Keywords

1 Introduction

Carbon nanomaterials, including carbon nanotubes (CNTs), graphene oxide and carbon nanosheets, have been widely recognized for their enhancement of cement-based composites  [1, 2]. The strengthening mechanism of CNTs in cement-based composites has been the focus in recent decades [3, 4], and it is though physical properties such as extremely high mechanical strength and large specific surface area that CNTs fill microscopic pores in cement matrix and promote the development of hydration products [5]. The resulting compact microstructure greatly optimizes the macroscopic mechanical properties of cement-based composites, including compressive strength, impermeability and corrosion resistance [6, 7]. At the same time, the addition of CNTs is potentially a way to reduce CO2 emissions from cement production [8].

In construction and mining projects, cemented rockfill with higher mechanical properties is significant for structural stability and production safety. CNTs improve the mechanical strength of cemented rockfill materials [9], but due to the lack of sensitivity of current monitoring systems, the effect of CNTs on failure patterns during the loading process needs further research. Acoustic emission (AE) technology is a nondestructive testing method widely used in the study of the mechanical properties of rock materials [10]. During the failure process, the gradual development of microfractures will cause acoustic signals, which can be monitored by the AE system. The waveform of the AE signal reflects the microscopic failure pattern. For example, the RA (Rise time to Amplitude ratio) and AF (AE counts to Duration ratio) can be used as criteria for the type of failure [11]. Therefore, the AE technique is effective for studying the destruction mode of the cemented rockfill.

In this study, we add a very low content of CNTs with fly ash (FA) to partially replace cement in the cemented rockfill. The tensile failure strength of the specimens was measured by the Brazilian split test and simultaneously the AE system collected signals during the destruction process. Combing the AE activity, and the stress–-strain curve, the distribution and intensity of the failure events in each loading stage of the CNT-enhanced specimens were analyzed. The failure event accumulation of the specimen is discussed according to the peak and growth rate of the cumulative AE counts curve. The b-value in AE technology is calculated to characterize the severity of the overall failure process. The findings in this study promote the engineering application of CNTs and we discuss the enhancement mechanism of CNTs in cemented rockfill materials.

2 Methods

Waste coal gangue extracted from a coal mine in Shanxi province, China, was selected as the solid particles in cement rockfill. The composition of coal gangue does not react chemically with the cement. Ordinary Portland cement (PO. 42.5), conforming to the Chinese Standard GB-175-2007 [12], was the binder material. Multiwall CNTs (MWCNTs; manufactured by Nanjing XFNANO Materials Tec. Co., Nanjing City, Jiangsu Province, China) were added to the rockfill materials. Polycarboxylate superplasticizer was added to promote CNT dispersion in the suspension.

The preparation of CNT-enhanced cement rockfill samples mainly comprised the preparation of CNT–cement slurry and mixing of the slurry with solid particles. The process is detailed in our previous report [13]. A Ref-group and CNT-group were set in this study, and a 0.4 water-to-cement ratio was applied in both groups. In the CNT-group mortar, there was 20% FA with 0.05 wt% CNTs to replace 20% cement. The other preparation processes of the two groups were consistent.

We chose 28-day-old standard-cured disc samples with a diameter of 50 mm and a thickness of 25 mm for the Brazilian split tests. The loading rate was 0.10 mm/min. To avoid the influence of errors in the test and improve accuracy, three samples were selected for each group. During the loading process, the AE signals were monitored by a Micro-II AE system (developed by the American Physical Acoustic Corporation), as shown in Fig. 1. The AE threshold and frequency were set at 30 dB and 10 Msps, respectively.

Fig. 1
A schematic representation. It indicates stress, specimen, A E signals, and loading procedure.

Brazilian split test and acoustic emission (AE) signal monitoring

Full size image

3 Results

3.1 Mechanical Properties of Cemented Rockfill

After curing for 28 days, the peak tensile strength of the specimens is shown in Fig. 2. The peak strength of the CNT-group was 3.4 MPa, an increase of 17.2% compared with the Ref-group, indicating that CNTs significantly enhanced the ability of the sample to resist tensile stress, which was consistent with previous studies [14]. Higher tensile strength can reduce the risk of sudden failure and improve engineering safety.

Fig. 2
A vertical bar graph with error bars plots tensile strength in megapascals for the ref-group and the C N Ts-group. The tensile strengths of Ref-group and C N Ts-group are 2.8 and 3.4, respectively. The values are approximated.

Tensile strength of two groups of cemented rockfill specimens. CNTs, carbon nanotubes

Full size image

The increase in tensile strength was attributed to the strengthening effect of CNTs in the cement matrix. CNTs show a bridging effect between microfractures in the cement matrix [15]. When the sample is under tensile stress, the high tensile strength of CNTs can help the cement matrix resist external stress and avoid microcrack development [16]. As the sample’s hydration reaction continued, the hydration products became more closely connected with the CNTs and the resultant compact internal structure enabled the specimen to resist failure and deformation.

3.2 Analysis of AE Activity During the Failure Process

The stress–strain curve and real-time AE signals of the samples are shown in Fig. 3. As the stress gradually increased, the stress–strain curve divided into four failure stages. In the initial compaction stage and elastic deformation stage, there was almost no fracture inside the sample, accompanied by sparse AE signals. During the plastic deformation stage, cracks gradually developed and the AE activity gradually became denser. When peak strength was reached, the samples suddenly produced a lot of AE counts, corresponding to the development of major fractures. In the post-peak stage, cracks continue to expand.

Fig. 3
2 graphs labeled a and b. They depict stress sigma in megapascals, A E counts rate in times 10 power 2 seconds power negative 1, and cumulative counts in times 10 power 3 versus epsilon in times 10 power negative 3 for ref-1 and C N T-group.

Acoustic emission counts of the Ref-group (a) and CNT-group (b) during the Brazilian split test. CNTs, carbon nanotubes

Full size image

With the addition of CNTs to the cemented rockfill, the AE activity became sparse and AE counts decreased significantly during the failure process. The cumulative counts of the Ref-group reached 20(×103), much higher than in the CNT-group. At the peak strength point, the Ref-group showed a single and high-count AE event, whereas the CNT-group showed multiple peaks of AE activity. The reason for this phenomenon is the promotion of a dense microstructure by CNTs. The defect area is smaller, reducing the severity of failure, resulting in fewer counts and lower frequency of AE events, which leads to more gradual macro disruption.

The b-value reflects the ratio of the number of lower and higher grade AE events during the rock failure process, and can be calculated by the following equation [17]:

$$lgN_{(A/20)} = \, a - b \times \left( {A/20} \right)$$
(1)

where A is the amplitude of the AE event, N(A/20) is the cumulative frequency of AE events with an amplitude ≥ A, and b represents the b-value.

According to Eq. 1, the b-value of the CNT-group reached 1.27, increasing 14.8% more than the Ref-group (Fig. 4), which reflects the AE activity being mostly concentrated at lower energy levels during the failure of the sample. The reason behind this phenomenon is that CNTs promote the development of hydration products to reduce the size of micropores, further avoiding the high energy release of cracks during loading.

Fig. 4
A vertical bar graph with error bars plots the b-value for the ref-group and the C N Ts-group. The b-values of Ref-group and C N Ts-group are 1.19 and 1.27, respectively. The values are approximated.

The b-value of two groups of cemented rockfill specimens. CNTs, carbon nanotubes

Full size image

4 Discussion and Conclusions

In this study, CNTs and FA were added to a cemented rockfill mix to investigate the production of environmentally friendly and high-workability materials. The tensile properties of the three specimens in each group were tested by the Brazilian split test and AE signals were monitored by a PCI-2 system. The failure mode pattern of the CNT-reinforced cement rockfill material was studied by the stress–strain curve and AE waveform.

In the Brazilian split experiment, the CNT-reinforced sample showed increased peak tensile strength by 17.2%. At the same time, the AE events became sparser and the count decreased, indicating that the intensity of destruction decreased. Finally, the addition of CNTs increased the b-value by 14.8%, which indicated that CNT-reinforced cemented rockfill material will have fewer high energy release destruction events, thereby making it a safer option.