- Email: [email protected]
Optimal mix and freeze-thaw durability of polysulfide polymer concrete
Construction and Building Materials 127 (2016) 539–545
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Optimal mix and freeze-thaw durability of polysulfide polymer concrete Sungnam Hong a, Hyeonjun Kim b, Sun-Kyu Park c,⇑ a
College of Engineering, Sungkyunkwan University, Suwon 440-746, South Korea Samsung C&T Corporation, Seoul 137-956, South Korea c School of Civil and Architectural Engineering, Sungkyunkwan University, Suwon 440-746, South Korea b
h i g h l i g h t s Polysulfide polymer concrete mix design and freeze–thaw durability were examined. The optimal mix design was determined from two-stage binder tests and mixing tests. Specimen strengths for the optimal mix were compared with design code requirements. Strength and modulus losses after accelerated freeze–thaw cycling were measured.
a r t i c l e
i n f o
Article history: Received 19 May 2015 Received in revised form 27 September 2016 Accepted 6 October 2016
Keywords: Polysulfide Polymer concrete Freeze–thaw Mechanical properties Optimal mix
a b s t r a c t A study was conducted to determine the optimal mix design for polysulfide polymer concrete (PPC). Twostage binder tests and polymer concrete mixing tests were carried out for this purpose. The optimal mixing ratio was determined from the test results. In addition, the strength and freeze–thaw resistance of specimens produced using the optimal mixing ratio were evaluated. The results of the strength tests showed that the specimens satisfied the strength requirements of the relevant design codes. Repeated freezing and thawing significantly decreased the mechanical strength of the specimens but had an insignificant effect on the specimens’ relative dynamic modulus of elasticity (RDME). It was found that more than 300 freeze–thaw cycles could cause a problem for PPC in terms of its strength. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Improvements in the performance of construction materials and advancements in construction technology [1–2] have contributed to growth in public infrastructure. The structural demands on ultra-large bridges and the harsh environments to which they are exposed have prompted industry-wide efforts to reduce the self-weight of ultra-large bridges and improve their strength and durability [3]. There is therefore growing interest in new materials that could replace asphalt concrete and Portland cement concrete, which have historically been favored as pavement materials. A series of studies dating back to the 1950s have evaluated the feasibility of using polymer concrete as a pavement material [4–8]. Based on the results of these studies, the American Concrete Institute (ACI) and the American Association of State Highway and Transportation Officials (AASHTO) established a number of guidelines for polymer concrete use, including the Guide for Polymer Concrete Over⇑ Corresponding author. E-mail address: [email protected] (S.-K. Park). http://dx.doi.org/10.1016/j.conbuildmat.2016.10.056 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
lays and the Guide Specifications for Polymer Concrete Bridge Deck Overlays [9–10]. These publications provide comprehensive design and construction guidelines for the use of polymer concrete as a pavement material for bridges and other structures. However, there have been studies on concrete produced using polysulfide polymer or epoxy resin (referred to hereinafter as polysulfide polymer concrete, or PPC) [11–15]. Furthermore, no studies have been conducted on how to determine the optimal mix design for PPC, despite the fact that such findings are essential if PPC is to be employed in structures. The freeze–thaw durability of PPC should also be assessed before it is used as a pavement material for bridges. This study was conducted to attempt to determine the optimal mix design for PPC. Laboratory tests were first performed to establish the optimal formulation for the binder, which is composed of a polysulfide polymer, an epoxy resin, a hardener, and a catalyst. The optimal mix design for PPC was then determined from a series of laboratory tests. The strength and freeze–thaw durability of PPC specimens developed using the optimal mix design were then tested.
540
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
produced and tested in accordance with ASTM standard test methods [17,18].
2. Experimental program 2.1. Materials PPC is typically composed of a binder, a catalyst, and coarse and fine aggregates. The binder used in this study consisted of a polysulfide polymer, an epoxy resin, a hardener, and a catalyst. For this study, YD-128, a bisphenol A epoxy resin (Kukdo Chemical, Seoul, South Korea), and LP-3, a polysulfide liquid polymer (SPI Chem, Philadelphia, PA, USA), were used as the main materials. JEFFAMINE D-230, an amine hardener manufactured by Huntsman International LLC (Salt Lake City, UT, USA), and HIESCAT HI-54K, a metal salt catalyst manufactured by Keumjung Co., Ltd. (Ulsan, South Korea), were used as the hardener and catalyst, respectively. Information on the physical properties of the epoxy resin, polymer, hardener, and catalyst provided by the manufacturers is shown in Table 1. Silica sand with an average diameter in the range of 0.35– 0.7 mm was selected for use as the coarse aggregate. To ensure excellent workability of the aggregate, silica powder was used as the fine aggregate. The powder was made by grinding the silica sand used as the coarse aggregate and filtering it with a No. 230 (63-l) sieve. The physical properties and chemical composition of the silica aggregates are listed in Table 2.
2.2.2. Polymer concrete The recommended ratio of binder to aggregate (coarse + filler) for polymer concrete is typically 1:3.7 [8]. However, it was reported in a previous study that problems could occur with the mixing properties and liquidity of polymer concrete if PPC was produced with this mixture ratio. It was also reported that the overall performance of PPC could be improved if the ratio of coarse aggregate to filler aggregate was maintained at a rate of 7:3 by weight [15]. Therefore, in this study, the mixture ratio between the coarse and filler aggregates was maintained at 7:3. The container residue (the runoff amount), the 10-min flow, and the thickness after hardening were each measured for binder-to-aggregate ratios of 1:2.0– 1:3.7 by weight. The flow and thickness of each mixture were measured in accordance with ASTM C1362 [19] and ASTM D35649 [20]. However, a standard method for assessing concrete mixture properties has yet to be established. Therefore, the properties of the PPC mixtures prepared in this study were assessed in accordance with KS M5000 [21], which includes some approximate test methods require that 700 g of a PPC specimen be poured into a container of a specific size and that the PPC’s mixture properties be evaluated by measuring the residual amount in the container when the PPC specimen flows out of the container.
2.2. Mixing tests 2.3. Mechanical and durability tests 2.2.1. Binder It was reported in a previous study that the optimal binding result was obtained when the primary materials (bisphenol A epoxy resin and polysulfide liquid polymer) were mixed at a ratio of 6:4 [14]. Accordingly, the primary binding materials used in this study were mixed at this same ratio. Over the two stages of this study, a series of lab tests was performed to establish the formulation of the binder. In the first stage, the variations in the tensile strength, tensile elongation, and gel time of the binder were observed with respect to the proportion of the hardener, which ranged from 3% to 43% relative to the weight of the primary materials. Based on the results, the appropriate range for the optimal mixture ratio for the hardener, relative to the weight of the primary materials, was determined. In the second stage, the effectiveness of the catalyst was examined for each optimal mixture ratio determined in the first-stage tests. The ratio of catalyst content to the primary materials content varied from 1 to 3% by weight, and the optimal binder ratio was determined from a series of results obtained at each mixture rate. Finally, the optimal binder ratio was confirmed by verifying that the test results for the binder specimens for each mixture ratio satisfied the recommendations of ACI 548.9-08 [16]. The binder specimens were
2.3.1. Strength tests Strength tests were carried out to evaluate the performance of the PPC specimens prepared based on the optimal ratio determined from the results of the mixing tests. Compressive strength tests were performed in accordance with ASTM C579 Test Method B [22]. Several cubic specimens 50 50 50 mm in size (height width length) were produced at room temperature for use in the compressive strength testing. For comparison with the compressive strengths of polymer concrete at 3- and 24-h material ages, as recommended by ACI 548.9-08 [16], the specimens were cured at 23 °C for 3 and 24 h, respectively, before being subjected to compressive strength testing. Flexural strength testing was performed using prismatic beam samples 25 25 300 mm in size, in accordance with ASTM C580 Test Method A [23]. The prismatic beam samples were cured at 23 °C for seven days, and the results of the flexural strength tests were compared with the epoxy polymer concrete flexural strength value recommended by ACI 548.5-98 [9]. Various methods can be used to measure the bond strength. In this study, the bond strength was measured using the direct pull-off method, for which the specimen production and measurement procedure are easy. The cylindrical
Table 1 Physical properties of the resin, polymer, hardener, and catalyst. YD-128 (Bisphenol A epoxy resin)
Epoxy equivalent weight (g/eq) 184–190
Viscosity (25 °C, mPas) 11,500–13,500
Hy-Cl (%, max) 0.05
Specific gravity (20 °C) 1.17
LP-3 (Polysulfide liquid polymer)
Molecular weight (g/mol) 1000
Viscosity (25 °C, mPas) 940–1440
Moisture (%, max) 0.1
Specific gravity (20 °C) 1.29
JEFFAMINE D-230 (Amine hardener)
Molecular weight (g/mol) 230
Viscosity (25 °C, mPas) 9
Specific gravity (20 °C) 0.948
Density (20 °C, kg/m3) 946.7
HIESCAT HI-54 K (Epoxy catalyst)
Specific gravity (20 °C) 0.97–0.99
Amine value (KOH mg/g) 610–630
Viscosity (25 °C, mPas) 150–250
Water content (%) <0.5
Mercaptan content (%) 5.9–7.7
541
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545 Table 2 Physical properties of the silica sand and chemical composition of the silica powder. Physical property
Size (mm) 0.35–0.7
Specific gravity (20 °C) 2.64
Unit weight (kg/m3) 1672
Absorption (%) 1.56
Solid volume (%) 64.9
Fineness modulus (%) 2.65
Chemical composition
SiO2 (%) 83.22
Al2O3 (%) 10.00
Fe2O3 (%) 2.21
CaO (%) 0.39
MgO (%) 0.19
All others (%) 3.99
specimens used in the bond strength testing were 50 12 mm in size (diameter height), and the specimens were cured at 23 °C for seven days before testing. The strength of the bond of the specimens to the surface of a prepared steel base was measured. The bond strength tests were performed in accordance with ASTM C1583 [24], and the test results were compared with the bond strength value recommended by ACI 548.9-08 [16]. Tem specimens were prepared for the 3- and 24-h compressive strength tests. Five specimens were prepared for the flexural strength tests, and five specimens were prepared for the bond strength tests. One set of tests consisted of tests of five specimens, and the average value for three specimens, excluding the highest and lowest values of the five, was used as the representative value. 2.3.2. Durability tests A series of tests designed to measure the freeze–thaw durability of the PPC used in this study was performed in accordance with ASTM C666 [25]. Prismatic beam samples measuring 100 100 400 mm in size (height width length) were used to measure the fundamental transverse frequency by using the force resonance method. Specimens that were supported so that they could freely vibrate in the transverse mode were forced by an electromechanical driving unit to vibrate, and the response of each specimen was recorded by a lightweight pickup unit on the specimen. The driving frequency was varied until the measured specimen response reached the maximum amplitude. The frequency that induced the maximum response was considered the fundamental transverse frequency of the specimen. The specimens used in the freeze–thaw testing and the specimens used in the mechanical strength testing were cured at 23 °C for 14 and 7 days, respectively, in accordance with the relevant ASTM standard test methods [22–25]. After being cured, the specimens used in the freeze–thaw testing were placed in a freeze–thaw tester (FTA, Sewon Sys., Ansan, South Korea). Each freeze–thaw cycle lasted 4 h, and the temperature was set to drop from 4 °C to 18 °C at the center of each specimen before it rose again from 18 °C to 4 °C. Five specimens were prepared for use in the freeze–thaw testing. Fifteen specimens were prepared for use in the compressive, flexural, and bond strength testing. As with the earlier strength tests, one set of tests consisted of five individual tests, and the average value for three specimens after excluding the highest and lowest values was used as the representative value. The fundamental transverse frequencies and mechanical strengths were measured at 0, 150, and 300 cycles. The relative dynamic moduli of elasticity (RDME) was calculated from the measured fundamental transverse frequencies. 3. Mixing test results and discussion 3.1. Binder The binder was optimized to maximize the tensile strength and tensile elongation at failure while minimizing the gel time. The objective was to make the uncured and physical properties of the optimized binder satisfy the recommendations of ACI 548-08
Water content (%) <0.1
[16] and the MLTM (Ministry of Land, Transport, and Maritime Affairs of South Korea) design code [26], i.e., a viscosity of 700– 2800 mPas, a gel time of 20–30 min, a tensile strength of 12– 34 MPa, and a tensile elongation of 30–70%. Fig. 1 shows the first set of results from the lab tests performed in the two stages to establish the binder formulation. The results show that the tensile strength of the binder increases as the hardener content, relative to the primary materials content, increases. The results of the tensile elongation tests for the binder were contrary to those of the tensile strength tests. The tensile elongation decreased linearly as the hardener content increased. The reason for this is that the brittleness of the binder increased as the tensile strength of the binder increased. The relationship between the tensile strength, tensile elongation, and rate of the hardener content (the ratio of the hardener content to the primary materials content) was analyzed using linear regression analysis. The regression analysis yielded high correlation coefficients for the tensile strength versus the hardener content (0.9965) and the tensile elongation versus the hardener content (0.9904). However, unlike the tensile strength, the gel time did not decrease linearly with increasing hardener content. As Fig. 1, the decrease in gel time was divided into three approximately linear sections (A, B, and C). Of these, Section exhibited the steepest rate of descent of the gel time with increasing hardener content. Fig. 1 clearly shows that the ratio of the hardener content to the primary materials content influences the physical properties of the binder. If the optimal hardener ratio is selected based on only the tensile strength and elongation, the value would be approximately 50, as indicated by the two linear regression lines shown in Fig. 1 for the tensile strength and elongation. Kim [15] reported, however, that there could be an agitation problem when the ratio of the hardener content to the primary materials content is more than 45% and that this could cause a problem with the use of PPC in construction. This researcher also recommended that a hardener ratio in the range of approximately 10–30% be used to ensure both the desired tensile strength and the desired tensile elongation. None of the gel times measured in this study were within the range (20–30 min) recommended in the design code [26]. Metallic salts were therefore added to accelerate the hardening in the secondstage laboratory tests. Section B, in which the gel time reduction exhibited the most rapid decrease and in which the desired tensile strength and elongation were obtained, was determined to be the best section in which to apply the catalyst. Fig. 2 shows the tensile strength, elongation, and gel time plotted with respect to the ratio of the hardener content to the primary materials content, as well as the effectiveness of the catalyst. The impact of the catalyst on the physical properties of the binder was obvious. For a given ratio of the hardener content to the primary materials content, the tensile strength increased as the amount of catalyst increased, whereas the tensile elongation and gel time decreased. These trends were consistent for all of the cases considered. At a hardener-to-primary materials content ratio of 22%, the longest gel time and the largest tensile elongation were observed when the catalyst content was 1%, but the largest tensile strength was observed when the catalyst content was 3%. For a given hardener content, however, the variation in the tensile strength with
542
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
16 10 %
a
22 %
Tensile strength [MPa]
Considered range
360
8
320 280
12
240 10
Tensile strength Tensile elongation Gel time
b 8
200 160
6 y=4 .1 0 2 R = 6x + 224 0 .9 9 04 .25
4
c 2
120 80 40
0
Tensile elongation [%], Gel time [min]
14
0 .27 +3 3x 965 4 7 0.2 0.9 y= R=
0 3
5
7
9 11 13 15 17 19 21 23 25 27
30
33
36
39
43
Hardener/main material [wt.%] Fig. 1. Effects of the ratio of hardener content to primary materials content on the gel time, tensile strength, and elongation.
18
Tensile strength [MPa]
16
70
Optimal ratio
Tensile strength Tensile elongation Gel time Catalyst
11% 10.6%
-28.6%
50
14 -10.6%
12 10
11.1%
7.1%
40
e -30%
19.5%
30
8
Gel time recommended in the design code
6
20
-52%
4
-42.1%
-47.1%
-37.5%
10
2 0
60
1
2
3
10
10
10
1
2
3
15
15
15
1
2
3
20
20
20
1
2
3
22
22
22
Tensile elongation or Catalyst [%], Gel time [min]
20
0
Hardener/main material [wt.%] Fig. 2. Effect of the ratio of hardener content to primary materials content and the catalyst content on the gel time, tensile strength, and elongation.
variation in the catalyst content was less significant than the variations in the tensile elongation and gel time, as shown in Fig. 2. In addition, the gel time recommended in the design code [26] was achieved only when the hardener-to-primary materials ratio and catalyst content were 22 and 1%, respectively. Therefore, the binder formulation selected, based on the test results, included a hardenerto-primary materials ratio of 22% and a catalyst content of 1%. Multiple binder specimens were produced using the optimal ratio determined from the laboratory test results shown in Fig. 2. Next, the viscosity, gel time, tensile strength, and tensile elongation of the specimens were assessed. The results were compared with the values recommended in ACI 548-08 [16] and the design code (MLTM) [26] and are presented in Table 3. The test results were satisfactory. 3.2. Polymer concrete As Fig. 3 shows, the container residue, which is the amount of residual material in the container and indirectly represents the
Table 3 Comparison of the ACI/MLTM recommendations and the binder test results. Tests
ACI 548-08/MLTM
Binder test
Viscosity (mPas) Gel time (min) Tensile strength (MPa) Tensile elongation (%)
700–2800 20–30 12–34 30–70
1453 25 16.3 66
mixing properties of PPC, increases linearly with the ratio of the binder content to the aggregate content. A higher amount of container residue corresponds to poorer mixing properties. In fact, concretes with high container residues pose difficulties during the stirring process. However, the lower the binder content ratio is, the smaller the 10-min flow value is. Thus, to achieve good mixing and flowability of the polymer concrete, a ratio of 1:3.0, which is close to the ratio of 1:2.98 that corresponds to the point at which the linear regression equations for the container residue and flowability values shown in Fig. 3 intersect, should be selected.
543
440
Container residue 10-min flow Thickness after hardening
400
Intersection at the ratio 1 : 2.98
.2 76 -2 4 x 6 0 0.7 .98 17 = 0 y= R
360 Optimal ratio
4
14 12
320 10
y = -6 0.02x R = 0.9 + 411.26 890
280 240
8
200 6 160 120
4
80 2 40
Thickness after hardening [mm]
Container residue [g] or 10 min flow [mm]
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
0
0 1:2.0
1:2.5
1:3.0
1:3.5
1:3.7
Binder/aggregate ratio Fig. 3. Effect of the binder/aggregate ratio on the residue, flow, and thickness.
However, some bubbles and separation of materials that could have a detrimental effect on the durability of PPC after curing were observed with the naked eye in the PPC specimens produced with the aforementioned ratio of 1:3.0 between the binder and the aggregates. Furthermore, because the PPC specimens produced at a ratio of 1:2.0 were too thin compared to the other specimens, there were concerns that they would pose usability problems after construction. The change in the thickness of the specimens was due to the contraction of PPC during the curing process. The final aggregate ratio, selected based on the test results and the various considerations discussed previously, was 1:2.5. The optimal mixture proportions determined for the binder and the PPC are listed in Table 4.
dynamic modulus of elasticity (RDME) calculated from the measurement results are shown in Fig. 5. The figure shows that the fundamental transverse frequency decreased by an average of 2.67% between the start of the freeze–thaw testing and the 150th freeze–thaw cycle but only decreased by 0.17% between the 150th and 300th freeze–thaw cycles. In other words, beyond 150 freeze–thaw cycles, the number of freeze–thaw cycles did not decrease the fundamental transverse frequency significantly. In addition, the RDME was calculated from the measured fundamental transverse frequencies. The calculated values were, on average, 94.8% at 150 cycles and 94.5% at 300 cycles. It was therefore concluded that the decrease in the resistance of PPC to rapidly repeated cycles of freezing and thawing was negligible.
4. Strength and durability test results and discussion 4.1. Strength tests The results of the laboratory tests conducted to measure the strength of the PPC specimens produced with the optimal mixture proportions listed in Table 4 are shown in the bar graph in Fig. 4. The test results satisfied the strength recommendation in the ACI guidelines [9,16]. The compressive and flexural strengths (41.4 MPa and 23.4 MPa, respectively) were greater than those reported for acrylic polymer in earlier studies (22 MPa and 7 MPa, respectively) [27]. The test results also showed that the compressive and flexural strengths of the PPC specimens produced with the optimal mixture proportions listed in Table 4 were greater than those reported for PPC in another study (39 MPa and 14 MPa, respectively) [14]. 4.2. Durability tests 4.2.1. Variation in the relative dynamic modulus of elasticity The fundamental transverse frequencies measured after various numbers of freeze–thaw cycles and the change in the relative
4.2.2. Variation in strength Fig. 6 shows the compressive, flexural, and bond strengths of the PPC specimens after various numbers of freeze–thaw cycles. As the figure shows, the mechanical strength of PPC decreases steadily as the number of freeze–thaw cycles increases. The results show that the rate at which the bond strength decreases over a given number of freeze–thaw cycles is higher than the rates at which the compressive and flexural strengths decrease. It can also be deduced from Fig. 6 that the strengths are linearly related to the freeze–thaw cycles for the range of cycles considered. The correlation between the number of freeze–thaw cycles and the compressive, tensile, and bond strengths was analyzed using linear regression. The results showed strong correlations between the number of freeze–thaw cycles and the compressive strength (0.9771), the tensile strength (0.9948, and the bond strength (0.9918). These correlation coefficients suggest that an increasing number of freeze–thaw cycles is closely related to decreases in various measures of strength. Accordingly, the number of freeze–thaw cycles at which PPC produced using the mix design described herein would satisfy the strength recommendations of the ACI [9,16] was calculated
Table 4 Optimal mixture proportions for binder and PPC. Binder content
Binder formation (wt.%)
(wt.%)
Bisphenol A epoxy resin
Polysulfide liquid polymer
Hardener
Catalyst
Filler (wt.%)
Aggregate (wt.%)
28.60
13.98
9.32
5.1
0.2
21.40
50.00
544
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
50 Present study ACI recommendation 41.4
40
Strength [MPa]
34.0
30 23.4
20 14.0
10
9.6 7.0 4.4 1.7
0 Comp. (3 h)
Comp. (24 h)
Flexural
Bond
Mechanical properties
RDME
Frequency
10400
No. 1 No. 2 No. 3
No. 1 No. 2 No. 3
Ave. 96.1
10000 Ave.
9838.9
9600
9200
-2.67%
-0.17%
Ave. Ave.
9204.4
98
Ave. 96.0
96
94.1
94
94.7
93.5 93.3
9055.6 94.8
94.5
92
8800 0
150
300
150
300
Relative dynamic modulus of elasticity [%]
Fundamental transverse frequency [Hz]
Fig. 4. Comparison of specimen strengths with ACI recommendations.
Freeze-thaw [cycle] Fig. 5. Effect of freeze–thaw cycles on the fundamental transverse frequencies and RDME of PPC.
using the linear regression equations presented in Fig. 6. The calculated numbers of cycles were 413, 711, and 343 for the compressive, flexural, and bond strengths, respectively. These results suggest that some problems may arise as a result of the low freeze–thaw durability of PPC relative to its compressive strength and bond strength, should the PPC be used in existing or new building structures in environments in which high levels of exposure to freeze–thaw cycles were expected. 5. Conclusions The following conclusions were drawn from the results of the mixture, strength, and freeze–thaw experiments on PPC specimens performed in this study: (1) A series of laboratory tests was performed in two stages to determine the optimal formulation of the binder. The results of the first stage of laboratory tests confirmed that a hardener alone could not satisfy the gel time requirement for the various test conditions considered. In the second-stage
laboratory tests, therefore, metallic salts were added to accelerate the gel time. The content of metallic salts was approximately 1–3% of the primary materials by weight. The optimal ratio of the binder that satisfied all the test requirements was determined with the addition of the catalyst. (2) After PPC specimens were produced with the optimal binder formulation, the container residue, 10-min flow, and thickness after hardening were measured. Based on the test results, a 1:2.5 ratio of the binder to the aggregates was determined to be optimal. (3) The mechanical strength values of the PPC specimens prepared at the optimal mixing ratio suggested in this paper were greater than the strength values recommended by the design code. (4) Exposure to accelerated freeze–thaw cycles had a significant effect on the mechanical strength of PPC, but this was not the case for the relative dynamic modulus of elasticity. This indicates that there is a durability issue associated with PPC under freeze–thaw conditions. The results of the tests
545
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
y = -0.0187x + 41.722 R = 0.9771
50
4.2%
9.9%
Compressive strength Flexural strength Bond strength
16.8%
y = -0.0131x + 23.303 R = 0.9948
13.6%
40 30
ACI recommendation: Over 34 MPa
ACI recommendation: Over 14 MPa y = -0.0076x + 4.305 R = 0.9918
20 10
31.8%
52.0%
150
300
ACI recommendation: Over 1.7 MPa
0 0
150
300
0
150
300
0
28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
Flexural and bond strength [MPa]
Compressive strength [MPa]
60
Freeze-thaw [cycle] Fig. 6. Effect of freeze–thaw cycles on the compressive, flexural, and bond strengths.
showed that there could be a problem with the compressive and bond strengths of PPC at high numbers of freeze–thaw cycles (more than 300 cycles). Therefore, care needs to be taken when PPC is employed in structures constructed in regions in which high numbers of freeze–thaw cycles occur.
[12] [13]
[14]
Acknowledgments
[15] [16]
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) and funded by the Ministry of Education (2013R1A1A2059122). And this work was also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2011-0030040).
[17] [18]
[19]
References [20] [1] M.M. Ali, K.S. Moon, Structural developments in tall buildings: Current trends and future prospects, Archit. Sci. Rev. 50 (3) (2007) 205–223. [2] P.C. Das, D.M. Frangopol, A.S. Nowak, Current and Future Trends in Bridge Design, Construction, and Maintenance 2, Thomas Telford, London, UK, 2001. [3] K.F. Meyer, F.K. Lawrence, Lightweight concrete reduces the weight and increases the span length of pre-tensioned concrete bridge girders, PCI J. 47 (1) (2002) 68–75. [4] E. Kirlikovali, Polymer/concrete composites: A review, Polym. Eng. Sci. 21 (8) (1981) 507–509. [5] Y. Ohama, Recent progress in concrete-polymer composites, Adv. Cem. Based Mater. 5 (2) (1997) 31–40. [6] J.M.L. Reis, A.J.M. Ferreira, Freeze-thaw and thermal degradation influence on the fracture properties of carbon- and glass fiber-reinforced polymer concrete, Constr. Build. Mater. 20 (10) (2006) 888–892. [7] N. Ahn, Moisture sensitivity of polyester and acrylic polymer concrete with metallic monomer powders, J. Appl. Polym. Sci. 107 (2008) 319–323. [8] ACI Committee 548, Report on Polymer-modified Concrete (ACI 548.3-09), American Concrete Institute, Michigan, USA, 2009. [9] ACI Committee 548, Guide for Polymer Concrete Overlays (ACI 548.5-98), American Concrete Institute, Michigan, USA, 1998. [10] AASHTO, Guide Specifications for Polymer Concrete Bridge Deck Overlays, AASHTO-AGC-ARTBA Task Force 34 Report, American Association of State Highway and Transportation Officials, Washington, USA, 1995. [11] A.M. Dinitz, S.S. Michael, The successful use of thin polysulfide epoxy polymer concrete overlays on concrete and steel orthotropic bridge decks, in:
[21] [22]
[23]
[24]
[25]
[26] [27]
Proceedings of the 2010 Structures Congress, Orlando, May 2010, 2010, pp. 530–540. S.S. Michael, J.C. Arif, Thin polysulfide epoxy bridge deck overlays, Transp. Res. Rec. 1749 (2001) 64–67. FHWA, Understanding and Mitigating the Effects of Chloride Deicer Exposure on Concrete. Report No. FHWA-OR-RD-15-08, Federal Highway Administration, Washington, USA, 2014. J.H. Kim, Y.C. Suh, Laboratory evaluation of the polysulfide epoxy overlay material for bridge decks, J. Korean Soc. Road Eng. 13 (2) (2011) 159–166. HJ. Kim, The Physical Properties of Polymer Concrete for Ultra-thin Bridge Deck Pavements (M.E. thesis), Sungkyunkwan University, Seoul, Korea, 2013. ACI Committee 548, Specifications for Type-ES (Epoxy Slurry) Polymer Overlays for Bridge and Parking Garage Decks (ACI 548.9-08), American Concrete Institute, Michigan, USA, 2008. ASTM D 638, Standard Test Method for the Tensile Properties of Plastics, American Society for Testing and Materials, Pennsylvania, USA, 2010. ASTM C881, Standard Specifications of Epoxy-Resin-Base Bonding Systems for Concrete, American Society for Testing and Materials, Pennsylvania, USA, 2014. ASTM C1362, Standard Test Method for the Flow of Freshly Mixed Hydraulic Cement Concrete, American Society for Testing and Materials, Pennsylvania, USA, 2009. ASTM D3549, Standard Test Method for the Thickness or Height of Compacted Bituminous Paving Mixture Specimens, American Society for Testing and Materials, Pennsylvania, USA, 2003. KS M5000, Testing Method for Organic Coatings and Their Related Materials Seoul, Korea:, Korean Industrial Standards (2014). ASTM C579, Standard Test Methods for the Compressive Strength of Chemicalresistant Mortars, Grouts, Monolithic Surfacings, and Polymer Concretes, American Society for Testing and Materials, Pennsylvania, USA, 2012. ASTM C580, Standard Test Method for the Flexural Strength and Modulus of Elasticity of Chemical-resistant Mortars, Grouts, Monolithic Surfacings, and Polymer Concretes, American Society for Testing and Materials, Pennsylvania, USA, 2012. ASTM C1583, Standard Test Method for the Tensile Strength of Concrete Surfaces and the Bond Strength or Tensile Strength of Concrete Repair and Overlay Materials by Direct Tension (Pull-off Method), American Society for Testing and Materials, Pennsylvania, USA, 2008. ASTM C666, Standard Test Method for Resistance of Concrete to Rapid Freezing and Thawing, American Society for Testing and Materials, Pennsylvania, USA, 2008. MLTM, Bridge Deck Pavement Design and Construction, Ministry of Land, Transport, and Maritime Affairs, Seoul, Korea, 2011. T.W. Kim, D.Y. Kim, M.T. Nguyen, H.J. Lee, A study on the physical characteristics of acryl concretes for thin bridge deck pavements, J. Korean Soc. Road Eng. 11 (3) (2009) 1–11.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Optimal mix and freeze-thaw durability of polysulfide polymer concrete Sungnam Hong a, Hyeonjun Kim b, Sun-Kyu Park c,⇑ a
College of Engineering, Sungkyunkwan University, Suwon 440-746, South Korea Samsung C&T Corporation, Seoul 137-956, South Korea c School of Civil and Architectural Engineering, Sungkyunkwan University, Suwon 440-746, South Korea b
h i g h l i g h t s Polysulfide polymer concrete mix design and freeze–thaw durability were examined. The optimal mix design was determined from two-stage binder tests and mixing tests. Specimen strengths for the optimal mix were compared with design code requirements. Strength and modulus losses after accelerated freeze–thaw cycling were measured.
a r t i c l e
i n f o
Article history: Received 19 May 2015 Received in revised form 27 September 2016 Accepted 6 October 2016
Keywords: Polysulfide Polymer concrete Freeze–thaw Mechanical properties Optimal mix
a b s t r a c t A study was conducted to determine the optimal mix design for polysulfide polymer concrete (PPC). Twostage binder tests and polymer concrete mixing tests were carried out for this purpose. The optimal mixing ratio was determined from the test results. In addition, the strength and freeze–thaw resistance of specimens produced using the optimal mixing ratio were evaluated. The results of the strength tests showed that the specimens satisfied the strength requirements of the relevant design codes. Repeated freezing and thawing significantly decreased the mechanical strength of the specimens but had an insignificant effect on the specimens’ relative dynamic modulus of elasticity (RDME). It was found that more than 300 freeze–thaw cycles could cause a problem for PPC in terms of its strength. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Improvements in the performance of construction materials and advancements in construction technology [1–2] have contributed to growth in public infrastructure. The structural demands on ultra-large bridges and the harsh environments to which they are exposed have prompted industry-wide efforts to reduce the self-weight of ultra-large bridges and improve their strength and durability [3]. There is therefore growing interest in new materials that could replace asphalt concrete and Portland cement concrete, which have historically been favored as pavement materials. A series of studies dating back to the 1950s have evaluated the feasibility of using polymer concrete as a pavement material [4–8]. Based on the results of these studies, the American Concrete Institute (ACI) and the American Association of State Highway and Transportation Officials (AASHTO) established a number of guidelines for polymer concrete use, including the Guide for Polymer Concrete Over⇑ Corresponding author. E-mail address: [email protected] (S.-K. Park). http://dx.doi.org/10.1016/j.conbuildmat.2016.10.056 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
lays and the Guide Specifications for Polymer Concrete Bridge Deck Overlays [9–10]. These publications provide comprehensive design and construction guidelines for the use of polymer concrete as a pavement material for bridges and other structures. However, there have been studies on concrete produced using polysulfide polymer or epoxy resin (referred to hereinafter as polysulfide polymer concrete, or PPC) [11–15]. Furthermore, no studies have been conducted on how to determine the optimal mix design for PPC, despite the fact that such findings are essential if PPC is to be employed in structures. The freeze–thaw durability of PPC should also be assessed before it is used as a pavement material for bridges. This study was conducted to attempt to determine the optimal mix design for PPC. Laboratory tests were first performed to establish the optimal formulation for the binder, which is composed of a polysulfide polymer, an epoxy resin, a hardener, and a catalyst. The optimal mix design for PPC was then determined from a series of laboratory tests. The strength and freeze–thaw durability of PPC specimens developed using the optimal mix design were then tested.
540
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
produced and tested in accordance with ASTM standard test methods [17,18].
2. Experimental program 2.1. Materials PPC is typically composed of a binder, a catalyst, and coarse and fine aggregates. The binder used in this study consisted of a polysulfide polymer, an epoxy resin, a hardener, and a catalyst. For this study, YD-128, a bisphenol A epoxy resin (Kukdo Chemical, Seoul, South Korea), and LP-3, a polysulfide liquid polymer (SPI Chem, Philadelphia, PA, USA), were used as the main materials. JEFFAMINE D-230, an amine hardener manufactured by Huntsman International LLC (Salt Lake City, UT, USA), and HIESCAT HI-54K, a metal salt catalyst manufactured by Keumjung Co., Ltd. (Ulsan, South Korea), were used as the hardener and catalyst, respectively. Information on the physical properties of the epoxy resin, polymer, hardener, and catalyst provided by the manufacturers is shown in Table 1. Silica sand with an average diameter in the range of 0.35– 0.7 mm was selected for use as the coarse aggregate. To ensure excellent workability of the aggregate, silica powder was used as the fine aggregate. The powder was made by grinding the silica sand used as the coarse aggregate and filtering it with a No. 230 (63-l) sieve. The physical properties and chemical composition of the silica aggregates are listed in Table 2.
2.2.2. Polymer concrete The recommended ratio of binder to aggregate (coarse + filler) for polymer concrete is typically 1:3.7 [8]. However, it was reported in a previous study that problems could occur with the mixing properties and liquidity of polymer concrete if PPC was produced with this mixture ratio. It was also reported that the overall performance of PPC could be improved if the ratio of coarse aggregate to filler aggregate was maintained at a rate of 7:3 by weight [15]. Therefore, in this study, the mixture ratio between the coarse and filler aggregates was maintained at 7:3. The container residue (the runoff amount), the 10-min flow, and the thickness after hardening were each measured for binder-to-aggregate ratios of 1:2.0– 1:3.7 by weight. The flow and thickness of each mixture were measured in accordance with ASTM C1362 [19] and ASTM D35649 [20]. However, a standard method for assessing concrete mixture properties has yet to be established. Therefore, the properties of the PPC mixtures prepared in this study were assessed in accordance with KS M5000 [21], which includes some approximate test methods require that 700 g of a PPC specimen be poured into a container of a specific size and that the PPC’s mixture properties be evaluated by measuring the residual amount in the container when the PPC specimen flows out of the container.
2.2. Mixing tests 2.3. Mechanical and durability tests 2.2.1. Binder It was reported in a previous study that the optimal binding result was obtained when the primary materials (bisphenol A epoxy resin and polysulfide liquid polymer) were mixed at a ratio of 6:4 [14]. Accordingly, the primary binding materials used in this study were mixed at this same ratio. Over the two stages of this study, a series of lab tests was performed to establish the formulation of the binder. In the first stage, the variations in the tensile strength, tensile elongation, and gel time of the binder were observed with respect to the proportion of the hardener, which ranged from 3% to 43% relative to the weight of the primary materials. Based on the results, the appropriate range for the optimal mixture ratio for the hardener, relative to the weight of the primary materials, was determined. In the second stage, the effectiveness of the catalyst was examined for each optimal mixture ratio determined in the first-stage tests. The ratio of catalyst content to the primary materials content varied from 1 to 3% by weight, and the optimal binder ratio was determined from a series of results obtained at each mixture rate. Finally, the optimal binder ratio was confirmed by verifying that the test results for the binder specimens for each mixture ratio satisfied the recommendations of ACI 548.9-08 [16]. The binder specimens were
2.3.1. Strength tests Strength tests were carried out to evaluate the performance of the PPC specimens prepared based on the optimal ratio determined from the results of the mixing tests. Compressive strength tests were performed in accordance with ASTM C579 Test Method B [22]. Several cubic specimens 50 50 50 mm in size (height width length) were produced at room temperature for use in the compressive strength testing. For comparison with the compressive strengths of polymer concrete at 3- and 24-h material ages, as recommended by ACI 548.9-08 [16], the specimens were cured at 23 °C for 3 and 24 h, respectively, before being subjected to compressive strength testing. Flexural strength testing was performed using prismatic beam samples 25 25 300 mm in size, in accordance with ASTM C580 Test Method A [23]. The prismatic beam samples were cured at 23 °C for seven days, and the results of the flexural strength tests were compared with the epoxy polymer concrete flexural strength value recommended by ACI 548.5-98 [9]. Various methods can be used to measure the bond strength. In this study, the bond strength was measured using the direct pull-off method, for which the specimen production and measurement procedure are easy. The cylindrical
Table 1 Physical properties of the resin, polymer, hardener, and catalyst. YD-128 (Bisphenol A epoxy resin)
Epoxy equivalent weight (g/eq) 184–190
Viscosity (25 °C, mPas) 11,500–13,500
Hy-Cl (%, max) 0.05
Specific gravity (20 °C) 1.17
LP-3 (Polysulfide liquid polymer)
Molecular weight (g/mol) 1000
Viscosity (25 °C, mPas) 940–1440
Moisture (%, max) 0.1
Specific gravity (20 °C) 1.29
JEFFAMINE D-230 (Amine hardener)
Molecular weight (g/mol) 230
Viscosity (25 °C, mPas) 9
Specific gravity (20 °C) 0.948
Density (20 °C, kg/m3) 946.7
HIESCAT HI-54 K (Epoxy catalyst)
Specific gravity (20 °C) 0.97–0.99
Amine value (KOH mg/g) 610–630
Viscosity (25 °C, mPas) 150–250
Water content (%) <0.5
Mercaptan content (%) 5.9–7.7
541
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545 Table 2 Physical properties of the silica sand and chemical composition of the silica powder. Physical property
Size (mm) 0.35–0.7
Specific gravity (20 °C) 2.64
Unit weight (kg/m3) 1672
Absorption (%) 1.56
Solid volume (%) 64.9
Fineness modulus (%) 2.65
Chemical composition
SiO2 (%) 83.22
Al2O3 (%) 10.00
Fe2O3 (%) 2.21
CaO (%) 0.39
MgO (%) 0.19
All others (%) 3.99
specimens used in the bond strength testing were 50 12 mm in size (diameter height), and the specimens were cured at 23 °C for seven days before testing. The strength of the bond of the specimens to the surface of a prepared steel base was measured. The bond strength tests were performed in accordance with ASTM C1583 [24], and the test results were compared with the bond strength value recommended by ACI 548.9-08 [16]. Tem specimens were prepared for the 3- and 24-h compressive strength tests. Five specimens were prepared for the flexural strength tests, and five specimens were prepared for the bond strength tests. One set of tests consisted of tests of five specimens, and the average value for three specimens, excluding the highest and lowest values of the five, was used as the representative value. 2.3.2. Durability tests A series of tests designed to measure the freeze–thaw durability of the PPC used in this study was performed in accordance with ASTM C666 [25]. Prismatic beam samples measuring 100 100 400 mm in size (height width length) were used to measure the fundamental transverse frequency by using the force resonance method. Specimens that were supported so that they could freely vibrate in the transverse mode were forced by an electromechanical driving unit to vibrate, and the response of each specimen was recorded by a lightweight pickup unit on the specimen. The driving frequency was varied until the measured specimen response reached the maximum amplitude. The frequency that induced the maximum response was considered the fundamental transverse frequency of the specimen. The specimens used in the freeze–thaw testing and the specimens used in the mechanical strength testing were cured at 23 °C for 14 and 7 days, respectively, in accordance with the relevant ASTM standard test methods [22–25]. After being cured, the specimens used in the freeze–thaw testing were placed in a freeze–thaw tester (FTA, Sewon Sys., Ansan, South Korea). Each freeze–thaw cycle lasted 4 h, and the temperature was set to drop from 4 °C to 18 °C at the center of each specimen before it rose again from 18 °C to 4 °C. Five specimens were prepared for use in the freeze–thaw testing. Fifteen specimens were prepared for use in the compressive, flexural, and bond strength testing. As with the earlier strength tests, one set of tests consisted of five individual tests, and the average value for three specimens after excluding the highest and lowest values was used as the representative value. The fundamental transverse frequencies and mechanical strengths were measured at 0, 150, and 300 cycles. The relative dynamic moduli of elasticity (RDME) was calculated from the measured fundamental transverse frequencies. 3. Mixing test results and discussion 3.1. Binder The binder was optimized to maximize the tensile strength and tensile elongation at failure while minimizing the gel time. The objective was to make the uncured and physical properties of the optimized binder satisfy the recommendations of ACI 548-08
Water content (%) <0.1
[16] and the MLTM (Ministry of Land, Transport, and Maritime Affairs of South Korea) design code [26], i.e., a viscosity of 700– 2800 mPas, a gel time of 20–30 min, a tensile strength of 12– 34 MPa, and a tensile elongation of 30–70%. Fig. 1 shows the first set of results from the lab tests performed in the two stages to establish the binder formulation. The results show that the tensile strength of the binder increases as the hardener content, relative to the primary materials content, increases. The results of the tensile elongation tests for the binder were contrary to those of the tensile strength tests. The tensile elongation decreased linearly as the hardener content increased. The reason for this is that the brittleness of the binder increased as the tensile strength of the binder increased. The relationship between the tensile strength, tensile elongation, and rate of the hardener content (the ratio of the hardener content to the primary materials content) was analyzed using linear regression analysis. The regression analysis yielded high correlation coefficients for the tensile strength versus the hardener content (0.9965) and the tensile elongation versus the hardener content (0.9904). However, unlike the tensile strength, the gel time did not decrease linearly with increasing hardener content. As Fig. 1, the decrease in gel time was divided into three approximately linear sections (A, B, and C). Of these, Section exhibited the steepest rate of descent of the gel time with increasing hardener content. Fig. 1 clearly shows that the ratio of the hardener content to the primary materials content influences the physical properties of the binder. If the optimal hardener ratio is selected based on only the tensile strength and elongation, the value would be approximately 50, as indicated by the two linear regression lines shown in Fig. 1 for the tensile strength and elongation. Kim [15] reported, however, that there could be an agitation problem when the ratio of the hardener content to the primary materials content is more than 45% and that this could cause a problem with the use of PPC in construction. This researcher also recommended that a hardener ratio in the range of approximately 10–30% be used to ensure both the desired tensile strength and the desired tensile elongation. None of the gel times measured in this study were within the range (20–30 min) recommended in the design code [26]. Metallic salts were therefore added to accelerate the hardening in the secondstage laboratory tests. Section B, in which the gel time reduction exhibited the most rapid decrease and in which the desired tensile strength and elongation were obtained, was determined to be the best section in which to apply the catalyst. Fig. 2 shows the tensile strength, elongation, and gel time plotted with respect to the ratio of the hardener content to the primary materials content, as well as the effectiveness of the catalyst. The impact of the catalyst on the physical properties of the binder was obvious. For a given ratio of the hardener content to the primary materials content, the tensile strength increased as the amount of catalyst increased, whereas the tensile elongation and gel time decreased. These trends were consistent for all of the cases considered. At a hardener-to-primary materials content ratio of 22%, the longest gel time and the largest tensile elongation were observed when the catalyst content was 1%, but the largest tensile strength was observed when the catalyst content was 3%. For a given hardener content, however, the variation in the tensile strength with
542
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
16 10 %
a
22 %
Tensile strength [MPa]
Considered range
360
8
320 280
12
240 10
Tensile strength Tensile elongation Gel time
b 8
200 160
6 y=4 .1 0 2 R = 6x + 224 0 .9 9 04 .25
4
c 2
120 80 40
0
Tensile elongation [%], Gel time [min]
14
0 .27 +3 3x 965 4 7 0.2 0.9 y= R=
0 3
5
7
9 11 13 15 17 19 21 23 25 27
30
33
36
39
43
Hardener/main material [wt.%] Fig. 1. Effects of the ratio of hardener content to primary materials content on the gel time, tensile strength, and elongation.
18
Tensile strength [MPa]
16
70
Optimal ratio
Tensile strength Tensile elongation Gel time Catalyst
11% 10.6%
-28.6%
50
14 -10.6%
12 10
11.1%
7.1%
40
e -30%
19.5%
30
8
Gel time recommended in the design code
6
20
-52%
4
-42.1%
-47.1%
-37.5%
10
2 0
60
1
2
3
10
10
10
1
2
3
15
15
15
1
2
3
20
20
20
1
2
3
22
22
22
Tensile elongation or Catalyst [%], Gel time [min]
20
0
Hardener/main material [wt.%] Fig. 2. Effect of the ratio of hardener content to primary materials content and the catalyst content on the gel time, tensile strength, and elongation.
variation in the catalyst content was less significant than the variations in the tensile elongation and gel time, as shown in Fig. 2. In addition, the gel time recommended in the design code [26] was achieved only when the hardener-to-primary materials ratio and catalyst content were 22 and 1%, respectively. Therefore, the binder formulation selected, based on the test results, included a hardenerto-primary materials ratio of 22% and a catalyst content of 1%. Multiple binder specimens were produced using the optimal ratio determined from the laboratory test results shown in Fig. 2. Next, the viscosity, gel time, tensile strength, and tensile elongation of the specimens were assessed. The results were compared with the values recommended in ACI 548-08 [16] and the design code (MLTM) [26] and are presented in Table 3. The test results were satisfactory. 3.2. Polymer concrete As Fig. 3 shows, the container residue, which is the amount of residual material in the container and indirectly represents the
Table 3 Comparison of the ACI/MLTM recommendations and the binder test results. Tests
ACI 548-08/MLTM
Binder test
Viscosity (mPas) Gel time (min) Tensile strength (MPa) Tensile elongation (%)
700–2800 20–30 12–34 30–70
1453 25 16.3 66
mixing properties of PPC, increases linearly with the ratio of the binder content to the aggregate content. A higher amount of container residue corresponds to poorer mixing properties. In fact, concretes with high container residues pose difficulties during the stirring process. However, the lower the binder content ratio is, the smaller the 10-min flow value is. Thus, to achieve good mixing and flowability of the polymer concrete, a ratio of 1:3.0, which is close to the ratio of 1:2.98 that corresponds to the point at which the linear regression equations for the container residue and flowability values shown in Fig. 3 intersect, should be selected.
543
440
Container residue 10-min flow Thickness after hardening
400
Intersection at the ratio 1 : 2.98
.2 76 -2 4 x 6 0 0.7 .98 17 = 0 y= R
360 Optimal ratio
4
14 12
320 10
y = -6 0.02x R = 0.9 + 411.26 890
280 240
8
200 6 160 120
4
80 2 40
Thickness after hardening [mm]
Container residue [g] or 10 min flow [mm]
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
0
0 1:2.0
1:2.5
1:3.0
1:3.5
1:3.7
Binder/aggregate ratio Fig. 3. Effect of the binder/aggregate ratio on the residue, flow, and thickness.
However, some bubbles and separation of materials that could have a detrimental effect on the durability of PPC after curing were observed with the naked eye in the PPC specimens produced with the aforementioned ratio of 1:3.0 between the binder and the aggregates. Furthermore, because the PPC specimens produced at a ratio of 1:2.0 were too thin compared to the other specimens, there were concerns that they would pose usability problems after construction. The change in the thickness of the specimens was due to the contraction of PPC during the curing process. The final aggregate ratio, selected based on the test results and the various considerations discussed previously, was 1:2.5. The optimal mixture proportions determined for the binder and the PPC are listed in Table 4.
dynamic modulus of elasticity (RDME) calculated from the measurement results are shown in Fig. 5. The figure shows that the fundamental transverse frequency decreased by an average of 2.67% between the start of the freeze–thaw testing and the 150th freeze–thaw cycle but only decreased by 0.17% between the 150th and 300th freeze–thaw cycles. In other words, beyond 150 freeze–thaw cycles, the number of freeze–thaw cycles did not decrease the fundamental transverse frequency significantly. In addition, the RDME was calculated from the measured fundamental transverse frequencies. The calculated values were, on average, 94.8% at 150 cycles and 94.5% at 300 cycles. It was therefore concluded that the decrease in the resistance of PPC to rapidly repeated cycles of freezing and thawing was negligible.
4. Strength and durability test results and discussion 4.1. Strength tests The results of the laboratory tests conducted to measure the strength of the PPC specimens produced with the optimal mixture proportions listed in Table 4 are shown in the bar graph in Fig. 4. The test results satisfied the strength recommendation in the ACI guidelines [9,16]. The compressive and flexural strengths (41.4 MPa and 23.4 MPa, respectively) were greater than those reported for acrylic polymer in earlier studies (22 MPa and 7 MPa, respectively) [27]. The test results also showed that the compressive and flexural strengths of the PPC specimens produced with the optimal mixture proportions listed in Table 4 were greater than those reported for PPC in another study (39 MPa and 14 MPa, respectively) [14]. 4.2. Durability tests 4.2.1. Variation in the relative dynamic modulus of elasticity The fundamental transverse frequencies measured after various numbers of freeze–thaw cycles and the change in the relative
4.2.2. Variation in strength Fig. 6 shows the compressive, flexural, and bond strengths of the PPC specimens after various numbers of freeze–thaw cycles. As the figure shows, the mechanical strength of PPC decreases steadily as the number of freeze–thaw cycles increases. The results show that the rate at which the bond strength decreases over a given number of freeze–thaw cycles is higher than the rates at which the compressive and flexural strengths decrease. It can also be deduced from Fig. 6 that the strengths are linearly related to the freeze–thaw cycles for the range of cycles considered. The correlation between the number of freeze–thaw cycles and the compressive, tensile, and bond strengths was analyzed using linear regression. The results showed strong correlations between the number of freeze–thaw cycles and the compressive strength (0.9771), the tensile strength (0.9948, and the bond strength (0.9918). These correlation coefficients suggest that an increasing number of freeze–thaw cycles is closely related to decreases in various measures of strength. Accordingly, the number of freeze–thaw cycles at which PPC produced using the mix design described herein would satisfy the strength recommendations of the ACI [9,16] was calculated
Table 4 Optimal mixture proportions for binder and PPC. Binder content
Binder formation (wt.%)
(wt.%)
Bisphenol A epoxy resin
Polysulfide liquid polymer
Hardener
Catalyst
Filler (wt.%)
Aggregate (wt.%)
28.60
13.98
9.32
5.1
0.2
21.40
50.00
544
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
50 Present study ACI recommendation 41.4
40
Strength [MPa]
34.0
30 23.4
20 14.0
10
9.6 7.0 4.4 1.7
0 Comp. (3 h)
Comp. (24 h)
Flexural
Bond
Mechanical properties
RDME
Frequency
10400
No. 1 No. 2 No. 3
No. 1 No. 2 No. 3
Ave. 96.1
10000 Ave.
9838.9
9600
9200
-2.67%
-0.17%
Ave. Ave.
9204.4
98
Ave. 96.0
96
94.1
94
94.7
93.5 93.3
9055.6 94.8
94.5
92
8800 0
150
300
150
300
Relative dynamic modulus of elasticity [%]
Fundamental transverse frequency [Hz]
Fig. 4. Comparison of specimen strengths with ACI recommendations.
Freeze-thaw [cycle] Fig. 5. Effect of freeze–thaw cycles on the fundamental transverse frequencies and RDME of PPC.
using the linear regression equations presented in Fig. 6. The calculated numbers of cycles were 413, 711, and 343 for the compressive, flexural, and bond strengths, respectively. These results suggest that some problems may arise as a result of the low freeze–thaw durability of PPC relative to its compressive strength and bond strength, should the PPC be used in existing or new building structures in environments in which high levels of exposure to freeze–thaw cycles were expected. 5. Conclusions The following conclusions were drawn from the results of the mixture, strength, and freeze–thaw experiments on PPC specimens performed in this study: (1) A series of laboratory tests was performed in two stages to determine the optimal formulation of the binder. The results of the first stage of laboratory tests confirmed that a hardener alone could not satisfy the gel time requirement for the various test conditions considered. In the second-stage
laboratory tests, therefore, metallic salts were added to accelerate the gel time. The content of metallic salts was approximately 1–3% of the primary materials by weight. The optimal ratio of the binder that satisfied all the test requirements was determined with the addition of the catalyst. (2) After PPC specimens were produced with the optimal binder formulation, the container residue, 10-min flow, and thickness after hardening were measured. Based on the test results, a 1:2.5 ratio of the binder to the aggregates was determined to be optimal. (3) The mechanical strength values of the PPC specimens prepared at the optimal mixing ratio suggested in this paper were greater than the strength values recommended by the design code. (4) Exposure to accelerated freeze–thaw cycles had a significant effect on the mechanical strength of PPC, but this was not the case for the relative dynamic modulus of elasticity. This indicates that there is a durability issue associated with PPC under freeze–thaw conditions. The results of the tests
545
S. Hong et al. / Construction and Building Materials 127 (2016) 539–545
y = -0.0187x + 41.722 R = 0.9771
50
4.2%
9.9%
Compressive strength Flexural strength Bond strength
16.8%
y = -0.0131x + 23.303 R = 0.9948
13.6%
40 30
ACI recommendation: Over 34 MPa
ACI recommendation: Over 14 MPa y = -0.0076x + 4.305 R = 0.9918
20 10
31.8%
52.0%
150
300
ACI recommendation: Over 1.7 MPa
0 0
150
300
0
150
300
0
28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
Flexural and bond strength [MPa]
Compressive strength [MPa]
60
Freeze-thaw [cycle] Fig. 6. Effect of freeze–thaw cycles on the compressive, flexural, and bond strengths.
showed that there could be a problem with the compressive and bond strengths of PPC at high numbers of freeze–thaw cycles (more than 300 cycles). Therefore, care needs to be taken when PPC is employed in structures constructed in regions in which high numbers of freeze–thaw cycles occur.
[12] [13]
[14]
Acknowledgments
[15] [16]
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) and funded by the Ministry of Education (2013R1A1A2059122). And this work was also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2011-0030040).
[17] [18]
[19]
References [20] [1] M.M. Ali, K.S. Moon, Structural developments in tall buildings: Current trends and future prospects, Archit. Sci. Rev. 50 (3) (2007) 205–223. [2] P.C. Das, D.M. Frangopol, A.S. Nowak, Current and Future Trends in Bridge Design, Construction, and Maintenance 2, Thomas Telford, London, UK, 2001. [3] K.F. Meyer, F.K. Lawrence, Lightweight concrete reduces the weight and increases the span length of pre-tensioned concrete bridge girders, PCI J. 47 (1) (2002) 68–75. [4] E. Kirlikovali, Polymer/concrete composites: A review, Polym. Eng. Sci. 21 (8) (1981) 507–509. [5] Y. Ohama, Recent progress in concrete-polymer composites, Adv. Cem. Based Mater. 5 (2) (1997) 31–40. [6] J.M.L. Reis, A.J.M. Ferreira, Freeze-thaw and thermal degradation influence on the fracture properties of carbon- and glass fiber-reinforced polymer concrete, Constr. Build. Mater. 20 (10) (2006) 888–892. [7] N. Ahn, Moisture sensitivity of polyester and acrylic polymer concrete with metallic monomer powders, J. Appl. Polym. Sci. 107 (2008) 319–323. [8] ACI Committee 548, Report on Polymer-modified Concrete (ACI 548.3-09), American Concrete Institute, Michigan, USA, 2009. [9] ACI Committee 548, Guide for Polymer Concrete Overlays (ACI 548.5-98), American Concrete Institute, Michigan, USA, 1998. [10] AASHTO, Guide Specifications for Polymer Concrete Bridge Deck Overlays, AASHTO-AGC-ARTBA Task Force 34 Report, American Association of State Highway and Transportation Officials, Washington, USA, 1995. [11] A.M. Dinitz, S.S. Michael, The successful use of thin polysulfide epoxy polymer concrete overlays on concrete and steel orthotropic bridge decks, in:
[21] [22]
[23]
[24]
[25]
[26] [27]
Proceedings of the 2010 Structures Congress, Orlando, May 2010, 2010, pp. 530–540. S.S. Michael, J.C. Arif, Thin polysulfide epoxy bridge deck overlays, Transp. Res. Rec. 1749 (2001) 64–67. FHWA, Understanding and Mitigating the Effects of Chloride Deicer Exposure on Concrete. Report No. FHWA-OR-RD-15-08, Federal Highway Administration, Washington, USA, 2014. J.H. Kim, Y.C. Suh, Laboratory evaluation of the polysulfide epoxy overlay material for bridge decks, J. Korean Soc. Road Eng. 13 (2) (2011) 159–166. HJ. Kim, The Physical Properties of Polymer Concrete for Ultra-thin Bridge Deck Pavements (M.E. thesis), Sungkyunkwan University, Seoul, Korea, 2013. ACI Committee 548, Specifications for Type-ES (Epoxy Slurry) Polymer Overlays for Bridge and Parking Garage Decks (ACI 548.9-08), American Concrete Institute, Michigan, USA, 2008. ASTM D 638, Standard Test Method for the Tensile Properties of Plastics, American Society for Testing and Materials, Pennsylvania, USA, 2010. ASTM C881, Standard Specifications of Epoxy-Resin-Base Bonding Systems for Concrete, American Society for Testing and Materials, Pennsylvania, USA, 2014. ASTM C1362, Standard Test Method for the Flow of Freshly Mixed Hydraulic Cement Concrete, American Society for Testing and Materials, Pennsylvania, USA, 2009. ASTM D3549, Standard Test Method for the Thickness or Height of Compacted Bituminous Paving Mixture Specimens, American Society for Testing and Materials, Pennsylvania, USA, 2003. KS M5000, Testing Method for Organic Coatings and Their Related Materials Seoul, Korea:, Korean Industrial Standards (2014). ASTM C579, Standard Test Methods for the Compressive Strength of Chemicalresistant Mortars, Grouts, Monolithic Surfacings, and Polymer Concretes, American Society for Testing and Materials, Pennsylvania, USA, 2012. ASTM C580, Standard Test Method for the Flexural Strength and Modulus of Elasticity of Chemical-resistant Mortars, Grouts, Monolithic Surfacings, and Polymer Concretes, American Society for Testing and Materials, Pennsylvania, USA, 2012. ASTM C1583, Standard Test Method for the Tensile Strength of Concrete Surfaces and the Bond Strength or Tensile Strength of Concrete Repair and Overlay Materials by Direct Tension (Pull-off Method), American Society for Testing and Materials, Pennsylvania, USA, 2008. ASTM C666, Standard Test Method for Resistance of Concrete to Rapid Freezing and Thawing, American Society for Testing and Materials, Pennsylvania, USA, 2008. MLTM, Bridge Deck Pavement Design and Construction, Ministry of Land, Transport, and Maritime Affairs, Seoul, Korea, 2011. T.W. Kim, D.Y. Kim, M.T. Nguyen, H.J. Lee, A study on the physical characteristics of acryl concretes for thin bridge deck pavements, J. Korean Soc. Road Eng. 11 (3) (2009) 1–11.