulting in a decrease in the bearing capacity of the overall structure. Interfacial nail implantation can effectively improve the interfacial bonding effect of the pipeline, so that the overall structure reaches a superimposed structure state under an external load, and the bearing performance of the pipeline is greatly improved. When the reinforced concrete pipeline is repaired using the spraying method, the coordinated deformation relationship between the lining layer and the existing pipeline can be improved by means of interfacial nail implantation, so as to reduce the repair thickness as much as possible and ensure the flow capacity of the pipeline.