1. Introduction
Chemical reaction between compound of hydraulic cement and water yields ceramic products that achieve the binding property after hardening. This complex process or reaction between cement and water is called hydration. The cement hydration reaction varying in time involves the change from the suspension state to the solid with pore state. Solid hydration products usually form at the surfaces of the cement particles and in the pore solution by nucleation and aggregation. As a result, the solid phase becomes highly connected and the material transforms from a viscous suspension of irregularly-shaped cement particles into a porous elastic solid. The connectivity of the solid phase is responsible for the setting and hardening behaviours and load-bearing capacity of cement. The pore structure has significant implications for the transport property and further the durability of cement-based materials.
The physical setting, strength gain and lifetime of cements are accompanied by a series of mineralogical and microstructural changes during the hydration. Such hydration process has attracted a considerable attention and has been studied by many methods, like the commonly used Vicat setting time, strength measurements, scanning electronic microscope (SEM), X-ray diffraction (XRD) and so on. However, the thus measured setting time and strength parameters can not directly show the materials and the microstructural changes in hydration process. For example, the setting time defined by penetration values of Vicat needles (ASTM C191) correspond to two particular practical points, which are loosely defined as the limit of handling and the beginning of mechanical strength development, respectively. And also the test’s accuracy depends largely on the skill and experience of the person performing it . SEM and XRD need to stop the hydration process and do the certain sample pretreatments. Such methods can not provide the continuous information about hydration and moreover the sample will be influenced by the drying and cutting preparation. Therefore, alternative techniques, which are accurate and non-destructive, are highly needed to monitor the hydration process of cement-based material. Thus, the details about the complex hydration can be clarified. During the last decade, non-destructive testing (NDT) techniques have attracted increasing attention for the characterization of the behavior of concrete at early age. These techniques include electrical properties measurements and ultrasonic pulse velocity (UPV) tests . In this paper, the newly developed non-contacting electrical resistivity measurement and embedded piezoelectric ultrasonic system will be presented and discussed. They concern the continuous monitoring of the setting and hardening of the fresh cement paste and have proper control by reliable and more objective for the advanced process technology.
In the electrical resistivity method, the electrical conduction of cement-based media relies on ionic conduction through water-filled pores. The factors influencing electrical conduction for porous media are mainly controlled by microstructure and liquid solution conductivity. The research had shown that electrical measurement can be applied to provide useful information of the hydration process of the cement-based materials. To overcome the contacting problem between the sample with the electrodes commonly used in the traditional measurement, a non-contacting electrical resistivity device was invented [10]. Because of the transformer principle used, this method successfully eliminated the contact problem and accurately monitored the electrical resistivity of cement-based materials.
By using the ultrasonic measurement, as the hydration goes on, ultrasonic pulse propagation path in the cement system switches from the liquid phase to the solid phase. It was shown that this change in behavior results from a change in character of the observed wave as hydration proceeds. At early times the observed wave involves essentially motion of the fluid phase while at longer times it involves essentially motion of the solid frame. The UPV reveals a significant increase after the appearance of the connectivity of solid phase, as already shown by Krautkraemer and Krautkraemer [11] and Sayers and Dahlin[12]. Ultrasonic waves are therefore sensitive to the point at which the solid phase becomes interconnected. This point is of practical significance since the connectivity of the solid phase is responsible for the load bearing capacity of set cement. Because of its non-destructive and reproducible benefits, ultrasonic technique has been employed to monitor the hydration process. Different from the conventional measurement using the surface contact commercial ultrasonic transducers, a new ultrasonic measurement system, which uses embedded piezoelectric composites as the ultrasonic transducers, has been developed and used in this study. This method can eliminate the coupling and contacting problems caused by the traditional transducer; thus it makes the measurement more accurate and more suitable for the in-situ hydration monitor .
The principle aim of this study is to use the characterization of changes in the electrical resistivity and ultrasonic pulse velocity respectively to follow cement hydration and monitor the development of the microstructure at early age.
2. Experimental
2.1 Non-contacting resistivity measurement
The electrical resistivity of the specimen was measured by a non-contacting electrical resistivity measurement device, CCR-II (BCT,
2.2 Embedded piezoelectric ultrasonic method
In this research, a new ultrasonic measurement system, which uses embedded piezoelectric composites as the ultrasonic transducers, is adopted (Fig. 2). The working principle is that the piezo-composite vibrates when an electrical pulse is applied on it, and then the ultrasonic wave propagates along the direction of vibration. The ultrasonic waves are received by other piezoelectric composites sensor, working as receiver. Comparing with traditional ultrasonic non-destructive methods, which have conventional commercial ultrasonic transducers fixed on the surfaces, the new method eliminates the coupling and surface contacting problems and is inexpensive and effective for any-scale structures. For construction of a concrete structure, early-age performance monitoring can be performed by the new system to provide guidance for construction control. After the concrete is cured and the strength is fully developed, the system can be used to conduct structural health monitoring to detect the damage accumulation or even disaster evolving. It meets all the requirements of concrete structure health monitoring from fresh stage to hardened stage.
In this study, the transducers were fixed on the bottom of a plastic box of 250 × 320 × 100 mm３ to monitor the early-hydration of cement paste. Cement paste was poured into the box and vibrated to remove the entrapped air bubbles. An electrical pulse was first generated by an Agilient 33120A functional generator. After amplifying by a power amplifier, the input electrical pulse excited the transmitter, and then vibrated in its resonance modes, emitting ultrasonic waves. A longitudinal ultrasonic wave was received by the receiver through changing mechanical energy into electrical energy. The voltage of the receiver could be measured and recorded by the oscilloscope though a pre-amplifier. The oscilloscope was a 12 bit Agilent 5462A one with a sampling frequency of 1 MHz. After casting the fresh sample, the measurement was immediately started.
2.3 Specimens (Materials and mix proportion)
In this study, the cement-based specimen used was ordinary Portland cement (OPC) paste with 0.4 water/cement ratio (w/c). The specimens for both resistivity and ultrasonic tests were from the same mixing batch and cast at the same time. The testing ambient condition was kept consistent with the specimen sealed with 100% humidity and the same laboratory temperature. Testing continues for 7 days without any breakage and hydration cease. The traditional Vicat method was also applied in this study to verify the setting times.
3. Result and discussion
3.1 Electrical resistivity result
Figure 3(a) and Figure 3(b) show the bulk electrical resistivity development with time up to 1 day (24 hours) and 7 days (168 hours), respectively. The resistivity-time curve begins with a small decline corresponding to the dissolution of ions from cement in water. After reaching the minimum point at half an hour after mixing time, the resistivity of the specimen increases a little for several minutes and then shows a nearly level development at very low value and keeps this level process for 3 hours. In this flat stage, the initial hydration products form in solution, which reduces the ion concentration and allows additional ions from cement particles dissolved. After 3 hours of hydration, the resistivity increases gradually due to the growth of hydrates. This increase continues through the whole hydration.
From the plot of rate change of resistivity with time shown in Figure 4, the hydration process stages can be easily identified. There are four stages can be divided in the first 7 days hydration. In the first stage (Stage I), the differential of resistivity is below zero. This negative value correlates with the initial decline of resistivity. On first contact with water, the mobile ions are rapidly released from the surface of each cement grains; the pH values rise to over 12 within a few minutes. This process is hydrolysis. Because of ion dissolution, the conductivity of cement water mixture will increase and the resistivity decreases inversely. After reaching the saturation, this initial hydrolysis slows down quickly, which is corresponding to the minimum point in the resistivity and also the zero value of the differential resistivity. The first peak in Figure 4 is related to the first small increase of resistivity. Then, in the next 3 hours, near zero values in differential of resistivity indicate that the mixture is relatively inactive which refers to a competition between the dissolution and precipitation, identified as a dynamic balance, or called competition period (Stage II). The dissolving and consuming ions compete to reach a balance. As the hydration continues, the thickness of the hydrate layer increases and forms a barrier through which water must flow to transport the free ions and through which ions must diffuse to conduct the electricity. The resulting rapid increase of differential resistivity corresponding to the accelerated increase of resistivity is the third stage (Stage III), which is observed from 3 to 10 hours. At about 10 hours of hydration, the differential resistivity reaches the maximum, showing the fastest increase of resistivity. After that, a decelerated increase of resistivity continues for longer than 7 days (Stage IV). With hydration degree increase, the resistivity of the sample increases due to the growing solid phase and their interconnection, which will block the way of transportation of the ionic conductors in the pore solution. Eventually, movement through the C-S-H layer determines the rate of reaction and hydration becomes diffusion controlled. The resistivity increases much slowly.
3.2 UPV result
Figure 5 shows the development of ultrasonic pulse velocity in fresh concrete paste measured by the piezoelectric ultrasonic method. In the first three hours, the UPV remains nearly constant around 1500 m/s. This period (Period 1) is related to the stage I and II of cement hydration in Figure 4. It is considered the paste in this period as an emulsion with dispersed solid particles uniformly. In this measurement, it was observed that the velocity started from about 1550 m/s, and decreased slightly before the rapid rise in velocity. The average value of 1500 m/s is very close to 1490 m/s, the wave velocity in water. Based on the property of emulsion-like cement paste in this stage, the velocity value looks very reasonable. A period of rapid increase (Period 2) in the UPV, which corresponds to the rapid development of hydration products, appears after the flat velocity period. This period started at the similar time as that of stage III in Figure 4. When a critical quantity of hydration product is reached, the percolation of solid phase seems to occur and UPV starts to increase. In this period, more and more hydration products continues to be intersected, stiffness or modulus of material increases rapidly. As a result, the UPV increases notably. Period 3 in UPV corresponds to period IV in resisting measurement, implying that the hydration process of cement proceeds into
C3
C3
As the water within the pore system becomes saturated with the ions, the resistivity reaches the minimum point, showing the end of the dissolution period. Therefore, it may be well concluded that the resistivity development during this stage is dominated by the ions dissolution. There are also chemical reaction occurrences and the products were detected by Xiao .
In this stage, ultrasonic wave propagates through the water-like viscous suspension. Previous study [12] considered the system in this period as a dispersion of solid particles in a liquid. Cement particles are isolated from each other, while the capillary pore water is connected. The observed ultrasonic wave involves essentially motion of the emulsion phase. The UVP remains constant at a very low value close to the wave velocity in water.
• Stage II: competition stage (0.7-3 hour);
Studies show that at high dilutions a gel layer or membrane forms over the surfaces of the grains soon . Short rods of ettringite are probably more abundant near to the surfaces of the aluminate phase, and appear to nucleate in the solution and on the outer surface of a layer of gel. The ions in the solution are readily absorbed by the formation of a thin layer of hydration products. The initial hydration reaction consumes a few ions so that the resistivity of the specimen shows a small increase. This small increase detected by the sensitive and accurate non-contacting resistivity measurement is the first discovery on the resistivity study of the fresh concrete. The initial hydrates form an envelope around the unhydrated cement grains. These effects reduce the ion concentration and meanwhile allow additional ions from cement particles dissolved. Therefore, this competition of consuming and releasing ions leads to a dynamic balance in the resistivity-time curve and a slow increase of the temperature curve as well.
In this stage, ultrasonic waves also propagate through the viscous suspension. There is no obvious change in UPV curve compared with the value in stage I. UPV keeps very low, though a slight decrease caused by an increasing tortousity of the pore solution due to the formation of hydration products. This result is similar as the previous studies [6,8,18]. There is no significant change in UPV curve in stage I and II, showing that the ultrasonic method is not sensitive at the very beginning of the hydration because during these two stages the chemical reaction plays an important role in the cement hydration process.
• Stage III: acceleration stage (3-10 hour);
In stage III, the rate of cement hydration increases sharply because of the growth of hydrates, leading to hardening. In this stage, C3S reaction dominates hydration. C-H continues to crystallize from solution with the concomitant development of C-S-H (C3S2H8) from the surface of C3S to form a coating covering grain and the reaction of C3S proceeds rapidly. During this period, which begins at about 3 h and ends at about 10 h, some 30% of the cement reacts . It coincides with that of strong heat evolution and is characterized by the rapid formation of C-S-H, ettringite and
Studies show that the undried C-S-H has a filmy, foil-like morphology. The C-S-H forms a thickening layer around the cement grains which engulfs and perhaps nucleates on the ettringite rods. The CH forms massive crystals in the originally water-filled space. Nucleation sites appear to be relatively few in number, and the growing crystals may engulf some of the smaller cement grains. Small isolated clusters of solid substance are formed. These small isolated clusters are the basis of the solid network. The solid network continues to develop until it becomes connected throughout the material. At this critical time, a solid percolation formed, which is the start of setting. As the hydration going on, more and more hydrates form the solid network. The shells grow outwards and those surrounding adjacent grains are beginning to coalesce. At this stage, fracture through the shells begins to supplant fracture between them. It coincides with the maximum rate of heat evolution and corresponds approximately to the completion of setting. Thus, the stage III completes.
The structure of interconnected shells has been considered to play an important part in determining the mechanical and other properties, which thus depend on the particle size distribution of the cement. In this stage, the rapid increase of resistivity is related to a relatively small but significantly decrease in the number of ionic species. The pores inside the specimen are likely to be filled with a highly concentrated or colloidal solution, and the shells are evidently sufficiently porous at this stage that ions can readily migrate through them. The temperature in the samples increases rapidly to reach the maximum mainly due to the hydration of the tricalcium silicate and the tricalcium aluminate. The UPV propagates through the solid phase instead of through the liquid phase; this fact leads to a steep increase in UPV (Fig. 6). This critical degree of hydration, when the system changes from a suspension of cement particles in water into an interconnected solid phase, can be considered as the percolation threshold [19]. From Figure 6, we find that the stage III superposes with the Period
Two dots in Figure 6 represent the initial and final setting times measured by the standard Vicat method. Setting refers to the stiffness change of the paste. Setting corresponds to a noticeable formation of hydration products. The beginning of solidification is called the initial set. Further buildup of hydration products is followed by start of the fully rigid, responsible for the strength of concrete, which is known as the final set. Setting times are commonly defined arbitrarily by a Vicat needle. In this apparatus, weighted needles of standard design are allowed to sink into the paste, and initial and final set defined as the times when the degree of penetration falls below specified levels. However, using this method, setting times are loosely defined as the fixed penetration values that do not fully reflect the chemical and microstructural developments of the cement hydration. The skill and experience of handler also limit its accuracy and reproducibility. In this study, the resistivity and ultrasonic methods have been applied as the alternative methods for determining the setting behavior of cement-based materials. In the stages derivation here, we can use the start time of Stage III as the initial setting and the end time as the final setting.
• Stage IV: deceleration stage (after 10 hour).
As the hydration continues, the hydrates form a thicker barrier which blocks the way of solution exposure to the unhydrated cement particles. Finally, ion diffusion through the C-S-H layer determines the rate of reaction. The sample’s resistivity development becomes ion diffusion controlled and shows a larger value. A slight increase in the UPV is found after the plateau is reached. As the volume of pores is decreased by filling of hydration products, the UPV levels off and approaches its asymptotic value in the solid structure. The major connected solid frame is formed, so the increase in UPV is limited and follows the evolution of the total solid volume fraction.
4. Conclusions
The non-contacting resistivity device and the embedded piezoelectric ultrasonic measurement used in this study provide the alternative ways to accurately monitor the hydration process of cement-based materials. The two systems are effective, accurate, non-destructive and in-situ. By utilizing the critical points on resistivity curve, four stages can be identified, which are the dissolution stage, the competition stage, the acceleration stage and the deceleration stage. Both resistivity and UPV can detect the commencement of acceleration period. This point corresponds to the threshold of solid percolation.
The resistivity measurement reflects both chemical and physical change in cement paste. It is sensitive to the conductivity change of the pore solution. The early-time decrease in resistivity results is due to the ion dissolution from the cement to the water. The resistivity plateau is caused by competition of ions, and the resistivity increase with different rate results from the increasing tortuosity of the pore structures. Nevertheless, the ultrasonic results involve essentially in physical change of cement paste, the motion of the solid phase. It is sensitive to the point at which the solid phase becomes interconnected.
Acknowledgements
The financial support from Hong Kong Research Grant Council under projected HKUST6272/03E is greatly acknowledged.
Fig. 1 Schematic diagram and equivalent circle of the non-contacting electrical resistivity measurement device (CCR-II).
Fig. 2 Schematic arrangement of the piezoelectric ultrasonic measurement setup and
Cement paste container with transducers embedded
Fig. 3 Electrical resistivity development during the first (a) 1 day and (b) 7 days of hydration.
Fig. 5 Velocity development during the first 7 days of hydration.
Fig. 6 Velocity, resistivity, rate of resistivity and temperature developments during the first 7 days of hydration
