Abstract
The effect of curing conditions on the early age and long-term shrinkage of ultra high-performance fiber-reinforced concrete (UHPFRC) was systematically studied. The shrinkage of the early age (0–168 h) and long-term age (0–90 d) of UHPFRC material was measured based on three kinds of humidity conditions (dry, sealed, and soaked) and curing temperatures (25°C, 40°C, and 75°C), respectively. In this paper, the hydration degree of different shrinkage stages was studied in combination with chemical-bound water experiment. Meanwhile, the influencing mechanism of curing condition on the shrinkage of UHPFRC was analyzed. The results show that the early shrinkage rate of UHPFRC is accelerated with the increase of temperature, and the rate of shrinkage development at the latter stage is suppressed with the increase of temperature. With the increase of humidity, the early age shrinkage of UHPFRC and its increasing rate gradually decrease, which means drying condition > sealing condition > soaking condition. According to the long-term shrinkage results, increasing temperature has very significant inhibiting effect on the UHPFRC shrinkage in the sealed condition. Due to the majority of the in-site components of UHPFRC cured in the sealed condition, high-temperature curing has evident inhibition of early age shrinkage of UHPFRC. Therefore, promoting curing temperature is fairly effective at inhibiting the early age shrinkage of UHPFRC for the in-site structures.
1. Introduction
As a kind of typical fiber-reinforced concrete material, ultra high-performance fiber-reinforced concrete (UHPFRC) is known for the high strength, high durability, high toughness, and low defection. UHPFRC has a broad application prospect, which can be widely used in several environments [1]. In pursuit of excellent properties of UHPFRC, some methods including reducing water-cement ratio [1], adding reactive powders [2], and adding compound chemical admixtures were carried out. These design approaches will accelerate the hydration of cementitious materials to form dense structure, while the larger self-drying shrinkage is also generated simultaneously. However, the internal structure of ordinary concrete with less autogenous shrinkage is relatively loose. The shrinkage mechanism of ordinary concrete is also quite different from that of UHPFRC. Studies have shown that the autogenous shrinkage of cement hydration was only 50–100 με after 5 years [3], while the early age shrinkage of UHPFRC material reached up to 400 με at the age of 14 d [4]. Since the early strength and shrinkage of UHPFRC grow rapidly, the shrinkage stress will be generated inside concrete by constraint effect. Then the probability of cracking also increases, which causes negative effect on mechanical properties and durability of the structure. Therefore, it is of great significance to further study the shrinkage mechanism of UHPFRC.
UHPFRC is usually cured by heating and steaming. The curing condition has a very important effect on the concrete shrinkage. There is great difference for the shrinkage between the steam curing and the standard curing. Studies showed that [5] the shrinkage of UHPFRC among 14 d age reached up to 400–800 με in the standard curing condition, while that of 28 d was only about 100–300 με in the curing condition of high-temperature steam. Fehling et al. [6] believed that the concrete was denser after the heat curing, which would not produce a larger shrinkage. Bouziadi et al. [7] studied the effect of different curing temperatures (20°C, 35°C, and 50°C) on the total shrinkage of high-strength concrete; the results showed that the total shrinkage and the growth rate increased with the increase of curing temperature. Besides, the effect of curing temperature on early age shrinkage was higher than that of long-term shrinkage. Mounanga et al. [8] studied the shrinkage development in different curing temperatures (10–50°C). It was found that there was a certain limiting value for the growth of shrinkage with increasing curing temperature. The shrinkage was lower than 40 με at 50°C. Han et al. [9] studied the effect of environmental humidity and temperature on the drying shrinkage before the end of curing. The results showed that higher environment humidity during curing led to higher drying shrinkage after the curing time. While the curing temperature increases, the drying shrinkage decreased accordingly. Tu [10] studied the development of UHPFRC shrinkage in different curing conditions (75°C steam curing, 75°C dry curing, 40°C purl curing, 40°C dry curing, and 20°C purl curing). It was indicated that temperature and humidity had large effect on the shrinkage during the curing period. Raising both the temperature and the humidity increases the shrinkage during curing, but high-temperature curing also decreases the drying shrinkage after the end of curing. Above all, the curing temperature and humidity presented significant influences on the shrinkage of UHPFRC. However, in existing researches, the effects of the curing condition on the shrinkage of UHPFRC in early age and long-term shrinkage have not been clearly studied. It is difficult to propose the effective methods to restrain the UHPFRC shrinkage.
In this study, based on the shrinkage development of UHPFRC cured in standard curing room, the shrinkage development of UHPFRC cured in various curing conditions was systematically measured. Three kinds of humidity conditions (dry, sealed, and soaked) and curing temperatures (25°C, 40°C, and 75°C) were chosen. The embedded vibrating wire extensometer was used to measure the early age and long-term shrinkage. At the same time, combining with the hydration degree in different stages of shrinkage, the changing mechanisms of UHPFRC shrinkage were discussed.
2. Materials and Methods
2.1. Materials and Mix Proportions of UHPFRC
42.5 ordinary Portland cement (fineness 3400 cm3/g; the initial set 160 min, final set 220 min) is used for cementitious materials. Its clinker mineral compositions are shown in Table 1. The chemical compositions of silica fume with an average particle size of 0.31μm are shown in Table 2. The fine aggregate is made of quartz sand customized by a sand factory in Beijing. It is divided into three grades: 1.25–2.5 mm, 0.63–1.25 mm, and 0.315–0.6 mm and a ratio of 2 : 4 : 1. The measured apparent density and bulk density of quartz sand in different grades are shown in Table 3. Shortcut steel fiber with the diameter of 0.22 mm, length of 13 mm, and tensile strength of 2800 MPa is used as reinforced material. PCE with 29% water reduction rate and 31% solid content is used as superplasticizer.
UHPFRC is composed of graded quartz sand, cement, silica fume, steel fiber, superplasticizer, and water. The standard mixture ratio of UHPFRC commonly used in this research group is designed based on the packing theory [11]. As shown in Table 4, the UHPFRC standard mixture ratio was regarded as the control group. The shrinkage mechanisms of UHPFRC in different conditions were investigated by varying the curing conditions.
The detailed curing process is described as follows: Before demolding and keeping in the standard curing room (constant temperature and humidity), the experimental specimens of the sealed and soaked conditions should be covered by the membrane. At 1 d after molding, the mold was removed. The specimen of dry curing was kept in the standard curing room. The specimens of sealed curing were sealed with a plastic membrane and tape. As for soaked curing, the specimen completely submerged in water was placed in a sealed box. Moreover, before putting it in the standard curing room until the stipulated age, the experimental group of high-temperature curing was cured 72 h in the corresponding temperature (40°C and 75°C, resp.).
2.2. Test Methods
According to the Chinese current code GB/T 50082-2009 (ordinary concrete long-term performance and durability test method standard), the measured methods of concrete shrinkage are divided into the contact measuring method and the noncontact measuring method. Because the UHPFRC needs to be cured at high temperature, the shrinkage is unable to be measured by these two methods during curing according to the current code. So the method of using the embedded vibrating wire extensometer was carried out. After comparing the two kinds of methods, Han et al. [12] found that the method of using the embedded vibrating wire extensometer could accurately measures the shrinkage of UHPFRC. The shrinkage of specimens in different curing conditions was measured by using the JMZX-212-type high-precision vibrating wire extensometer and the JMBV-1164-type automatic acquisition device from molding. The reading time was set for each hour in 7 d. After this, it changed to read a number every day until the age of 90 d. The SHT75 temperature and humidity sensor embedded inside the cube specimens (100 mm × 100 mm × 100 mm) was used to measure the internal humidity with cooperating SCTH2001 intelligent data collector. The internal humidity of hardened slurry was read every other day from pouring until the age of 28 d. The early age shrinkage of UHPFRC is closely related to the hydration reaction of cementitious materials. The hydration heat method can only measure the hydration degree in first 7 d, while the chemical-bounding water test can measure that of any age. For acquiring the data of hydration degree comprehensively, the chemical-bound water contents were measured at 2, 4, 7, 14, 28, and 90 d in different curing conditions, respectively. According to [13], after stopping the hydration reaction, the samples were dried 3 h at 110°C in vacuum from about 10 g to constant weight m1, and the constant weight m2 at 950°C through the same method. The chemical-bound water content was computed through the difference of two weights.
3. Results and Discussion
3.1. The Early Age Shrinkage of UHPFRC in Different Curing Temperatures
The development of early age shrinkage is mainly determined by the hydration degree of UHPFRC. Different curing temperatures have different effects on the degree of cement hydration. In the study process of UHPFRC early age shrinkage, the UHPFRC in the three conditions of dry, sealed, and soaked were placed in the environment of 25°C, 40°C, and 75°C, respectively. After curing for 72 h, the specimens were placed into the standard curing room until the prescribed curing time. The early age shrinkage in 168 h is shown in Figure 1.
(a)
(b)
(c)
The early age shrinkage of UHPFRC occurred mainly in the first 96 hours. After that, the shrinkage curve gradually tended to be flat. Meanwhile, the shrinkage remained basically unchanged. In different humidity conditions, raising the temperature would have a great impact on the early age shrinkage of UHPFRC. At room temperature (25°C), the shrinkage of UHPFRC 168 h cured in three conditions of soaked, sealed, and dry was 194.8 με, 651.7 με, and 770.7 με, respectively. With the increase of temperature, UHPFRC early age shrinkage had growth of varying degrees. When the temperature was raised to 75°C, the corresponding 168 h shrinkage was 551.6 με, 885.5 με, and 1321.9 με, which was increased by 183.2%, 35.9%, and 71.5%, respectively. The early age shrinkage of UHPFRC would increase with the increase of curing temperature in all conditions. Faster development of early age shrinkage led to a higher shrinkage value, which increased corresponding risk of material cracking. The initial cracking of UHPFRC could occur within a short time. This was because the increase of temperature promoted the hydration rate of the cementitious materials in the concrete, and the hydration-hardening process was also enhanced. Due to the increase of cement hydration, the chemical shrinkage also increased quickly. On the contrary, hydration consumed a large amount of water in the material, which leaded to an increase in self-shrinkage inside the concrete. It was noteworthy that UHPFRC undergone some expansion reaching 200 με within 24–36 h in soaked conditions. Due to the sudden increase of air humidity in the curing environment after demolding, the hygroscopic expansion of concrete resulted in the expansion of the UHPFRC specimen.
3.2. The Long-Term Shrinkage of UHPFRC in Different Curing Temperatures
After continuous measurement of the shrinkage development, the shrinkage in 90 days was obtained in different curing humidity conditions, as shown in Figure 2. In soaked and dry curing environments, the early age shrinkage rate of UHPFRC increased rapidly until the fifth day and then remained stable in the latter period. The long-term shrinkage results showed that higher curing temperature led to smaller long-term shrinkage. Even rebounded phenomenon would occur. However, higher curing temperature led to higher shrinkage in the same curing time. The development of chemical shrinkage and autogenous shrinkage resulting in larger early age shrinkage and rapid decrease of humidity inside the concrete would be promoted by the high-temperature during curing. However, the accelerated hydration of the cementitious material produced a large amount of hydration products, which makes the hardened paste more compact. As a result, the external water hardly enters the interior of concrete after curing, which inhibited the further development of hydration. In other words, curing temperature accelerated the developing rate of UHPFRC shrinkage. Therefore, curing temperature has a significant increase in the early shrinkage. It also has an inhibitory effect on the long-term shrinkage.
(a)
(b)
(c)
As shown in Figure 2(b), UHPFRC in the sealed condition exhibited different shrinkage developments within 90 d at three different curing temperatures. The shrinkage curve before 28 d at 25°C increased with age continuously. The shrinkage at the first 7 d developed rapidly but gradually slowed down in the latter period. The curve tended to be gentle after 28 d, and the shrinkage at 90 d was 909.9 με. The shrinkage at 40°C increased rapidly in the first 5 d and then continued to increase until up to 875.5 με at 90 d. The early age shrinkage at 75°C was similar to that at 40°C, while the shrinkage decreased continuously after 5 d and decreased to 612.6 με at 90 d. As shown in the long-term shrinkage curve, the increase of curing temperature inhibited the self-shrinkage in the latter age. The inhibition was also more obvious with the increase of curing temperature. The chemical-bound water of the cement paste in three curing temperatures was measured at 1, 4, 7, 14, 28, and 90 d, respectively. Analyzing the mechanism with the hydration degree, the hydration degree was calculated as shown in Figure 3.
The shrinkage developing curve of the sealed condition showed that the effect of curing temperature on shrinkage was phased. During curing, high temperature promoted the development of shrinkage, while the inhibiting effect of high temperature on shrinkage began to emerge at the end of curing. The higher the temperature was, the more obvious the inhibiting effect was. As could be seen from the hydration degree, the highest degree of hydration occurred at 75°C and the degree of hydration at 40°C is slightly higher than 25°C. High temperature promoted the hydration reaction, which made early relative humidity decrease rapidly. The autogenous shrinkage also developed rapidly during this period. As there was no water loss in the sealed condition, the hydration of the cementitious material would continue. The hydration products filled the pore to make the structure denser, which resulted that the increase of curing temperature would inhibit the shrinkage.
3.3. The Early Age Shrinkage of UHPFRC in Different Humidity Conditions
The early age shrinkage of UHPFRC in different humidity conditions was compared. The UHPFRC cured in different humidity conditions (dry, sealed, and soaked) was placed for 72 h at the environment temperatures of 25°C, 40°C, and 75°C, respectively. The early age shrinkage in 168 h is shown in Figure 4. With the increase of moisture in the curing environment, the shrinkage of UHPFRC decreased significantly. About three hours after molding, there was a difference appearing in the curing conditions of different humidity conditions. The shrinkage of UHPFRC in the drying condition was highest, the growth rate of which was faster. However, the shrinkage of UHPFRC in the soaked condition and the sealed condition was almost the same. The water absorption expansion occurred at 24 h after curing in the soaked condition. The expansion increased obviously with the increase of curing temperature. The early age shrinkage developed rapidly during the stage of 24 to 48 h. Especially in the normal temperature condition, the moister the environment was, the slower the shrinkage developed was. In addition, after curing in the high-humidity environment, the larger shrinkage occurred in the short term after the end of curing (96 hours). Due to the thermal shrinkage and humidity shrinkage caused by the reduction of temperature and moisture, respectively, higher temperature led to higher shrinkage. Therefore, when the UHPFRC components of high-temperature steam curing were translated from the curing phase into the use stage, the risk of cracking should be focused. Higher curing temperature also led to more obvious cracking risk.
(a)
(b)
(c)
3.4. The Long-Term Shrinkage of UHPFRC in Different Humidity Conditions
After continuous measurement of the shrinkage development, the shrinkage in 90 days was obtained in different curing humidity conditions, as shown in Figure 5. There was a significant difference in the long-term shrinkage of UHPFRC material in different curing conditions. At 25°C, the shrinkage in the soaked condition was always lower than that in the other two conditions. The shrinkage in the dry condition and the sealed condition was similar to each other. At 40°C, the sort by UHPFRC shrinkage was as follows: dry condition > sealing condition > soaking condition. The correlation between this rule and environmental humidity was very good. At a high-temperature curing of 75°C, the shrinkage of UHPFRC material in the dry condition significantly increased, while shrinkage of the other two curing conditions was relatively close. The change of curing temperature caused the difference in the effect of environmental humidity on the early age shrinkage of materials. On the whole, with the increase of curing temperature, the early age shrinkage of UHPFRC in the sealed condition was closer and closer to soaked curing, which indicated that it was significantly inhibited. The inhibiting effect was only significant for the UHPFRC in the sealed condition, but not for the other two conditions. The UHPFRC material was too dense to permeate after forming, so most of the actual structural components of UHPFRC material could be regard as a sealed curing state. The high temperature had the most obvious inhibiting effect on the early age shrinkage of UHPFRC in the sealed curing condition. Therefore, as for the practical engineering structures, improving temperature is a very effective method to inhibit the early age shrinkage of material.
(a)
(b)
(c)
4. Conclusions
The early age shrinkage rate of UHPFRC in different humidity conditions is as follows: dry condition > sealing condition > soaking condition, which is consistent with ordinary concrete. The increase of curing temperature will significantly accelerate the process of UHPFRC material shrinkage, which increases the early age shrinkage. However, the rate of long-term shrinkage will be accordingly reduced. The shrinkage of UHPFRC material at the same age always increases with the promotion of curing temperature. According to the development of long-term shrinkage, increasing temperature has very significant inhibiting effect on the UHPFRC shrinkage in the sealed condition, which is much higher than the other two conditions. Due to the majority of the engineering components of UHPFRC cured in the sealed condition, high-temperature curing has evident inhibition on early age shrinkage of them. Therefore, as for the practical engineering structures, improving temperature is a very effective method to inhibit the early age shrinkage of UHPFRC.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This research was supported by the National Natural Science Foundation of China (51408037 and 51578049), Major Scientific and Technological Projects of the Communications Ministry (2011318494160), and China Communications Construction R&D Project (2013-ZJKJ-11).