Evaluating Fresh Properties of Non-Dispersive Reactive Powder Concrete: A Novel Approach

Marine and hydraulic structures, particularly those damaged by abrasion-erosion, along with aging bridges, are a significant concern for construction experts. These structures often require underwater strengthening procedures to ensure their functional and structural safety [1-4].

The primary challenge with underwater concrete lies in its setting and curing process; the presence of water slows the chemical reactions that lead to hardening. To mitigate this challenge, construction professionals utilize special types of high early strength cements and admixtures that can set and cure rapidly, even in a water-rich environment.

One crucial property of underwater concrete is its resistance to chemical attack. Concrete can deteriorate upon exposure to various chemical agents, such as seawater, river water, and industrial effluents. Enhancing its chemical resistance involves the use of unique cements, such as sulfate-resistant Portland cement. Furthermore, pozzolanic materials like fly ash, slag, and silica fume are often included in the mix to bolster the concrete's resistance to chemical attack.

Workability and pumpability are critical aspects of underwater concrete. The presence of water can complicate handling the concrete mixture, and the pumping process can present its own set of challenges. To address these issues, admixtures like superplasticizers and air-entraining agents are employed, improving the concrete mixture's workability and pumpability.

Given the above, it's clear that cementitious materials significantly determine the properties and performance of underwater concrete. These special cements and admixtures help improve the concrete's strength, durability, and resistance to chemical attack. Therefore, the appropriate selection and use of these materials are vital for ensuring the long-term performance and durability of underwater structures.

In the quest to improve construction quality assurance and efficiency, several properties—like washout resistance, pumpability, low dispersibility, and viscosity—are primary considerations in the design of underwater concrete (UWC) [4-7]. Different UWC applications entail different fresh parameters. Recent research has improved the cohesiveness and viscosity of UWC by incorporating supplementary materials such as silica fume (SF), fly ash (FA), metakaolin, and ground granulated blast slag. SF, which boasts the highest specific surface area among other supplementary cementitious materials, has been proven to increase water resistance, densify the interfacial transition zone (ITZ), and enhance compressive strength through the pozzolanic reaction [8-10].

Conventional strengthening methods, while effective, are often time-consuming and costly [6, 7, 11-13]. Therefore, traditional strengthening and rehabilitation methods should ideally be performed without requiring dewatering and downtime. This study investigates the fresh performance of a novel type of concrete—non-dispersive reactive powder concrete (NDRPC). NDRPC is produced by combining reactive powder concrete and non-dispersive concrete. Reactive powder concrete (RPC) is composed of a precise blend of high-quality materials, including fine powders, micro silica, quartz powder, steel fibers, and chemical admixtures. These materials are carefully chosen for their unique properties and combined in specific proportions to achieve the desired performance characteristics.

Reactive Powder Concrete (RPC) is known for its high strength, with the capacity to achieve compressive strengths of up to 200 MPa (29,000 psi)—significantly higher than traditional concrete. Its durability properties are also impressive, with low permeability and high resistance to chemical and environmental degradation. Owing to its superior properties, RPC has found its way into several high-performance construction applications. These include bridge components, precast concrete elements, tunnel linings, and offshore structures. RPC is also used in creating intricate and visually pleasing architectural features, such as facades and sculptures [10, 14].

However, the inclusion of Silica Fume (SF) necessitates higher amounts of a viscosity agent, as SF reduces flowability and pumpability, consequently delaying the setting time [14-16]. Several studies have explored the effects of adding cementitious materials to Underwater Concrete (UWC). For instance, a UWC mix with a compressive strength of 435kg/m3, composed of 12% Fly Ash (FA) and 6% SF (as a replacement by cement weight) at 28 days, registered a strength of 62 MPa. This mix incorporated 0.08% welan gum by the mass of the binder material and recorded initial slumps of 250 and 275mm [17].

Other studies have experimented with SF replacements of 5%, 10%, and 15% alongside a high volume of class C FA to produce eco-friendly UWC. The primary conclusions related to the effect of SF involved a loss of workability and compressive strength as the SF percentage increased. Although this study focused on UWC, it did not record the type and doses of Anti-Wash Admixture (AWA) [18]. Another study explored the incorporation of 10% SF with two common AWAs: welan gum and a cellulose-based anti-washout. The test results reported increased washout resistance and water velocity, which led to losses in in-place strength [19].

From the literature mentioned above, it can be inferred that a UWC with adequate compressive strength and substantial washout resistance can be achieved by incorporating SF. Yet, the performance of Non-Dispersive Reactive Powder Concrete (NDRPC) remains unclear. This study aimed to determine how the gum-powder anti-wash influences the rheology of fresh concrete. Modifications to the properties of fresh concrete may enhance construction methods and concrete elements. To analyze the advantages and limitations of NDRPC, it is crucial to characterize its fresh properties.

Such a design approach is beneficial for advancing the engineering applications of UWC. Furthermore, this study sought to develop a new test to efficiently assess the non-dispersibility of a fresh Self-Compacted Mortar (SCM) or concrete mixture when placed underwater.

3.1 Flowability and washout resistance of NDRPC

Table 3 and Figure 6 present the slump flow results for a different amount of AWA made with 15% and 30% SF replacement. For constant w/b, Figure 7 shows the HRWRA needed to give a slump flow between 244-248 mm. The first impression with increment of AWA dosage, the percentage of HRWRA required increased. The addition of HRWRA was necessary at high levels of AWA to secure the target flowability, between 240-250 mm. As expected, 30% SF incorporation prevents the need for increased HRWRA; overdosages were added for 0.9%. The percentage of increased HRWRA due to increasing SF to 30% ranged between 4%-5%. This is partially attributed to the gel action of AWA, which contains long-chain polymer molecules that fix and adhere to some of the water present. In the presence of high levels of AWA, this action increased significantly [28, 29].

Moreover, the superior fineness of SF effectively enhances the cohesiveness and increases the internal friction due to flocculation action on other solid particles [30]. The levels of washout loss of the NDRPC mixtures determined by stream test are given in Table 3. As represented in Figure 8, regardless of the added AWA, the mixtures prepared with 30% SF have dramatically improved anti-dispersion. For example, dispersions of 41%, 53%, 64%, and 76% were recorded for mixtures made with 0.9%, 1.1%, 1.3%, 1.5%, and 2% AWA, respectively.

Figure 9 plots the inverse relationship between flowability and washout resistance. A strong positive correlation of 0.955 and 0.973 was established for 15% SF and 30% SF. It is interesting to note the strength effect of washout loss (WOL) on the degree of flowability. Similarly, a correlation has been established by mathematically modeling WOL and slump flow after a series of experimental results [13]. This linear relationship between slump flow and WOL% has been reported for conventional UWC previously [30], which are expressed in Eq. (2) and Eq. (3) for NDRPC mixtures made with 15% SF and 30% SF, respectively.

$S l u m p_{m m}=0.47 W O L_{\%}+238$       (2)

Slump $p_{m m}=0.41 W O L_{\%}+239$         (3)

6.png

Figure 6. Flowability behavior of NDRPC with different percentage of AWA and SF

7.png

Figure 7. Effect AWA dosages on HRWR content

8.png

Figure 8. Washout loss of NDRPC

9.png

Figure 9. Relationship between slump flow and washout resistance of NDRPC

Table 3. Flowability, filling ability passing ability and viscosity of NDRPC

3.2 pH factor and turbidity tests of NDRPC

The pH and turbidity tests are methods commonly used to investigate the washout resistance of underwater concrete [31, 32]. Figures 10 to 13 illustrate the values of pH and turbidity with time over 24 hours. A 100-ml sample of NDRPC was placed in a beaker containing 500 ml of tap water. The first reading was taken instantaneously, and the second was taken after 15 minutes. After that, readings were taken every 30 minutes until 6 hours had elapsed. The pH values were monitored every 2 hours for 24 hours, with an average of three test values used as the final result, Hanna pH meter have been used. The results show that MV, MVI, and MVII have the same pH, which indicates high washout resistance, high cohesiveness, and good stability.

In contrast, despite using a high level of AWA, the first group of NDRPC samples incorporating 15% SF had high alkalinity, ranging between 9-9.3. The second group showed a slight increase in alkalinity. This is attributed to the physicochemical effect of the extremely fine surface area of SF. This significantly enhances the action of AWA and improves the packing density of underwater concrete, at the same time as an earlier chemical effect of SF related to its reaction with calcium hydroxide to form calcium silicate hydrate [30].

A previous study has used X-ray diffraction to determine that adding high doses of powdered AWA gum can reduce the formation of calcium hydroxide and, thus, cement hydration [29]. Despite the water beaker remaining clear at a high level of AWA for approximately 1-2 hours, a pH of 8 indicates zero washouts. However, the degree of purity rapidly changed afterward for MI, MII, and MIII in the first group. These results are considered excellent compared with other studies that have investigated the addition of nano SiO₂ and nano MgO to underwater concrete, in which the pH values ranged between 11.1 and 11.9 [2].

From the above, it is worth monitoring the turbidity for the first 24 hours. A Hach 2100N turbidimeter was used to measure the turbidity of NDRPC. The pH and turbidity of the samples are shown in Figure 12. The experimental results of the turbidity test for NDRPC samples incorporating 15% and 30% SF are illustrated in Figure 13 and Figure 14, respectively. Adding AWA appears to decrease the turbidity values. In general, the results show an approximately symmetric probability distribution, which suggests the turbidity is somewhat constant because each AWA dose has an equal chance of affecting the outcome. This proves the inefficiency of instantaneous turbidity inspection for indirectly indicating the washout resistance.

Otherwise, previous studies [2, 28-30] have used this test to predict the washout resistance of underwater concrete. Samples of NDRPC incorporating 15% SF showed that 58 to 258 NTU at 1 minute became 85 to 569 NTU at 15 minutes. However, samples of NDRPC incorporating 30% SF showed an effective reduction in turbidity at 15 minutes, ranging from 25 NTU to 56 NTU. This effective influence of SF on turbidity values was continuous for 24 hours. The maximum turbidity value in the second group at 2 hours was approximately equivalent to the minimum value in the first group at 1 minute. For example, MI from GII showed 249 NTU at 2 hours while form GI showed 260 NTU at 1 minute.

10.png

Figure 10. pH values for water contains 0.2 NDRPC mixtures with 15% SF through 24 hrs

11.png

Figure 11. pH values for water contains 0.2 NDRPC mixtures with 30% SF through 24 hrs

12.png

Figure 12. Turbidimeter instrument and samples

13.png

Figure 13. Turbidity results of NDRPC incorporating with 15% SF through first 24 hrs

14.png

Figure 14. Turbidity results of NDRPC incorporating with 30% SF through first 24 hrs

3.3 Filling ability of NDRPC

Underwater concrete must easily flow out of the tremie, fill the placement area, and consolidate under its weight. Therefore, V-funnel and L-box tests have been performed for underwater concrete, and the results are given in Table 3. The filling required time limitations for SCM (using the V funnel), which were achieved at a low level of AWA ranging between 0.5% and 0.7%. In contrast, NDRPC resists the thrust of water using the force of gravity, and the surface tension starts at the tap gate, as shown in Figure 3. When V-funnel test have been conducted in conventional condition (in air) and underwater condition, the casting period takes approximately twice as long underwater as conventional tests in air. As expected, the addition of 30% SF increased the time to empty the V funnel due to the high degree of cohesiveness, as mentioned before. This is plotted in Figures 15 and 16 for mixtures from group I incorporating 15% SF and group II incorporating 30% SF. It is interesting to note that the decrease in variation in filling ability with increase SF contents. The variance in the filling ability for group I was increased from approximately 50%-116%. However, the significance reduction in variance ranged between 35% and 81% approximately.

Nevertheless, contrary to the previously stated effects of AWA and SF replacement on flowability, this was observed despite the overdosages of HRWRA used to reach the target flowability and satisfy SCM requirements. This effect is mainly related to the high internal friction and collision of grain particles in NDRPC, which increases the binding capacity and cohesiveness due to the high amount of cementitious materials [31, 33-36]. Thus, it is practical to use pumps for consolidated concrete underwater or at high levels of underground water containing 0.9%-1.5% of AWA.

15.png

Figure 15. Variation of filling ability of NDRPC with 15% SF

16.png

Figure 16. Variation of filling ability of NDRPC with 30% SF

17.png

Figure 17. Spread ability of NDRPC incorporation with 15% and 30% SF

The passing ability has been investigated through two types of barrier steel reinforcement bars and thrusts of water; this was specifically realistically simulated. Figure 17 shows the ratio of H2/H1 for each NDRPC mixture. The proper passing ability was obtained when 0.5% AWA was added to 0.95 and 0.92 for 15% SF and 30% SF, respectively. The incorporation of 30% SF resulted in a slight reduction in the passing ability of approximately 3%. However, a significant reduction in the passing ability was observed due to 30% SF replacement at high levels of AWA (7%, 9%, 19%, 13%, 12%, and 11% for MII, MIII, MIV, MV, MVI, and MVII, respectively). Moreover, H2/H1 became very small, as low as 0.4, at overdosages of AWA for MVII. This drop in H2/H1 values occurred for two reasons: first, the extra cohesiveness due to the addition of 1.3%-2% AWA and second, friction between the NDRPC and the reinforced bars. MV, MVI, and MVII were confirmed to meet the requirements for underwater casting exposed to crushing water because the lowest H2/H1 could be lower than 0.4.

3.4 Effect AWA dosage and SF content on setting time of NDRPC

A standard Vicat apparatus was used according to ASTM C953 to determine the setting time of mixtures with and without AWA. Table 4 and Figures 18 and 19 present the initial and final setting times of NDRPC samples incorporating 15% and 30% SF. Generally, adding powdered AWA caused a significant delay in the setting time. However, replacement with 30% SF produced dramatic reductions in the delay in setting time. From the experimental results, the mechanism of the chemical additives is presumed. When admixtures with high absorptivity, such as HRWRAA or a water reducer, are used, the AWA is adsorbed by SF or the fine aggregate, leaving a reduced amount to be adsorbed by the supplementary materials, resulting in a shorter delay in setting time, especially for the initial setting time [8, 34]. The synergistic effect between AWA and HRWRA from the side and SF incorporation from the other side have been observed to counterbalance the setting time delay, especially for 30% SF replacement.

18.png

Figure 18. Setting time of NDRPC incorporating with 15%SF

19.png

Figure 19. Setting time of NDRPC incorporating with 30%SF

The experimental results showed a slight delay in the initial setting time ranging from 15 minutes to approximately 3 hours. However, the retardation was increased for 15% SF replacement, and the setting time ranged from 45 minutes to approximately 4 hours. A setting time similar to the control mix was observed for MI and MII from the second set of mixtures. Generally, the initial setting time should be less than 45 min, while the final setting time should be not more than 10 hours [37]. The high concentration of HRWRA, which was required for workability enhancement, contributed substantially to promoting the setting time retardation. This could be due to the adsorbing action on cement grains or the dispersing action of HRWRA retarding the initial setting time and binder hydration, which can reduce cementitious hydration products [2]. Another reason that has been reported is that AWA molecules become adsorbed on cement grains initially in alite and particularly C3A [8].

Table 4. Setting time and washout resistance of NDRPC*

*Note: Mass before test was 2000 gm.