Effects of admixtures on the rheological properties of high-performance wet-mix shotcrete mixtures

Construction and Building Materials 78 (2015) 194–202

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Effects of admixtures on the rheological properties of high-performance wet-mix shotcrete mixtures Kyong-Ku Yun a, Sung-Yong Choi a, Jung Heum Yeon b,⇑ a b

Department of Civil Engineering, Kangwon National University, Chuncheon 200-701, South Korea Department of Civil and Environmental Engineering, Gachon University, Seongnam 461-701, South Korea

h i g h l i g h t s  Effects of various admixtures on the rheological properties of HPWMS were examined.  Rheological properties such as yield stress and plastic viscosity were measured using IBB rheometer.  The measured rheological properties were correlated to shootability and pumpability.  Silica fume is the most effective admixture for enhancing the rheological properties of HPWMS.

a r t i c l e

i n f o

Article history: Received 4 July 2014 Received in revised form 25 November 2014 Accepted 31 December 2014 Available online 16 January 2015 Keywords: High-performance wet-mix shotcrete Rheology Admixtures Pumpability Shootability

a b s t r a c t This study investigates the effects of various admixtures on the rheological properties of high-performance wet-mix shotcrete (HPWMS) in an attempt to resolve practical issues faced in conventional wet-mix shotcrete processing. The admixtures used in this study were silica fume, air-entraining agent (AEA), superplasticizer, synthetic fiber, powdered polymer, and a viscosity agent. Representative rheological properties such as yield stress and plastic viscosity were measured using an IBB rheometer to evaluate the pumpability and shootability of HPWMS with varying admixture types and contents. The results demonstrated that the use of AEA tended to reduce both flow resistance and torque viscosity of HPWMS almost proportionally. A superplasticizer had a relatively greater impact on the flow resistance rather than torque viscosity. Also, it was observed that silica fume led to a remarkable increase in flow resistance while it slightly reduced torque viscosity. This behavior trend indicates that silica fume is quite effective in enhancing the rheological properties of HPWMS, particularly in terms of shootability and pumpability. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction

s ¼ s0 þ lc_

Fresh concrete exhibits an elastic behavior at a low-stress range, and it begins to show a plastic flow once a certain level of deviator stress is reached. Because of such nature, from a rheological perspective, fresh concrete is classified as a Bingham fluid rather than a Newtonian fluid [1–3]. It is well known that, in fresh concrete, the shear stress is not in perfect linear proportion to the shear strain rate. However, if assumed that their correlation is approximately linear, the Bingham model can be an effective tool to characterize the flow behavior of fresh concrete. Applications of the Bingham model require at least two parameters, yield stress and plastic viscosity, as indicated in Eq. (1):

where s is the shear stress; s0 is the shear yield stress; l is the plastic viscosity; and c_ is the shear strain rate. The Bingham model has been primarily applied to high-concentration viscous fluids such as fresh cement paste. However, over the years, this well-defined model has been accepted to fresh concrete containing various-sized aggregates as the development of cutting-edge test apparatuses, such as rotational and oscillatory rheometers [4–7], allowed easy measurements of yield stress and plastic viscosity of fresh concrete. In particular, the Bingham model yields a strong fitness when applied to fresh concrete with a high slump (over 15 cm) and consistency because, in such highly workable concrete, the flow characteristics largely depend on the properties of cement paste matrices [8]. Previous research on shotcrete often focused on wet-mix types; a number of studies attempted to document the strength characteris-

⇑ Corresponding author. Tel.: +82 31 750 5498. E-mail address: [email protected] (J.H. Yeon). http://dx.doi.org/10.1016/j.conbuildmat.2014.12.117 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.

ð1Þ

195

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

Planetary gears: 16 DP and 45 DP

Bowl = 360 mm in diameter x 250 mm in height Concrete level = 200 mm

100 mm

130 mm

(a)

(b)

Fig. 1. IBB rheometer: (a) front view; (b) details of H-shape impeller, bowl, and planetary gear.

Table 1 Experimental variables and target levels.

Table 3 Physical properties of polymer.

Variables

Target levels

w/cm () AEA (% of cement content) Superplasticizer (% of cement content) Silica fume (% of cement content) Polymer (% of cement content) Synthetic fiber (% of cement content) Viscosity agent (% of cement content)

0.50, 0.55, 0.60 0, 0.01, 0.02, 0.05 0, 0.1, 0.2 0, 9 0, 4 0, 0.2 0, 0.3

Table 2 Chemical compositions of cement (unit: %). SiO2

Al2O3

Fe2O3

CaO

MgO

SO3

20.8

6.3

3.2

61.2

3.3

2.3

Percentage passed by weight [%]

100

80

60

40 Fine agg. limits Fine agg. gradation

20

Coarse agg. limits Coarse agg. gradation

0 0

1

10

100

Screen size [mm] Fig. 2. Aggregate gradations with recommended limits.

tics and rebound rate of fiber-reinforced wet-mix shotcrete [9,10]. Also, over the years, there has been an increasing interest in highstrength and high-durability shotcrete [11–15]. Only few studies are available that address pumpability and shootability, as well as other rheological properties of high-performance shotcrete. Beaupre [7] studied the rheological properties of high-performance shotcrete using a self-developed UBC rheometer, in which pumpability and shootability of fresh shotcrete were evaluated. Another study

Bulk Ash Min. film Solid content content density forming (kg/m3) temp. (°C) (%) (%)

Particle Appearance Protective size colloid/ (lm) emulsifier system

98–100 9–13

400+ (max. 4%)

490– 590

4

White powder

Polyvinyl alcohol

[8] has confirmed that flow characteristics of fresh concrete largely depend on the rheological properties of cement paste matrices. Moreover, Ko et al. [16] evaluated how the binder type, water-tobinder ratio, admixture content, and time of testing initiation affect the mechanical and rheological properties of mortar to provide some basic information on high-fluidity concrete. Szecsy [17] found the effects of fine aggregate fraction, water-to-cement ratio, coarse aggregate type, fly ash replacement, and superplasticizer addition on the rheological properties of fresh concrete. Kang et al. [18] developed a theoretical mixture design for high-fluidity concrete by assessing the influences of microfine and mineral admixtures on the rheological properties of cement paste. Other former research studies incorporate modeling of pump pressure considering the rheological properties and frictional conditions [19] and experimental investigations on the effects of silica fume [20] and accelerators [21]. The key components affecting the rheological properties of fresh concrete include: (1) binder formulations including the type and content of chemical and mineral admixtures; (2) type, shape, and gradation of aggregates; (3) water-to-cementitious ratio (w/ cm); and (4) properties of cementitious materials. Even similar mixtures could display quite different flow characteristics with slight variations in those components. In this study, the effects of various types of additions (i.e., silica fume, air-entraining agent (AEA), superplasticizer, synthetic fiber, powdered polymer, and viscosity agent) on the rheological properties of high-performance wet-mix shotcrete (HPWMS) are comprehensively evaluated. Also, the identified rheological properties were correlated to the practical indicators of shotcrete behavior, such as shootability and pumpability. The outcomes of this study are expected to provide valuable information to resolve practical issues faced in conventional wet-mix shotcrete processing.

196

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

Table 4 Physical properties of superplasticizer. Active component

pH (5% solution)

Bulk density (kg/m3)

Appearance

Equivalent Na2O (%)

Chloride content (%)

Polycarboxylate

6.5 ± 1.0

370

Brownish powder

6

0.05

Table 5 Physical properties of AEA. Active component

pH (1 wt.% solution)

Bulk density (kg/m3)

Appearance

Moisture content (%)

Sulfonate silica

7.0

900–1200

Light ivory powder

<2.0

2. Experimental program 2.1. Test apparatus There have been continued modifications and advances in test apparatuses to accurately measure the rheological properties of fresh concrete. Representative types of concrete rheometers include coaxial cylinder, MK II/III rheometers (or Tattersall rheometer), IBB rheometer (or UBC rheometer), BML viscometer, and LCPC rheometer. In this study, the IBB rheometer, which was first devised by The University of British Columbia and later modified by IBB Rheology Inc., was employed (see Fig. 1). The IBB rheometer allows direct measurements of flow resistance and torque viscosity from mixture samples.

2.2. Experimental variables A total of seven experimental variables were established to comprehensively investigate the rheological properties of HPWMS. Table 1 summarizes the variables considered in the experimental program, all of which are recognized to have substantial influences on the flow characteristics of fresh concrete. In addition, the table contains selected levels of each variable at which the rheological properties were measured.

2.3. Materials 2.3.1. Cement This study included normal Portland cement (Type I) with a fineness of 3200 cm2/g and a specific gravity of 3.15. The chemical composition of the Portland cement used is presented in Table 2.

2.3.2. Aggregates Sound aggregates meeting the physical property requirements in the Korean Industrial Standards (KS) were collected and used for the experiments. For coarse aggregates, washed crushed rock with a maximum size of 10 mm was employed. Washed river sand was used as fine aggregate. The specific gravities of the coarse and fine aggregates were 2.65 and 2.57, respectively. The fineness moduli of the coarse and fine aggregates were found to be 5.70 and 2.66, respectively. The gradation curves for both coarse and fine aggregates used are depicted in Fig. 2, along with the gradation limits recommended by ASTM C33.

2.3.3. Admixtures 2.3.3.1. Silica fume. The silica fume employed in this study had a specific surface area of 150,000–300,000 cm2/g and a specific gravity of 2.22. The silica fume was comprised of up to 97% SiO2 with less than 1% CaO.

2.3.3.2. Powdered polymer. Powdered polymers were prepared by polymerizing vinyl acetate and ethylene. The physical properties of the polymers are tabulated in Table 3.

2.3.3.4. Chemical admixtures. The present study included various powder-type chemical admixtures such as superplasticizer, AEA, and viscosity agent. Tables 4– 6 display the physical properties of each chemical admixture employed, respectively. 2.3.4. Mixture proportions and mechanical properties The HPWMS mixtures tested in this study had a fixed cementitious content of 440 kg/m3, maximum aggregate size of 10 mm, and fine aggregate fraction of 70%, while the dosages of the admixtures varied as required—which generates a slump of 120 ± 30 mm. The selected levels of w/cm were 0.50, 0.55, and 0.60, at which wet-mix shotcrete could be produced without any mineral and chemical admixtures. Table 7 shows the mixture proportions of HPWMS used in this study. For the mixtures before shotcreting, the compressive strength varied between 16.9 (SF9A-FP) and 39.6 MPa (SF9), while the flexural strength varied between 3.3 (SF0A) and 6.2 MPa (SF9). For the mixtures after shotcreting, the compressive strength varied between 39.9 (SF9A-FPV) and 57.6 MPa (SF9A), whereas the flexural strength varied between 5.4 (SF0A) and 8.0 MPa (SF9A-FPV). 2.4. Experimental methods 2.4.1. Air content test An air content, which is known to have significant effects on the strength and durability of hardened concrete, was measured as per the method outlined in KS F 2421 (method of test for air content of fresh concrete by pressure method). 2.4.2. Rheology test This study used the IBB rheometer equipped with an H-shape impeller. As the impeller stirs mixtures with a specified rotational speed, the torque exerted on the impeller is measured, assessing both yield stress and plastic viscosity. Typical rheological parameters such as torque viscosity and flow resistance were computed from the obtained yield stress and plastic viscosity based on linear regression analysis, presenting the speed of rotation in the x-axis and the exerted torque in the yaxis. An example of the result analysis is shown in Fig. 3. In this plot, an inverse of the slope is defined as the torque viscosity (H in Fig. 3) while the x-intercept is defined as the flow resistance (G in Fig. 3). The obtained data set can be generalized by the following form of the Bingham model:

T ¼ G þ HN

ð2Þ

in which, T is the torque exerted on the impeller [N m]; G is the flow resistance [N m]; H is the torque viscosity [N m s]; and N is the angular speed of the impeller [rev/s].

3. Results and discussion 3.1. Effect of w/cm Fig. 4 presents the effect of w/cm on the flow resistance and torque viscosity of HPWMS. The result shows that as the w/cm increased from 0.50 to 0.60, the slope of the fitting lines tended to gradually increase. Also observed is that the fitting lines shifted to the left side of the given chart with the increased w/cm, thereby decreasing the flow resistance. This behavior trend well agreed with the published data from Tattersall [22], which measured the rheological parameters using the MK II rheometer (a former version of the IBB rheometer). Particularly, the flow resistance value obtained in this study was found to be quite similar to Tattersall’s results when the w/cm was 0.55 (2.62 N m in this study and 2.40 N m in Tattersall’s work). It should be also noted that as the w/cm increased from 0.50 to 0.60, both flow resistance and torque viscosity decreased. From a practical standpoint, higher w/cm may be beneficial for improving the pumpability of HPWMS because the higher the w/ cm, the lower the torque viscosity. However, when it comes to the shootability, higher w/cm may result in poor performance since the reduced flow resistance could allow HPWMS to flow down immediately after shotcreting operation. 3.2. Effect of AEA (air content)

2.3.3.3. Synthetic fiber. 100% nylon synthetic fibers with a specific gravity of 1.16 were adopted as a reinforcing material. The filament diameter and fiber length were 23 microns and 12 mm, respectively. The melting point was approximately 260 °C. The tensile strength, elastic modulus, and toughness were 890 MPa, 5.1 GPa, and 107 MPa, respectively.

Fig. 5(a) shows the correlation between the exerted torque and rotational speed for different AEA contents (or air contents). Fig. 5(b) illustrates the variations in flow resistance and torque

197

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202 Table 6 Physical properties of viscosity agent. Active component

Hydroxypropyl content (wt.%)

Methoxyl (wt.%)

Ash content (wt.%)

Moisture content (%)

Viscosity of 2 wt.% solution (mPa s)

Bulk density (kg/m3)

Hydroxypropyl methyl cellulose

7.0–12.0

28.0–30.0

<1.0

<5.0

400

0.25–0.7

Table 7 Mixture proportions of HPWMS. Variable

Mixture ID

w/cm ()

S/a (%)

Water (kg/ m3)

Ca (kg/ m3)

FAb (kg/ m3)

CAc (kg/ m3)

SFd (kg/ m3)

AEA (kg/ m3)

SPe (kg/ m3)

Fiber (kg/ m3)

Polymer (kg/ m3)

VAf (kg/ m3)

w/cm

WC50 WC55 WC60

0.50 0.55 0.60

70 70 70

220 242 264

440 440 440

1092 1052 1011

473 456 438

– – –

– – –

– – –

– – –

– – –

– – –

AEA

AE0 AE0.01

0.50 0.50

70 70

220 220

440 440

1092 1092

473 473

– –

– –

– –

– –

– –

AE0.02

0.50

70

220

440

1092

473

–

–

–

–

–

AE0.05

0.50

70

220

440

1092

473

–

– 0.044 (0.01%) 0.088 (0.02%) 0.22 (0.05%)

–

–

–

–

SP0 SP0.1

0.50 0.50

70 70

220 220

440 440

1092 1092

473 473

– –

– –

– –

– –

– –

SP0.2

0.50

70

220

440

1092

473

–

–

– 0.44 (0.1%) 0.88 (0.2%)

–

–

–

SFd

SF0 SF9

0.55 0.55

70 70

242 242

440 400

1052 1018

456 441

– 39.6 (9%)

– –

– –

– –

– –

– –

Polymer

Polymer0 Polymer4

0.55 0.55

70 70

242 242

440 440

1052 1035

456 449

– –

– –

– –

– –

– 17.6 (4%)

– –

Fiber

Fiber0 Fiber0.2

0.55 0.55

70 70

242 242

440 440

1052 1050

456 455

– –

– –

– –

– –

– –

Fiber0.4

0.55

70

242

440

1050

454

–

–

–

– 0.88 (0.2%) 1.76 (0.4%)

–

–

Visco0 Visco0.3

0.55 0.55

70 70

242 242

440 440

1052 1050

456 455

– –

– –

– –

– –

– –

Visco0.6

0.55

70

242

440

1050

455

–

–

–

–

–

– 1.32 (0.3%) 2.64 (0.6%)

SPe

VAf

a b c d e f

Cement. Fine aggregate. Coarse aggregate. Silica fume. Superplasticizer. Viscosity agent.

viscosity estimated based on Fig. 5(a). Note that as the AEA content (or air content) increased, the torque viscosity significantly decreased. The reduction rate of torque viscosity continued to decrease with the increased air content. In contrast, the flow resistance exhibited a sudden fluctuation around the air content of 18%, which appears to result from an experimental error. The result of this study showed a good agreement with that of the previous work [23], where both flow resistance and torque viscosity continued to decrease until the air content reached 5% and then stabilized beyond that point. The findings demonstrate that AEA effectively improves the pumpability of HPWMS as it reduces both torque viscosity and flow resistance dramatically. As mentioned previously, the lower flow resistance may be harmful to the shootability; for air-entrained shotcrete, however, this should not cause critical problems because air-entrained shotcrete typically loses some entrained air during shotcreting operations, thereby regaining the shootability. 3.3. Effect of superplasticizer Fig. 6(a) and (b) present the relationship between the exerted torque and rotational speed for different superplasticizer contents

and the corresponding variations in flow resistance and torque viscosity, respectively. Observed was that as the dosage of superplasticizer increased from 0% to 0.2%, the flow resistance was considerably reduced. The reduction in the rate of flow resistance tended to decrease as the dosage increased, which was very similar to the data published by Banfill [24]; however, the torque viscosity tended to decrease with the increased superplasticizer dosage, and this was contrary to the result of the former study. This discrepancy appears to be mainly attributed to the type and compositions of superplasticizer used and the w/cm of mixtures tested; the w/ cms examined in the former study were 0.65 and 0.73, which are far higher than those used in this study.

3.4. Effect of silica fume The effect of silica fume on the rheological properties of HPWMS is plotted in Fig. 7. It is interesting to note that, upon addition of 9% silica fume, the flow resistance considerably increased whereas the torque viscosity rather slightly decreased. The results shown are practically and technically quite important in that silica fume aids the ease of pumping by reducing the torque viscosity,

198

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

1.5

Speed of rotation [rev/s]

Speed of rotation (rev/s)

1.0 y= M x+ B y= 0.12x-0.75

0.8

R 2= 0.96 M = 1/ H

0.6

1

0.4 H

0.2

G =B /M

y = 0.65 x - 2.15 R2 = 0.98

y = 0.55 x - 1.25 R2 = 1.00

1.2

y = 0.27 x - 1.84 R2 = 1.00

0.9

0.6

AE = 0% y = 0.81 x - 2.47 R2 = 0.99

AE = 0.01%

0.3

AE = 0.02% AE = 0.05%

0.0

0.0 4.0

8.0

12.0

16.0

0

Torque (N·m)

3

6

Flow resistance [N·m]

Speed of rotation [rev/s]

1.5 y = 0.37 x - 0.42 R2 = 0.97 y = 0.27 x - 1.96 R2 = 0.96

y = 0.31 x - 0.82 R2 = 0.98

0.9

0.6

15

10

5

8

4

6

3

4

2

2

0.3

1

Flow resistance Torque viscosity

w/cm = 0.50

0

w/cm = 0.55

0 0

w/cm = 0.60

5

10

15

20

25

Air content [%]

0.0 0

3

6

9

12

(b)

15

Torque [N·m]

(a)

Fig. 5. Effect of AEA (air content): (a) torque vs. speed of rotation; (b) variations of G and H.

10

5

8

4

6

3

4

2

2

1

Flow resistance Torque viscosity

0 0.45

0.50

0.55

0.60

Torque viscosity [N·m·s]

Flow resistance [N·m]

12

(a)

Fig. 3. Example of test result from IBB rheometer (G = 6.25; H = 8.33).

1.2

9

Torque [N·m]

Torque viscosity [N·m·s]

0.0

while the torque viscosity slightly increased. This is most likely because polymers play a role of AEA (which gives rise to surfactant actions), allowing fresh concrete to keep plenty of entrained air bubbles. Again, the reduced flow resistance is unfavorable for shootability, and in turn, for shotcreting thickness. However, because some entrained air would be eliminated upon shotcreting operations, it is expected that the flow resistance would be recovered, and thus polymers would positively affect the shootability of HPWMS.

0 0.65

w/cm [-]

(b) Fig. 4. Effect of w/cm: (a) torque vs. speed of rotation; (b) variations of G and H.

while even providing enhanced shootability by maintaining an appropriate level of flow resistance. Accordingly, silica fume is found to be a quite effective admixture that increases shotcreting thickness and minimizes rebound rate. The result achieved in this paper is comparable to that in the previous work by Gjørv [25], which reported that the flow resistance remains nearly constant until the silica fume replacement reaches 7%, and then it considerably increases as the replacement becomes greater than 7%, and by Beaupre [7], who observed that incorporating silica fume by up to 10–15% has a positive effect on both pumpability and shootability of wet-mix shotcrete. 3.5. Effect of polymer Fig. 8 shows that, for the 0.55 w/cm HPWMS mixtures, the flow resistance substantially decreased when 4% polymer was used,

3.6. Effect of synthetic fiber Fig. 9(a) illustrates the torque versus rotational speed relationship for different synthetic fiber contents. The associated changes in flow resistance and torque viscosity are shown in Fig. 9(b). The results clearly present that the flow resistance rapidly increased as the content of synthetic fiber increased, which well coincided with the finding confirmed by Tattersall [22]. As for the torque viscosity, it tended to slightly increase for the first 0.2% addition, and then decreased by about 40% for additional 0.2%, which is different from the result of Tattersall’s study. On the other hand, another study by Llewellyn [26] has shown that both flow resistance and torque viscosity increased when 0.1% polypropylene fiber was employed. 3.7. Effect of viscosity agent Fig. 10 illustrates the variations in the rheological parameters when the content of viscosity agent changed between 0% and 0.6%. Note that the flow resistance was reduced by approximately 50% when the content of a viscosity agent was 0.3%. For an addi-

199

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

1.5 y = 0.33 x - 0.72 R2 = 0.98

y = 0.39 x - 0.77 R2 = 0.98

1.2

y = 0.27 x - 1.84 R2 = 1.00

0.9

0.6 SP = 0%

0.3

SP = 0.1%

Speed of rotation [rev/s]

y = 0.31 x - 0.82 R2 = 0.98

1.2

y = 0.33 x - 2.10 R2 = 0.98

0.9

0.6

0.3

SF = 0% SF = 9%

SP = 0.2%

0.0

0.0 0

3

6

9

12

0

15

3

6

Torque [N·m]

(a) 5

15

4

6

3

4

2 1

Flow resistance

5 Flow resistance

Flow resistance [N·m]

8

10

Torque viscosity [N·m·s]

Flow resistance [N·m]

12

(a)

10

2

9

Torque [N·m]

8

Torque viscosity

4

6

3

4

2

2

1

Torque viscosity

0

0 0

0.1

0.2

Superplasticizer [%]

(b) Fig. 6. Effect of superplasticizer: (a) torque vs. speed of rotation; (b) variations of G and H.

tional 0.3%, the flow resistance recovered by as much as 80% of the initial value. On the other hand, the torque viscosity continued to increase upon addition of a viscosity agent. The obtained data confirms that a viscosity agent adversely affects the rheological characteristics of HPWMS (reducing the flow resistance and increasing the torque viscosity), which worsens both shootability and pumpability.

3.8. Results discussion Beaupre [7] reported that both flow resistance and torque viscosity should be kept lower to ensure better pumpability and at the same time flow resistance should be kept higher for better shootability. These conflicting requirements imply that an optimum mixture design for wet-mix shotcrete can be achieved when there is a good balance between flow resistance and torque viscosity. Fig. 11 compares the relationship between the torque viscosity and flow resistance for different levels of w/cm, AEA, and superplasticizer. As can be seen in the results, AEA tended to reduce the flow resistance and torque viscosity in a more balanced manner compared to a superplasticizer. This result demonstrates that the use of AEA is one of the most practical measures to effectively improve the pumpability of HPWMS. However, because the excessive use of AEA could result in strength reduction, the content of AEA needs to be controlled with caution when incorporated into HPWMS. Time-dependent changes in the rheological properties of HPWMS with 0.2% superplasticizer are displayed in Fig. 12. The fitting line shifted to the right side along the x-axis as time elapsed (after 10 min), indicating that the flow resistance increased with time when a plasticizer was added. From a practical point of view,

0

Torque viscosity [N·m·s]

Speed of rotation [rev/s]

1.5

0 0

9

Silica fume [%]

(b) Fig. 7. Effect of silica fume: (a) torque vs. speed of rotation; (b) variations of G and H.

this behavior is highly unfavorable since the mixtures could suffer a significant loss of pumpability even prior to initiating a shotcreting operation. Therefore, it can be concluded that, compared with a superplasticizer, AEA provides a more effective way to promote the pumpability of HPWMS while keeping an adequate level of shootability. As discussed previously, polymers also generate a comparable effect with AEA, but they tended to slightly increase the torque viscosity. Thus, adequate control of the polymer content is essential to avoid a pumpability reduction when polymer is selected as an admixture. From a shootability aspect, it is highly recommended to choose admixtures that increase the flow resistance. On the basis of the results obtained in this study, silica fume and synthetic fiber could be considered as feasible candidates because both of them enhance the flow resistance, and in turn, the shotcreting thickness. However, it should be noted that, whereas silica fume decreased the torque viscosity to some degree, synthetic fiber had a tendency to somewhat increase the torque viscosity when the fiber content increased up to 0.2%. Using synthetic fiber less than 0.2% therefore is unfavorable for pumpability, and thus silica fume can be more effectively used for HPWMS. Fig. 13 shows the results of IBB rheometer tests, clearly highlighting the influence of AEA. Note that, when AEA was used, the linear fitting lines slightly moved to the left along the x-axis and the slope of the fitting lines appreciably increased. The variations in the torque versus rotational speed relationship for the HPWMS mixtures incorporating 9% silica fume and 0.05% AEA are shown in Fig. 14. The results reveal that the fitting line shifted to the right when some fiber was added. Also observed was that the slope of the regression line substantially decreased when polymer and viscosity agent were simultaneously added.

200

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

0.9

0.6 Polymer = 0%

0.3

Polymer = 4%

3

6

9

12

y = 0.27 x - 0.33 R2 = 1.00

0.9

y = 0.21 x - 0.42 R2 = 1.00

0.6 VA = 0%

0.3

VA = 0.3% VA = 0.6%

0.0 0

1.2

0.0

15

0

Torque [N·m]

3

6

(a) 10

5

6

3

4

2

2

1

Flow resistance [N·m]

4

Torque viscosity [N·m·s]

Flow resistance [N·m]

Torque viscosity

0

5 Torque viscosity

8

4

6

3

4

2

2

1

0

0 0

0 0.0

4

0.3

0.6

Viscosity agent [%]

Polymers [%]

(b)

(b) Fig. 8. Effect of polymer: (a) torque vs. speed of rotation; (b) variations of G and H.

Fig. 10. Effect of viscosity agent: (a) torque vs. speed of rotation; (b) variations of G and H.

y = 0.29 x - 1.46 R2 = 0.98

y = 0.31 x - 0.81 R2 = 0.98

8 w/cm

1.2 y = 0.50 x - 4.91 R2 = 0.91

0.9

0.6 Fiber = 0%

0.3

Fiber = 0.2% Fiber = 0.4%

Flow resistance [N·m]

Speed of rotation [rev/s]

15

Flow resistance

Flow resistance

1.5

12

(a)

10 8

9

Torque [N·m]

Torque viscosity [N·m·s]

1.2

y = 0.31 x - 0.81 R2 = 0.98

1.5

y = 0.31 x - 0.82 R2 = 0.98

y = 0.30 x + 0.05 R2 = 0.98

Speed of rotation [rev/s]

Speed of rotation [rev/s]

1.5

AE agent

6

Superplasticizer

4

2

0.0 0

3

6

9

12

15

0

Torque [N·m]

1

2

5

8

4

6

3

4

2

Flow resistance

1

Torque viscosity

0

0 0.0

0.2

4

Fig. 11. Effects of w/cm, AEA, and superplasticizer on G and H.

1.5 SP = 0.2%

Speed of rotation [rev/s]

10

2

3

Torque viscosity [N·m·s]

Torque viscosity [N·m·s]

Flow resistance [N·m]

(a)

SP = 0.2% (after 10 min)

1.2

0.9 y = 0.39 x - 0.77 R2 = 0.98

0.6

y = 0.37 x - 1.08 R2 = 0.98

0.3

0.4

Fibers [%]

(b) Fig. 9. Effect of synthetic fiber: (a) torque vs. speed of rotation; (b) variations of G and H.

0.0 0

2

4

6

Torque [N·m] Fig. 12. Effect of age on the rheological properties.

8

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

y = 0.34 x - 1.09 2 R = 1.00

Speed of rotation [rev/s]

1.5

y = 0.20 x - 0.79 R2 = 1.00

y = 0.23 x - 0.85 R2 = 1.00

1.2 y = 0.39 x - 1.26 R2 = 1.00

0.9

y = 0.17 x - 0.55 R2 = 1.00

0.6 y = 0.46 x - 1.33 R2 = 1.00

0.3

SF0

SF4.5

SF9

SF0A

SF4.5A

SF9A

8

10

0.0 0

2

4

6

12

Torque [N·m]

201

considerably decreased upon addition of another 0.2%. This result indicates that addition of synthetic fiber could be one of the alternatives for improving the rheological performance of HPWMS. (e) A viscosity agent had a tendency to increase the torque viscosity while reducing the flow resistance of HPWMS. This behavior is quite unfavorable for both shootability and pumpability. Although silica fume and AEA had positive influences on the shootability and pumpability of HPWMS, if excessive rebound occurs during shotcreting operations, it could be beneficial to add some viscosity agent, even if it could lead to an associated partial loss of pumpability.

Fig. 13. Results of rheometer tests (SF0, SF4.5, SF9, SF0A, SF4.5A, and SF9A).

Acknowledgements

y = 0.35 x - 1.48 R2 = 0.99

Speed of rotation [rev/s]

1.5

1.2

y = 0.27 x - 0.89 R2 = 1.00 y = 0.39 x - 1.27 R2 = 1.00

0.9

y = 0.15 x - 0.68 2 R = 1.00

This research was supported by the grant (13RDRP-B066780) from the Regional Development Research Program funded by the Ministry of Land, Infrastructure and Transport of Korean government and was performed using the facilities of the Institute for Advanced Construction Materials at Kangwon National University, Chuncheon, Korea. References

0.6

SF9A SF9A-F

0.3

SF9A-FP SF9A-FPV

0.0 0

3

6

9

12

15

Torque [N·m] Fig. 14. Results of rheometer tests (SF9A, SF9A-F, SF9A-FP, and SF9A-FPV).

4. Conclusions and recommendations The primary goal of this paper was to shed some light on the effects of various admixtures on the rheological properties of high-performance wet-mix shotcrete (HPWMS). A special emphasis was placed on correlating the identified rheological properties to the practical indicators of shotcrete application, such as shootability and pumpability. Based on the experimental findings obtained in this study, the following conclusions can be reached: (a) Upon addition of air-entraining agent (AEA), both torque viscosity and flow resistance of HPWMS tended to decrease in a balanced manner. On the other hand, a superplasticizer had a more pronounced effect on the flow resistance rather than torque viscosity. This result demonstrates that AEA is more effective in enhancing the pumpability of HPWMS than a superplasticizer. (b) Silica fume remarkably promoted the flow resistance of HPWMS, while it slightly reduced the torque viscosity. It is evident from this result that the use of silica fume is one of the most effective measures to improve both shootability and pumpability. (c) The flow resistance of HPWMS substantially decreased upon addition of polymers; in contrast, the torque viscosity somewhat increased. This behavior trend shows that the use of polymers may be unfavorable for pumpability of HPWMS. (d) As the content of synthetic fiber increased, the flow resistance of HPWMS rapidly increased. The torque viscosity tended to slightly increase for the first 0.2% addition but

[1] Ferraris CF, de Larrard F. Testing and modelling of fresh concrete rheology, report No. NISTIR 6094, National Institute of Standards and Technology: Gaithersburg; 1998. [2] Tattersall GH, Banfill PFG. The rheology of fresh concrete. Boston: Pitman Advanced Publishing Program; 1983. [3] Koehler EP, Fowler DW. Development of a portable rheometer for fresh Portland cement concrete, Research report ICAR-105-3F, International Center for Aggregates Research, The University of Texas Austin; 2004. [4] Van Wazer JR, Lyons JW, Kim KY, Colwell RE. Viscosity and flow measurement: a handbook of rheology. New York: Interscience; 1963. [5] Barnes HA, Hutton JF, Walters K. An introduction to rheology. Amsterdam: Elsevier Science Publishers; 1989. [6] Wallevik OH, Gjørv OE, Development of a coaxial cylinder viscometer for fresh concrete, Properties of fresh concrete. In: Proceedings of the RILEM Colloquium, Hanover, Germany, 1990. [7] Beaupre D. Rheology of high performance shotcrete [Ph.D. dissertation]. University of British Columbia: Canada; 1994. [8] Ryu H-J, Choi Y-J, Kim J-H, Ji N-Y, Kim W-J. An experimental study on the rheology properties of cement paste using fly ash. J Korea Concr Inst 1997;9(2):241–6. [9] Leung CKY, Lai R, Lee AYF. Properties of wet-mixed fiber reinforced shotcrete and fiber reinforced concrete with similar composition. Cem Concr Res 2005;35(4):788–95. [10] Saw H, Villaescusa E, Windsor CR, Thompson AG. Laboratory testing of steel fibre reinforced shotcrete. Int J Rock Mech Min Sci 2013;57:167–71. [11] Ishida A, Shotcrete with high strength at early age. In: 10th international conference on shotcrete for underground support, Whistler, Canada; 2006. [12] Won JP, Hwang U-J, Kim C-K, Lee S-J. Mechanical performance of shotcrete made with a high-strength cement-based mineral accelerator. Constr Build Mater 2013;49:175–83. [13] Lee S, Kim D, Ryu J, Lee S, Kim J, Kim H, et al. An experimental study on the durability of high performance shotcrete for permanent tunnel support. Tunnelling Underground Space Technol 2006;21:431–7. [14] Chen J, Zhao X, Luo Y, Deng X, Liu Q. Investigating freeze-proof durability of C25 shotcrete. Constr Build Mater 2014;61:33–40. [15] Aïtcin PC. The durability characteristics of high performance concrete: a review. Cement Concr Compos 2003;25:409–20. [16] Ko E-H, Kim C-J, Hong Y-T, Oh S-G. The rheology and strength property of highfluidity mortar using fly ash as admixture. J Reg Assoc Archit Inst Korea 2006;2(1):405–10. [17] Szecsy RS. Concrete rheology [Ph.D. dissertation]. University of Illinois at Urbana-Champaign: Champaign, USA; 1997. [18] Kang B, Park M, Kim H, An J. The rheological properties of high flowability paste using cellulose type high viscosity agent. J Reg Assoc Archit Inst Korea 2006;8(2):85–92. [19] Burns D. Characterization of wet-mix shotcrete for small line pumping [Master Thesis]. University of Laval: Canada; 2008. [20] Morgan DR, Wolsiefer JT. Wet-mix shotcrete: effect of silica fume form, fly ash, silica fume, slag, and natural pozzolans in concrete. In: Proceedings of the 4th CANMET/ACI international conference, SP-132. Farmington Hills: American Concrete Institute; 1992.

202

K.-K. Yun et al. / Construction and Building Materials 78 (2015) 194–202

[21] Zampini D, Walliser A, Oppliger M, Melbye T, Maltese C, Pistolesi C, et al. Liquid-based set accelerating admixtures for sprayed concrete: a comparison between alkali-free and alkali-rich accelerators. Gallerie e Grandi Opere Sotterranee 2004;72:30–40. [22] Tattersall GH. Workability and quality control of concrete. London: Chapman & Hall; 1991. [23] Tattersall GH, Banfill PFG. The rheology of fresh concrete. London: Pitman Advanced Publishing Program; 1983.

[24] Banfill PFG. Workability of flowing concrete. Mag Concr Res 1980;32(110):17–27. [25] Gjørv OE, High strength concrete, Advances in concrete technology, report MSL 92-6(R), Energy Mines and Resources, Ottawa, Canada; 1992. [26] Llewellyn DH, The effect of polypropylene fibres on the properties of concrete and mortar, B Eng Project, Sheffield City Polytechnic, United Kingdom; 1990.