Mechanical properties of sand modified resins used for bonding CFRP to concrete substrates Academic research paper on "Civil engineering"

HOSTED BY

Gulf Organisation for Research and Development International Journal of Sustainable Built Environment

ELSEVIER

ScienceDirect www.sciencedirect.com

Original Article/Research

Mechanical properties of sand modified resins used for bonding

CFRP to concrete substrates

Aziz. I. Abdulla a'*, Hashim Abdul Razakb, Yassen Ali Salihc, Muataz Ibrahim Ali

a Civil Eng. Dept., College of Eng., Al-Ahliyya Amman University, Jordan Civil Eng. Dept., College of Eng., University Malaya, Malaysia c Civil Eng. Dept., College of Eng., Tikrit University, Iraq d Civil and Environmental Eng. Dept., College of Eng., Samarra University, Iraq

Received 5 April 2016; accepted 8 June 2016

Abstract

This study is an experimental investigation into the properties of adhesive before and after mixing with fine sand, and its behavior on reinforced concrete beams strengthened by CFRP to show the effects of modified adhesive on load-carrying capacity, ductility, stiffness and failure mode of the reinforced concrete beams. Compressive strength, flexural strength and the effect of high temperature on these properties were the focus of the current study in order to prove the efficiency of adding fine sand to improve adhesive properties and reduce cost. Based on the compressive and flexural tests, results indicated that the addition of sand to the adhesive improved its mechanical properties when sand is 50% of the total weight of the adhesive. However, its effect on the modulus of elasticity is minimal. Using adhesive with fine sand increased the ultimate load bearing capacity, ductility, stiffness and toughness of the reinforced concrete beams strengthened by CFRP. The ratio of the fine sand to the adhesive equal to 1 is considered the best in terms of the cost reduction, maintaining workability, as well as maintaining the mechanical properties. Lastly, the use of fine sand with adhesive ensured a significant reduction in the cost of the adhesive and increased the adhesive resistance to temperature. © 2016 Published by Elsevier B.V. on behalf of The Gulf Organisation for Research and Development.

Keywords: Concrete; Fibers; Resin; Sand; Thermal analysis

1. Introduction

The number of civil engineering structures in the world continues to increase, as well as average age. The need for increasing maintenance is inevitable (Gao et al., 2004). One

* Corresponding author. E-mail addresses: aziz_914@hotmail.com (A.I. Abdulla), hashim. abdulrazak@gmail.com (H.A. Razak), eng.yaseenali@gmail.com (Y.A. Salih), muitaz988@yahoo.com (M.I. Ali).

Peer review under responsibility of The Gulf Organisation for Research and Development.

of the most modern methods of strengthening or rehabilitating of concrete is the use of polymer fibers because of their high tensile forces as well as being lightweight and resistant to weather and chemical agents. Three types of FRP have been in use namely: glass fiber reinforced polymer (GFRP), carbon fiber reinforced polymer (CFRP), and aramid fiber reinforced polymer (AFRP). The most important one among these types is CFRP because of its good toughness, good stiffness, high resistance to creep, high resistance to chemical agents, and good resistance to high temperature (Al-Sulayfani and Mohammed, 2013).

http://dx.doi.org/10.1016/j.ijsbe.2016.06.001

2212-6090/© 2016 Published by Elsevier B.V. on behalf of The Gulf Organisation for Research and Development.

The efficiency of reinforced concrete beam strengthening with CFRP depends on proper bonding between the concrete and the CFRP with an epoxy or vinylester adhesive. The adhesive plays an important role in connecting fibers and concrete together. Particularly for interfacial debond-ing and concrete cover separation, the mechanisms relate to the adhesive mechanical properties and interlaminar fracture toughness. Therefore, there is a need to improve ductility and toughness of adhesive (Gao et al., 2004). Epoxy adhesive is used to paste the fibers on the concrete or steel, and is available in several types according to the manufacturers. It always consists of two parts, the first called resin and the second is called hardener. Strengthening of concrete structures with epoxy bonded carbon fiber reinforced polymers (CFRPs) has been proved to be a good strengthening technique. However, this strengthening technique with epoxy adhesives does contain some disadvantages such as diffusion closeness, thermal incompatibility to the base concrete, working environment and minimum application temperature (Taljsten and Blanksvard, 2007).

Sen and others (Sen et al., 2001) study the effect of Florida city climate on the concrete strengthening by CFRP, and the work included the effect of air temperature and cycles of moisturizing and drying. The study proved that carbon fiber and epoxy are highly resistant to weather conditions.

Camata and others (Camata et al., 2007) have studied the effect of exposure concrete models strengthening by plates of carbon fiber to cycles of heating up to 100 °C. The study proved that all kinds of epoxy used significantly affected heating cycles.

Klamer and others (Klamer et al., 2008) have studied the effect of temperature on the beams strengthening by CFRP. Test results have shown that, compared to room temperature, the type of failure and the failure load of the beams tested at 50 °C were not significantly affected. At 70 °C, the type of failure changed for one of the beams from failure in the concrete adjacent to the concrete adhesive interface to failure exactly in the concrete-adhesive interface. The failure loads of the beams tested at 70 °C were not significantly affected compared to room temperature, except for the beam with a relatively short laminate length. For this beam, the load capacity is expected to be mainly related to the capacity of the end anchorage zone, which was negatively affected by the effects of the elevated temperature.

Blanksvard (Blanksvard, 2009) used an adhesive material consisting of reinforced polymer mortar (polymer, mortar, superplastizer, and fibers) instead of epoxy, and has proved in experiments that this new adhesive is efficient for use as an adhesive for CFRP. However this material is expensive and difficult to use because of difficulty controlling the mixing ratio for many components, so this adhesive can be used as crack reinforcement in prefabricated concrete elements.

Fonseca and others (Cabral-Fonseca et al., 2011) study the effect of environmental conditions on three different

types of epoxy used to strengthen or the rehabilitation of concrete structures. The models immersion in sea water and alkaline solutions for 18 months and at a temperature of 40-60 °C. Mechanical properties and the weight loss of the samples have been measured. The study proved that the epoxy is greatly influenced by environmental conditions of exposure to sea water and alkaline solutions. Buyukoz turk and others (Buyukozturk et al., 2012) study the effect of moisture on the cohesion between epoxy and concrete, and tests showed that humidity reduces cohesion between the epoxy and the concrete.

Al-Safy and others (Al-Safy et al., 2012) have studied the thermal and mechanical properties of nanoclay-modified adhesives for use in civil engineering applications. Differential scanning calorimetry (DSC), X-ray diffraction (XRD) and transmission electron microscopy (TEM) were used to characterize the adhesive structure. The glass transition temperature (Tg), measured by DSC, was found to decrease with nanoclay addition. Measurements from XRD and TEM identified an intercalated/exfoliated structure of the nanoclay, nanomer I.30E in the epoxy matrix. The adhesive tensile strength showed a reduction with the addition of nanoclay at elevated temperatures, however, improvement in tensile modulus was found for all nanoclay addition. The bond-loss temperature of CFRP/concrete systems with modified adhesive was observed to be lower than for the control (0% NC) using adhesion (pull-off) tests at elevated temperatures. And also these materials were very expensive.

Nguyen and others (Nguyen et al., 2012) studied the effects of UV to the bond between steel and CFRP. Specimens (epoxy adhesive, CFRP laminates, and steel/CFRP adhesively bonded joints) were exposed to UV for various time periods and identical reference specimens were exposed to only thermal environments without UV. They found that UV exposure did not influence the tensile strength of CFRP composites. The tensile strength of the adhesive reduced by 13.9% while modulus showed a significant increase by 105% after 744 h of exposure. The tensile modulus of adhesive exposed to only thermal environment also increased by 38%, considerably less than that induced by UV exposure. The UV exposure also led to a decrease in joint strength but an increase in stiffness, caused by the temperature effect rather than the UV rays.

Table 1

Properties of adhesive Sikadur®-30.

Properties

Description

Service temperature

Density

Mixing

Pot life

Application temperature Tensile strength Compressive strength E-modulus: tensile E-modulus: compressive

A: white, B: black, A + B: light gray -40 °C to +45 °C (when cured at > +23 °C)

I.65 kg/l kg/l (parts A + B mixed) (at +23 °C) 1:3 (A:B) by weight

+8 °C:120 min, +20 °C:90 min, +35 °C:20 min +8 °C to +35 °C

(Curing 7 day, +35 °C) = 30 N/mm2 (Curing 7 day, +35 °C) = 95 N/mm2

II.200 N/mm2 (at +23 °C) 9.600 N/mm2 (at +23 °C)

Figure 1. Flow table, instant inflating, visible shrinkage of adhesive, and prisms before test.

Abdulla (Abdulla, 2016) has made an experimental investigation into the thermal properties of adhesive before and after the mixing with fine sand. The results of his study had showed that such an addition significantly improved the thermal characteristics, such as: reducing the initial

and the final shrinkage, reducing the heat of the reaction, coefficient of linear expansion, and the coefficient of the thermal conductivity. Also, this addition leads to a reduction in the adhesive cost and a small increase in the compressive strength and the modulus of rupture. The ratio

of the fine sand to the adhesive equal to 1 is considered the best in terms of the cost reduction, maintaining workability, as well as maintaining the mechanical properties.

The present study aims at getting the economic adhesive material used for bonding CFRP or others by mixing one well-known adhesives with cheap material (fine-sand). The sand is cheap compared to the price of epoxy where price of epoxy represents almost 15,000-2500 times the price of sand, so the proposed method leads to a large reduction in the cost of the adhesive (sand price consider negligible), in addition to improving their properties. The study includes evaluating the effects of fine-sand added to the mechanical properties of the adhesive, and the effects of high temperatures to the adhesive properties. Mechanical properties for adhesive with or without fine-sand were mainly investigated, to prove the efficiency of this adding to either improve or at least maintain their properties and reduce the cost. The behavior of reinforced concrete beams' strengthening by CFRP using adhesive with or without fine-sand was investigated to show the effects of modified epoxy adhesive on load-carrying capacity, ductility, stiffness and failure mode of the reinforced concrete beams.

2. Experimental works

sand, so the sand must added to part A or B before mixing A with B (Abdulla, 2016). Also adding fine-sand to part A or B leads to reduce the reaction temperature as will be explained in the results. The temperature of the adhesive is measured after pouring it by thermometer. The operation of the temperature continues until the temperature of the adhesive is settled.

2.3. Consistency of adhesive

The consistency of adhesive is expressed as an adhesive flow, determined according to the procedures of ASTM C

Table 2

Adhesive mixing ratio and flow values.

Sym S/E Flow (%)

Initial Final

134 200

127 188

118 182

110 175

104 158

100 140

S/E = by weight sand/epoxy ratio.

E1 E2 E3 E4 E5 E6

0.25 0.5 1

1.25 1.5

2.1. Materials

- Fine-sand: granular sand was passed through 300-75 im sieve. Thus, the fine sand retained is used for the purpose of this study. The granular sand was washed well and dried before use. The specific gravity of sand used is 2.57 with absorption of 2.41 and the sulfate content after washing is 0.04%.

- Adhesive: a high adhesive bonding paste known as Sikadur®-30 with a medium viscosity material and own density 1.31 kg/L for a mixture of two parts resin (part A) and hardner (part B) is used for the purpose of this study. The properties of the adhesive are as presented in Table 1.

- CFRP: the carbon fiber fabric sheet used in this study is SikaWrap- 300C. The CFRP fibers when loaded in tension did not exhibit any plastic behavior before rupture. The tensile behavior of the CFRP fibers is characterized as a linearly elastic stress-strain relationship up to failure.

2.2. Mixing

Adhesive consists of two parts namely resins (part A) and hardener (part B); which will be mixed in a (1:3) ratio. But, when adding, the sand must be mixed with either part A or B before mixing A and B together, in the case of adding sand to a mixture of A and B after mixing them will be less workability largely, the reason for this is the interaction of adhesive parts which prevents the penetration of

40 60 80 Time (min.)

Figure 2. Temperature of adhesive after mixing.

Setting time (days)

Figure 3. Compressive strength of adhesive cube in different curing ages.

50 100 150 200 Temp. (°C)

Figure 4. Effect of temp. on the adhesive cubes compressive strength.

230 and C 1437 by flow table. The flow table used for computing the consistency of cement mortar was suggested by Abdulla (Abdulla, 2016) for computing the consistency of adhesive. Epoxy adhesive is poured in a truncated cone. The cone is placed on a flow table whose top can be raised and dropped through to a certain height by means of a rotating cam. The mold is removed from the adhesive, and the table is dropped 25 times in 15 s (see Fig. 1). The flow is measured as the resulting increase in the average base diameter of the adhesive mass, measured as a percentage of the original diameter. The authors also suggests

computing initial flow (after removed truncated cone and before the dropping of table), this value is not important and not sensible for cement mortar but it's valuable for adhesive.

2.4. Mechanical properties

A- Casting adhesive cubes (40 x 40 x 40 mm) and prisms (40 x 40 x 160 mm), with different ratios of fine-sand, are as shown in Table 2. The samples were tested at 3, 7, 14, and 18 days. Compressive strength for cubes and modulus of rupture for prisms have been measured. The compressive strength was determined according to ASTM C109 (ASTMC109, 2004). The modulus of rupture was measured according to ASTM C 348 (ASTM-C348, 2004).The prisms were subjected to central point loading with a span of 100 mm. The neat adhesive and the adhesive which is mixed with fine sand have been compared with each other.

B- The effect of temperature on the mechanical properties of cubes and prisms by placing cubes and prisms in the oven for three hours at different temperatures, and then the compressive strength of cubes and the modulus of rupture to reflect influence heat on these properties were measured. The specimens were subjected to four different temperature cycles from 20 °C up to 150 °C, 250 °C, 350 °C and 400 °C. The first part of each cycle consisted of a heating at 10 °C/min up to the target temperature. After that,

Figure 5. Adhesive prisms broken into two parts after exposure to high temperature.

the temperature was held constant for 1 h in order to ensure a uniform temperature throughout the specimens.

C- Caste RC beams have section (150 mm x 150 mm) and length (1000 mm), reinforced by 206 mm at top and 306 mm at bottom with 06 mm@75 mm for shear. Six beams tested up to failure under one point load, then the CFRP was used for the rehabilitation four beams two using adhesive alone and two using the adhesive with fine sand. Strengthened and unstrengthened beams are then tested under one point load to see the type of failure and load capacity of these beams. The beams strengthened by CFRP using adhesive have a symbol SF1, the beams strengthened by CFRP using adhesive with fine-sand (Mix E4) have a symbol SF2, and the beams which are tested without strengthening have a symbol FR1.

All parameters in this table calculate according to the steps outlined by Abdullah and others (Abdulla et al., 2013).

3. Results and discussion

3.1. Mixing

As highlighted earlier, sand mixing must be done with one part of the adhesive (resin or hardener) to ensure getting good workability and to reduce the temperature of the reaction as shown in Fig. 2. Fig. 2 shows clearly the effect of sand in reducing of reaction temperature.

setting time (days)

Figure 7. Modulus of rupture of adhesive prisms in different curing ages.

Deflection mm

Figure 8. Load-deflection curve for adhesive prisms under flexural load (14-days age).

3.2. Consistency of adhesive

The flow of mixing will decrease with increasing sand ratio, for mixes E5 and E6 (sand/adhesive ratio = 1.25, and 1.5) the flow ability is unacceptable (see Table 1),

and the compressive strength will start decreasing as will be explained later in mechanical properties. The optimum ratio for sand to adhesive is 1 (mix E4), where E4 has good workability, and mechanical properties. This is properly discussed later in this study.

3.3. Mechanical properties

- Compressive strength: Fig. 3 shows the compressive strength test results for adhesive cubes with different percentages of fine-sand and different curing ages. Results clearly show that the increase in the proportion of sand increases the compressive strength and the best ratio of sand to adhesive material is the ratio of 1 (mix E4) where the sand is 50% of the total weight of the adhesive. When the ratio of sand to adhesive increases to more than 1, it will affect the workability and the adhesive compressive and tensile strength. This ratio of added (mix E4) gives an increase in the compressive strength equal to 6.12%, and a reduction in the cost of the adhesive to almost half, where the price of sand is negligible

compared to the price of adhesive; Thus, when sand is 50% by total weight of adhesive, this means that a reduction in the cost would be to half. Adding to that the addition of sand increases the resistance to high temperature and improves thermal characteristics. Fig. 4 shows that the compressive strength increases with increasing temperature and then begins to decline with carbonization of adhesive at 300 °C. The reason behind increasing the compressive strength of the adhesive when the curing temperature increases is the annealing phenomenon (temperature annealing). Annealing phenomenon is a phenomenon representing increasing of resistance when exposed to temperature due to reduction in the free spaces and change the molecular configuration (Odegard and Bandyopadhyay, 2011). Among these

Figure 10. Test of SF2.

Figure 11. Load-deflection curves for reinforced concrete beam strengthening with CFRP.

important notes, also there are parts of the core of adhesive cubes flowing at a temperature of 250 °C and above affecting the annealing process, and has not been observed with sand modified resin. At a temperature of approximately 170 °C epoxy begins to show unpleasant odors and these odors increase with rising temperature to be accompanied by a gray fume at 250 °C. It starts being carbonated at a temperature of 300 °C and will carbonize overall within 3 h. At 250 °C E1 shows many cracks and change in surface color on reverse from E4 as shown in Fig. 5.

Fig. 6a represents the relationship between load and deflection for adhesive cubes and adhesive cubes with optimal ratio of sand (E4). It clearly shows from the figure that the addition of sand doesn't affect much on the modulus of elasticity. Fig. 6b shows that the adhesive fragment with a pop sound during a failure, while the adhesive containing sand-powder, it does not fragment and be semi gradually failing.

- Modulus of rupture: Fig. 7 shows the relationship between the modulus of rupture and the added ratio as well as the effect of setting time on modulus of rupture. The figure shows that the best ratio of sand to adhesive is 1, namely the E4 mixes. Fig. 8 represents load-deflection curve for prisms of the adhesive under the influence of central flexural load.

- Reinforced concrete beam strengthening by CFRP: as it was explained previously, reinforced concrete beam has been strengthening by CFRP using adhesive with or without fine-sand. Figs. 9 and 10 illustrate failure stages. Fig. 11 shows load-deflection curve for beams. Using adhesive with fine-sand increases ultimate load carrying capacity by 10%, as well as it has significantly increases ductility, stiffness, toughness, and especially large increasing in stiffness (see Table 3). The failure of SF1 is interfacial debonding where it's concrete cover separation for SF2.

Table 3

Static properties of reinforced concrete beams strengthen by CRFP.

Sym. Fy (N) Fu (N) Ductility Stiffness (N/mm) Toughness (N mm)

RF1 11,300 28,600 35.55 41,852 321,000

SF1 16,600 43,200 14.04 37,727 523,000

SF2 17,700 47,400 81.00 177,000 611,800

All results represent the average of two beams.

4. Conclusions

An experimental study was conducted to investigate the properties of adhesive before and after mixing with fine sand, and its behavior on reinforced concrete beams strengthened by CFRP. Compressive strength, flexural strength, effects of high temperature on the adhesive properties and the cost implication of using sand adhesive were evaluated. Based on this study, the following conclusions can be drawn:

1- The flow of mixing (adhesive workability) is decreased with increasing fine-sand ratio, and the optimum weight ratio for sand to adhesive is 1.

2- Fine-sand must be mixed with one part of the adhesive (resin or hardener) to get a good workability and reduce the reaction temperature.

3- Mixing two parts of adhesive (without fine-sand) produces high reaction temperature, a large expansion and shrinkage during the first hour of reaction and after that until 24 h. One of the important observations is that it is not possible to mix a large amount of epoxy; otherwise the high heat of reaction will lead to inflating, rapid hardening, and losses of the strength.

4- Adding sand to adhesive will increase compressive strength and modulus of rupture (MOR) for adhesive, and also increase the resistance of adhesive to the high temperature.

5- For reinforced concrete beam strengthened with CFRP, using fine-sand with adhesive gives as significant increase in yield load, ultimate load capacity, ductility, stiffness, toughness, and it remarkable increases the stiffness.

6- Using fine-sand can provide adhesive with a useful reduction in cost. The cost could be up to half the cost of the original adhesive.

7- Using fine-sand reduces the effect of high temperatures on the properties of adhesive.

8- Using fine-sand changes the failure mode of RC beams strengthened with CFRP from interfacial debonding to concrete cover separation.

References

Abdulla, A.I., 2016. Thermal properties of sand modified resins used for bonding CFRP to concrete substrates. Int. J. Sustainable Built Environ.

Abdulla, A.I., Salih, Y.A., Saleh, H.M., 2013. Properties of ferrocement slabs containing sawdust. Tikrit J. Eng. Sci. 20 (1), 51-63.

Al-Safy, Rawaa, Al-Mahaidi, Riadh, Simon, George P., Habsuda, Jana, 2012. Experimental investigation on the thermal and mechanical properties of nanoclay-modified adhesives used for bonding CFRP to concrete substrates. Constr. Build. Mater. 28 (1), 769-778.

Al-Sulayfani, Bayar J., Mohammed Roaa, S., 2013. Studying flexural behavior of reinforced concrete beams strengthened with different lengths of FRP strips by using NSM technique. Tikrit J. Eng. Sci. 20 (2), 1-11.

Blanksvard, Thomas, 2009. Strengthening of concrete structures by the use of mineral-based composites (Ph.D.). Lulea University of Technology, Sweden.

Buyukozturk, Oral, Lau, Denvid, Tuakta, Chakrapan, 2012. Durability and long-term performance modeling of FRP-concrete system. In: Paper Presented at the CICE Conference, Rome.

Cabral-Fonseca, Susana, Nunes, Joao Pedro, Rodrigues, Maria Paula, Eusebio, Maria Isabel, 2011. Durability of carbon fibre reinforced polymer laminates used to reinforced concrete structures. Sci. Eng. Compos. Mater. 18, 201-207.

Camata, Guido, Pasquini, Franco, Spacone, Enrico, 2007. High temperature flexural strengthening with externally bonded FRP reinforcement. In: Paper Presented at the Proceedings of 8th International symPosium on Fiber Reinforced Polymer (FRP) Reinforcement for Concrete Structures (FRP8RCS). Patras, Greece.

Gao, Bo, Kim, Jang.-Kyo, Leung, Christopher K.Y., 2004. Experimental study on RC beams with FRP strips bonded with rubber modified resins. Compos. Sci. Technol. 64 (16), 2557-2564.

Klamer, Ernst L., Hordijk, Dick A., Hermes, Michael C.J., 2008. The influence of temperature on RC beams strengthened with externally bonded CFRP reinforcement. Heron 53 (3), 157-185.

Nguyen, Tien.-Cuong, Bai, Yu, Zhao, Xiao.-Ling, Al-Mahaidi, Riadh, 2012. Effects of ultraviolet radiation and associated elevated temperature on mechanical performance of steel/CFRP double strap joints. Compos. Struct.

Odegard, G.M., Bandyopadhyay, A., 2011. Physical aging of epoxy polymers and their composites. J. Polym. Sci., Part B: Polym. Phys. 49 (24), 1695-1716.

Sen, Rajan, Mullins, Gray, Shahawy, Mohsen, Spain, John, 2001. Effect of environment on the integrity of CFRP/concrete bond. In: Paper Presented at the Proceedings of the Eleventh (2001) International Offshore and Polar Engineering Conference, Stavanger, Norway.

Taljsten, Bjorn, Blanksvard, Thomas, 2007. Mineral-based bonding of carbon FRP to strengthen concrete structures. J. Compos. Constr. 11 (2), 120-128.