- Email: [email protected]
Effects of epoxy, hardener, and diluent types on the workability of epoxy mixtures
Construction and Building Materials 158 (2018) 369–377
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effects of epoxy, hardener, and diluent types on the workability of epoxy mixtures Eren Ozeren Ozgul, M. Hulusi Ozkul ⇑ Istanbul Technical University, Faculty of Civil Engineering, Maslak 34469, Istanbul, Turkey
h i g h l i g h t s The addition of 5% diluent caused a decrease in viscosities between 52 and 72%. Glycidylether C12-C14 alcohol is the most effective diluent in reducing the viscosities of epoxy-diluent mixtures. There is a correlation between the flow of the mortars and the viscosities of the epoxy-diluent mixtures. Altering the type of the hardener used in epoxy mortars may increase the flow rate up to 2.8-fold. The higher the functionality of a hardener the shorter is the pot life of epoxy-hardener-diluent mixtures.
a r t i c l e
i n f o
Article history: Received 21 August 2017 Received in revised form 28 September 2017 Accepted 2 October 2017
Keywords: Polymer concrete Epoxy mortar Viscosity Workability Flow Pot life
a b s t r a c t Three types of epoxy resins, namely, diglycidyl ether of bisphenol A and F, and a third obtained by mixing the two resins, and three types of reactive diluents, based on glycidylether, were used to form epoxy mixtures. The effect of diluent content up to 15 wt% on the viscosities of the three epoxies was investigated. The flow properties of mortars prepared with epoxy, hardener, and diluent, as a binder, were compared with respect to the viscosity and molecular weight of the epoxies and diluents, as well as the viscosity and shape of the hardeners used. Six types of hardeners, including four aliphatic and two based on cycloaliphatic amine, were used in the production of mortars. Furthermore, the effects of the type of epoxy, content, type and functionality of diluents, and the functionality and molecular structure of hardeners, on the pot lives of the mixtures were examined. In this study, it is found that not only the viscosities of resin and diluents but also viscosities and molecular structures of hardener and accelerator are effective on the workability properties of epoxy mortars. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction The term ‘‘Concrete-Polymer Composite” (C-PC) is commonly used to define concretes containing a polymer as a binder, or as part of a binder [1]. The former type of concrete, known as a ‘‘polymer concrete” (PC), contains a low viscosity resin as the polymer and does not include any cement. The latter type, named as ‘‘polymer cement concrete” (PCC), includes a polymer, usually in the form of latex or emulsion, which serves as a modifier and also as a part of the Portland cement binding system. A third type of C-PC, namely, ‘‘polymer impregnated concrete” (PIC), can be added to this list. Epoxy resins are widely used in civil engineering applications. In view of their high resistance to abrasion, epoxy mortars and concretes can be used as surface coatings in bridges, ⇑ Corresponding author. E-mail addresses: [email protected] (E. Ozeren Ozgul), [email protected] (M.H. Ozkul). https://doi.org/10.1016/j.conbuildmat.2017.10.008 0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
pavements, and in industrial floors. Due to their outstanding durability, epoxy resins can be used in industrial buildings, underground applications, and in constructions on seashores. Another important area of application of epoxy resins is in repairing reinforced concrete structures for either filling concrete cracks or patching damaged concrete, due to the excellent adhesion of epoxy to concrete surfaces. In addition, epoxy resins also find applications in corrosion repairing. However, epoxies can pose certain health hazards and care should be taken while using them. Diglycidyl ether of bisphenol A (DGEBA), based on epichlorohydrin, is the most widely used epoxy (Fig. 1), whose viscosity depends on the number of repeating units (n) and the molecular weight. The diglycidyl ether of bisphenol F (DGEBF) has a similar molecular structure to DGEBA, but with the difference that the two methyl groups attached to the carbon located between the benzene groups are replaced by H atoms. F-type epoxies have lower molecular weight than A-type and show lower viscosities.
370
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
For this reason, A and F types are mixed to lower the viscosity of the mixture and also to prevent the crystallization of DGEBA. In the pre-polymerized state, epoxy resins are in the liquid state and a crosslinking process is necessary to convert them into a thermoset polymer. A and F type epoxy resins contain epoxy groups at the end of the molecule (two functionalities) consisting of an oxygen atom bonded to two carbon atoms so as to form a threemembered ring. The strained character of the CAO bond and the high electron affinity of O make the epoxy group highly unstable [2], which means that a number of systems can open the ring to attach to the epoxy resin molecules. Molecules which can react with epoxy groups are known both as hardeners and curing or crosslinking agents. Hardener molecules enable the resin molecules to connect to each other to form a three-dimensional network. There are many types of hardeners, but the most important ones are classified into aliphatic and aromatic amines, anhydrides, and polyamides. Usually, hardener molecules are much smaller than resin molecules and therefore, have low viscosities. The addition of a hardener into a resin lowers the viscosity of the mixture to different extents depending on the viscosity and shape of the hardener and its stoichiometric amount used in the mixture. In some cases, it may be necessary to modify the viscosity of an epoxy resin by using diluents to improve its processability [3]. Diluents reduce the viscosity of epoxy mixtures by weakening the interaction between the resin molecules [4]. There are two main types of diluents, namely, non-reactive and reactive. Non-reactive diluents, which include aromatic hydrocarbons such as toluene or xylene, dibutyl phthalate, styrene and various phenolic compounds [5], do not participate in the reaction between the epoxy and the hardener. Reactive diluents can be classified into epoxy-based and those which are not based on epoxides. Among the reactive diluents, those with epoxy groups significantly influence the properties of epoxy resins. They can chemically bind to the resin and form a part of the network following crosslinking. For this reason, they are less volatile and therefore, more environmentally friendly than non-reactive diluents. Nonvolatile reactive diluents have been used to increase the workability of epoxy mortars, where a fivefold increase in fluidity was reported when 15% of a diluent was added [6]. Epoxidized soybean oil (20% ratio), was added as a reactive diluent to modify DGEBA, and the rheological properties were measured using a rheometer [7]. Test results showed that when the concentration of the oil in the mixture increased, the consistency index (zero shear viscosity) decreased, indicating improved flow properties. In another study, the reactive diluent vinyl cyclohexane dioxide at concentrations up to 25% was used in epoxy-amine systems and a reduction in both apparent viscosity and yield stress was reported [8]. To improve the impact strength of epoxy resins, polyol based reactive diluents were added [9], which, besides increasing the toughness, decreased viscosity, extended pot life and improved wetting properties. Ethanol was used as a diluent in an epoxy system, which allowed the maximum silica filler content to be increased from 40% to 60% [10]. Loos et al. [11] performed rheological measurements on DGEBA resin, using acetone as a diluent and showed that the addition of 10% acetone reduced the viscosity by 50%, which resulted in a better dispersion of nanofillers. In this study, three types of epoxy resins, six amine type hardeners (four aliphatic and two cycloaliphatic), and three reactive diluents (glycidyl ether-based) were used [12]. In the first part of the study, the viscosity of epoxy-diluent mixtures and workability properties of epoxy-diluent-hardener-sand mortars were investigated and the results were presented in this paper. In the second part of the study, the hardened state properties (compressive and flexural strengths, adherence strength to the concrete surface and
chemical resistance) of the same mortars have been explored and the results will be presented in a forthcoming paper.
2. Experimental 2.1. Materials 2.1.1. Epoxy resins The epoxy resins used in this study are diglycidyl ether of bisphenol A (DGEBA) (Fig. 1) and diglycidyl ether of bisphenol F (DGEBF), both of which are commercial products. As a third type of epoxy resin, the mixture of DGEBA and DGEBF (70% from the former and 30% from the latter) (DGEBAF), which is sold commercially, was used. Since DGEBF has lower viscosity relative to DGEBA, the two resins are blended to lower the viscosity; in this way each resin prevents the crystallization of the other one at the room temperature [13]. General information on the epoxy resins is given in Table 1.
2.1.2. Hardeners Six hardeners are used in this study and their properties are given in Table 2. For the hardeners trimethylhexamethylene diamine, cyanoethylated mixture of isomers (trimethylhexane -1,6-diamine) (CTMDA) and isophorone diamine (IPDA), the long pot-life times necessitated the use of benzyl alcohol (density 1.04 g/ml and viscosity@25 °C 8 cp.) as an accelerator (Fig. 1).
2.1.3. Reactive diluents The derivatives of glycidyl ether are the most widely used reactive diluents. Three types of glycidyl ether-based reactive diluents are employed, namely, a monofunctional reactive diluent (RDMF) (glycidylether C12-C14 alcohol), a difunctional reactive diluent (RDDF) (1,6 Hexanediol diglycidylether), and a trifunctional reactive diluent (RDTF) (trimethylopropane glycidylether) (Table 3 and Fig. 1).
2.1.4. Aggregate The aggregate mixture used in the study consists of 82% silica sand (density: 2.631 g/cm3) in size in the range 0.075–4 mm and 18% of fine silica (density: 2.67 g/cm3) with size <100 lm.
2.1.5. Resin mixtures In the mortar formulations, silica sand, fine silica and epoxy resin binders were used. Epoxy resin binder consists of an epoxy, a reactive diluent (if exists), stoichiometric amount hardener for both epoxy and diluent, and an accelerator (if exists). When a diluent is used, the amount of epoxy is reduced with the same amount of diluent. The quantities of hardeners were determined from stoichiometry calculations and the amounts of each hardener calculated for DGEBA are listed in Table 2; adjustments from these values were made for the other epoxies and diluents. For hardeners CTMDA or IPDA, pot–life times were very long, and hence, benzyl alcohol was used in 30 and 10% with respect to epoxy (and diluent if exists) as the accelerator. The resin binder/aggregate ratio was determined to have at least a flow rate of 100 ± 10%. In all mortar mixtures, epoxy resin binder, sand and fine silica contents were 15.1%, 69.6% and 15.3% by mass, respectively.
2.2. Testing procedures The tests were carried out at a temperature and R.H of 23 ± 2 °C and 60 ± 5%, respectively and all the materials used were conditioned for 24 h before testing.
2.2.1. Viscosity measurement Viscosity measurements were carried out using a Brookfield viscometer on mixtures of resin and diluent.
2.2.2. Flow measurement The flow of fresh mortar was determined using a cone of height 60 mm, a bottom surface diameter of 100 mm, and a top surface diameter of 70 mm.
2.2.3. Pot life Pot life is the time required for the epoxy resin-hardener-diluent mixtures to become rigid. For the mixtures containing CTMDA or IPDA, an accelerator (benzyl alcohol) was used as an additive. Pot life tests were carried out on 100 g of resin (with or without diluent) mixed with a stoichiometric amount of the hardener at room temperature in accordance with ISO 10364: 2007(E) [16]. The pot life of the mixture was measured from the amount of time between when it was first mixed, up to the point when the mixture could not be spread by hand.
371
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
Epoxy resin (DGEBA)
Hardeners
Meta-xylenediamine (MXDA) Trimethyl hexamethylene diamine (TMDA)
Triethylene tetraamine (TETA) Isophorone diamine (IPDA)
Accelerator (Benzyl alcohol)
Tris (dimethylaminomethyl) phenol (TDMAP) Reactive diluents
Hexanediol diglycidylether [14]
Trimethylopropane glycidylether [15]
C12-C14 Aliphatic glycidylether [14] Fig. 1. Molecular structures of epoxy, hardeners, accelerator and diluents used. (See above-mentioned references for further information).
Table 1 Properties of resins used in the study. Resin
Density (g/ml)
Viscosity (cp.)
Epoxy eq. weight (g/val)
Diglycidyl ether of bisphenol A (DGEBA) Diglycidyl ether of bisphenol F (DGEBF) A mixture of %70 DGEBA and %30 DGEBF (DGEBAF)
1.16 1.19 1.17
16700 5850 7500
194.9 171.2 182.5
3. Test results and discussion 3.1. Viscosity Fig. 2 shows the variation of viscosity of resins (DGEBA, DGEBF and DGEBAF) containing different amounts of the RDMF diluent. As
expected, viscosities drop with increase in diluent substitution ratio. The highest viscosities are observed for the DGEBA mixtures and the lowest for DGEBF-containing mixtures, similar to viscosities of the neat resins. The addition of 5% of diluent caused a 52– 72% decrease in viscosities with respect to the diluent-free samples; however, further increase in diluent rates to 10 and 15% does
372
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
Table 2 Properties of hardeners and stoichiometric quantities for 100 g of DGEBA. Hardener
Type
Viscosity (cp)
Density (g/ml)
Mol. Weight
Quantity (g)
Trimethyl hexamethylene diamine (TMDA) Meta-xylenediamine (MXDA) Triethylene tetraamine (TETA) Trimethyl hexamethylene diamine (cyanoethylated mixture of isomers) (CTMDA) Isophorone diamine (IPDA) Tris (dimethylaminomethyl) phenol (TDMAP)
Aliphatic Aliphatic Aliphatic Aliphatic Cycloaliphatic Accelerated cycloaliphatic
6 8 26 30 16 150
0.87 1.05 0.98 0.92 0.92 1.01
158.3 136.2 146.2 316.4 170.3 265.4
40 34 13 37.2 42.6 20
Table 3 Properties of reactive diluents used in the study. Reactive diluent
Density (g/ml)
Viscosity (cp.)
Epoxy eq. weight (g/val)
Glycidylether C12-C14 alcohol (monofunctional) (RDMF) 1,6 Hexanediol diglycidylether (difunctional) (RDDF) Trimethylopropane glycidylether (trifunctional) (RDTF)
0.89 1.07 1.15
10 23 149
289.4 142.9 142.0
4000
18000
3500 16000
VISCOSITY (cp)
12000 10000 DGEBA+RDMF DGEBF+RDMF DGEBAF+RDMF
8000 6000
VISCOSITY (cp)
3000 14000
2500 2000
DGEBA DGEBAF DGEBF
1500 1000 500 0
4000
RDMF
RDDF
RDTF
DILUENT TYPE
2000 0
Fig. 3. The effect of 10% diluent addition on the viscosities of epoxies.
0
5
10
15
20
DILUENT RATE (%) Fig. 2. Variation of viscosity of epoxies with diluent (RDMF) rate.
not proportionally decrease viscosities. Instead, the viscosities are reduced by 81–88% and 91–95%, respectively. Similar to our observations, it was reported that the addition of 5% glycidyl ether of eugenol reduced the viscosity of the diglycidyl ether of diphenolate n-pentyl ester (epoxy) by 38%, whereas, the viscosity was reduced by 70% and 100% when the diluent concentration was increased to 10% and 15%, respectively [17]. In addition, the differences between the viscosities of the mixtures decrease for higher diluent rates. For a diluent content of 5%, the difference between the maximum and minimum viscosity is 25.8% of the initial difference between the viscosities of the resins without diluents; however, this difference is reduced to 3.3% when the diluent content is increased to 15%. For liquid mixtures, many equations have been proposed to predict the viscosity of a mixture in terms of its components. One of them is the Grunberg and Nissan [18] equation, which is as follows:
ln gmix ¼ x1 lng1 þ x2 lng2 þ x1 x2 G12
ð1Þ
where g is the viscosity, x is the molar fraction of the components in the mixture and G12 is a temperature dependent constant. When the viscosities of the epoxies and diluents as given in Table 1 and 3, and the test results obtained after addition of diluents as shown in Fig. 2 are inserted in Eq. (1), negative G12 values are obtained for all
the binary systems tested, indicating that dispersive forces are predominant [19,20]. Under the assumption G12 = 0, Eq. (1) overestimates the viscosity of the resin-diluent mixtures; however, the trend in the viscosities of the mixtures can be estimated as shown in Fig. 2. The viscosities of resin mixtures with 10% added diluents are compared in Fig. 3. This figure shows that RDMF is the most effective diluent for reducing the viscosities of mixtures followed by RDDF. The highest viscosities are observed for mixtures containing RDTF for the three resins tested. The epoxy-diluent mixtures in increasing order of viscosity in terms of the diluent used are as follows:
glycidylether C12 C14 alcoholðRDMFÞ < 1; 6 Hexanediol diglycidyletherðRDDFÞ < trimethylopropane glycidyletherðRDTFÞ The molecular weight of a diluent also influences the viscosity of an epoxy-diluent mixture with the lower molecular weight leading to lower viscosity [21]. The molecular weights of the diluents used in this study are as follows (Table 3).
1; 6 Hexanediol diglycidylether ð285:7 gÞ < glycidylether C12 C14 alcohol ð289:4 gÞ < trimethylopropane glycidylether ð426:1 gÞ The viscosities of these diluents in increasing order are (Table 3):
373
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
glycidylether C12 C14 alcohol < 1;6 Hexanediol diglycidylether < trimethylopropane glycidylether Although the molecular weight of hexanediol diglycidylether is slightly smaller than that of C12-C14 glycidylether, the latter diluent lowers the viscosity of the resins to a greater extent than the former, indicating that the viscosity of a diluent is more important than its molecular weight in reducing the viscosity of the epoxy mixture. Hong et al. [22] tested four different reactive diluents in bisphenol-A epoxy resin with the aim to reduce the viscosity. Mixtures of the epoxy with 10% added diluents showed the following trend in viscosities (the mixtures are denoted by the names of the diluent used):
butyl glycidyl ether < styrene oxides < phenyl glycidyl ether < aliphatic glycidyl ether
to the viscosity and decreases exponentially, with an exponent of 1/8 [23]. Most of the studies relating the flow to rheological parameters were carried out on Portland cement based mixtures. Bouvet et al. [24] concluded from flow tests that for cement pastes, a lower viscosity led to a higher spread and proposed an expression to calculate the final spread as a function of the rheological parameters such as yield stress and viscosity. However, there are other studies showing that there is only a weak relationship between the spread and the viscosity [25]. In the slump flow test, a cone is first filled with a mixture, after which, it is removed vertically, while the diameter of the spread material is measured. A comparison of the flow of different mixtures is done by comparing the change in flow area (A) with respect to the initial area (A0). Flow rate:
w¼
If these diluents are arranged in order of increasing molecular weights, the only modifications in the above trend are in the positions of the butyl glycidyl ether and styrene oxides. Further comparison reveals that the viscosities of the diluents follow the same trend as the viscosities of the corresponding epoxy-diluent mixtures, showing that the diluent viscosity may be a more important parameter than the molecular weight in reducing the viscosity of the epoxy resin. When the different types of resins are compared for each diluent (Fig. 3), the highest viscosities are observed for DGEBA mixtures, with DGEBF having the lowest viscosities, which again shows a trend similar to the viscosities of the pure resins. The change in viscosities with the diluent content is given in Fig. 4 for DGEBA epoxy-diluent mixtures. As expected, the viscosities of the mixtures decrease with increase in diluents rate for the three reactive diluents used. A comparison of the diluents shows that RDMF is the most efficient with the lowest viscosity, except for the case of 5% diluent, where the viscosity is slightly greater than that of RDDF. Test results given in Fig. 4 reveal a 66–81% reduction in the viscosity of DGEBA for 5% diluent addition, between 78 and 88% for 10%, and finally, from 83% up to 95% when the diluent amount was 15%. The lowest viscosities are obtained for mixtures with RDMF, followed by RDDF, while mixtures with RDTF show the highest viscosities; this trend is similar to that observed for the viscosities of the individual diluents. 3.2. Flow
A A0 A0
The flow rates of epoxies (all the three types used in this study)hardeners TETA or TDMAP-with or without reactive diluents (all the three types used)-sand mortars just after the mixing process are given in Fig. 5 together with the viscosities of the same mixtures without the addition of any hardener or sand (from Figs. 2 and 3). Fig. 5 shows that there is a correlation between the flows of the mortars and the viscosities of the epoxies-with or without added diluent mixtures. Fig. 5 also reveals that the addition of TDMAP into the mortars leads to a greater improvement in the flow rate as compared to TETA. The flow properties of mortars prepared with DGEBA epoxy resin with 10% of the three reactive diluents together with the six different hardeners at different rates (Table 2), are displayed in Fig. 6. The viscosity reducing effect (diluent effect) of the hardeners on the flow behavior of the mortars can also be seen in the same figure. By only altering the type of the hardener used, a 100–280% (a 2.8-fold increase) change in the flow rates of the plain mortars without diluents can be achieved. Epoxy mortars, with CTMDA and IPDA as hardeners, with addition of 30% and 10% of accelerator BA, respectively, are also shown in the same figure. Fig. 6 reveals that the highest flows are observed for mixtures with CTMDA as the hardener and 30% benzyl alcohol (BA) as the accelerator, followed by IPDA hardener with 10% BA, demonstrating the utility of BA as a diluent in view of its low viscosity of 5.47 cp. Following mortars containing added accelerator, those containing hardeners MXDA and TMDA, which have the lowest viscosities among all the hardeners tested, show the highest flow rates among the remaining mortars. The flow rates of CTMDA and IPDA-added
Many studies have attempted to relate the slump or spread of mortars and concretes to their rheological parameters, measured using rheometers. Flow tests have been carried out on epoxy resin-diluent mixtures filled with sand and sand flour and it was reported that the spread of the mortars is inversely proportional
500
400
FLOW RATE (%)
VISCOSITY (cp)
20000 15000 10000
ð2Þ
300 R² = 0.6642
200
100
5000 RDTF RDDF
0
0
5
10
R² = 0.6736
0 0
5000
15
DILUENT RATE (%) Fig. 4. The effect of diluent type and rate on the viscosity of DGEBA.
10000
15000
20000
VISCOSITY (cp)
RDMF TDMAP
TETA
Log. (TDMAP)
Log. (TETA)
Fig. 5. The relationship between flow rates of mortars and viscosities of the epoxies used with or without diluent.
374
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
600
FLOW RATE (%)
500 400 300 200 100 0 NO DILUENT
RDMF
RDDF
RDTF
DILUENT TYPE TMDA
MXDA
TETA
CTMDA
CTMDA+AC
IPDA
IPDA+AC
TDMAP
Fig. 6. Variation of flow rate of DGEBA mortars with stoichiometric amount of hardener and 10% diluent.
mortars are close to each other, although the latter hardener has a lower viscosity, indicating that besides viscosity, the molecular interactions between the hardeners, resin, and diluents (if any diluent exists) are also important. The mortars with TDMAP come next, and finally, the lowest flow rates are obtained for TETA-containing mortars, which is in contradiction with the rule that a lower viscosity leads to a higher flow. The logarithmic additive rule of Eq. (1) was applied to the epoxy resin-hardener binary mixtures without accelerators (assuming G12 = 0). The values for viscosities and molecular fractions obtained from Tables 1 and 2, were inserted into Eq. (1) to predict the viscosities of the resin-hardener mixtures and the following sequence was obtained:
TMDA ðMW : 326 cpÞ < MXDA ð385 cpÞ < IPDA ð540 cpÞ < CTMDA ð2289 cpÞ < TETA ð3161 cpÞ < TDMAP ð5727 cpÞ where the mixtures have been named according to the name of the hardener used. When the flow rates of the mortars are considered, the following sequence is found.
than the one with TETA, although the viscosity of TETA is lower than TDMAP, which is once again due to the latter having pendant methyl groups. In Fig. 6, the effects of diluents on the flow rates of mortars are compared for each hardener type; in general, the addition of a diluent improved the workability. However, for the hardeners TETA and TDMAP, the addition of the diluent RDMF, and for CTMDA and IPDA, the addition of RDDF results in the highest observed flow. For the remaining two hardeners (TMDA and MXDA), RDMF and RDDF lead to similar performances and show higher workability than RDTF. The addition of 30% and 10% BA as an accelerator in CTMDA and IPDA-added mixtures increases the flows of the mortars for all the diluent types used. The viscosity of RDTF is the highest among the tested diluents and therefore, the lowest flows are obtained for RDTF-added mortars among all the hardeners tested. The effects of resin type, as well as hardeners TETA and TDMAP on the flow rates of plain and RDMF-added mortars are compared in Fig. 7. The mixtures with TETA exhibit lower flows than those with TDMAP for all resin types and diluent rates, which, as explained above, is due to the presence of pendant methyl groups in TDMAP. The addition of 5% of diluent into TETA-containing mortars enhances the flow by about 100%, but further increase of the diluent by 5% results in almost no change. On the other hand, increasing the diluent further to 15% causes the flow to increase by 200% with respect to the plain mortar. Similarly, the viscosity and shear yield properties of DGEBA with diamine-based hardener and vinyl cyclohexane dioxide diluent were investigated and it was found that addition 5% of diluent significantly increased the rheological properties [8]. However, further increasing the diluent concentration by 5% had no effect on the rheological properties. In contrast, when the diluent rate increased from 10% to 25%, the workability of the mixture once again increased considerably. The effect of resin type is not significant for mixtures with TETA as seen in Fig. 7, and close level flows were obtained for each diluent rate. Fig. 7 also shows the variation of workability of TDMAP mixtures. It can be seen that for plain mixtures, a lower viscosity of the resin results in a higher flow of the mortar. However, mixtures containing an added diluent do not follow this trend; the
MXDA TMDA < CTMDA < IPDA < TDMAP < TETA
400
FLOW RATE (%)
When the sequence of mixtures in increasing order of viscosities is compared with that arranged in increasing order of flows of the mortars prepared with the same mixtures, it is seen that the mixtures with the two lowest viscosities have similar flows (Fig. 6). For the other mixtures, the mixture which had the third lowest viscosity, is found to be in the fourth position if arranged in increasing order of flow; the same trend is observed for the fifth in the viscosity list, which is the sixth in the flow rate list. In the epoxy mixtures with added hardener, the viscosity of the molecules is not the only parameter affecting the viscosity of the mixtures, but the shape of the hardener is also found to be important. For example, the pendant methyl groups on the diluents hinder the interactions between adjacent resin molecules, which reduces the flow strength [26]; this effect is similar to that due to solvent molecules occupying the intermolecular space of a resin resulting in weaker inter-chain interactions [4]. When the flowaugmenting properties of the hardeners are compared (Fig. 6), although according to Grunberg Eq. (1) (assuming G12 = 0), the mixture with CTMDA should have a higher viscosity than IPDA, these two mixtures have similar flow behavior, which may be due to the presence of pendant methyl groups on both the hardeners (Fig. 1). Similarly, the mortar with TDMAP has a higher flow
500
300
200
100
DGEBA+TDMAP
DGEBF+TDMAP
DGEBAF+TDMAP
DGEBA+TETA
DGEBF+TETA
DGEBAF+TETA
0 0
5
10
15
DILUENT RDMF RATE (%) Fig. 7. Effects of epoxy and hardener types on variation of flow rate of mortars with RDMF rate.
375
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
POT LIFE (min)
250 200 150 100 50 0
NO DILUENT
10% RDMF
10% RDDF
10% RDTF
Fig. 8. Pot lives of DGEBA mixtures prepared with stoichiometric amounts of hardeners and 10% diluents.
mortar prepared with the resin DGEBF, has the worst workability, even when having the lowest viscosity, showing that the interaction between the epoxy type, hardener and diluent also influences the workability. 3.3. Pot life Fig. 8 shows the effects of different hardeners, in stoichiometric amounts, on the pot lives of DGEBA and mixtures prepared by adding 10% of different diluents. Fig. 8 reveals that the hardener triethylene tetra-amine (TETA), being an aliphatic amine and having a functionality of six (Fig. 1), exhibits the shortest pot life. The addition of 10% of a diluent only slightly increases the pot life. However, for all the diluents tested, the pot life remains under 1 h. Another aliphatic polyamine hardener tested is trimethyl hexamethylene diamine (TMDA), which shows a slightly longer pot life than TETA (Fig. 8); a similar behavior was reported by Brostow et al. [27]. Owing to the presence of an aromatic ring in its backbone, meta-xylenediamine (MXDA) (Fig. 1) behaves as a cycloaliphatic amine and shows properties, which are in between aliphatic and aromatic amines. Fig. 8 reveals that the mixtures with MXDA have longer pot lives as compared to those prepared with aliphatic amines. It is known that the reactivity of amines toward aromatic glycidyl ether resins is related to their nucleophilic nature [28] and decreases in the order:
30% of benzyl alcohol reduces the pot life to less than 1 h, except for the mixture containing RDTF as the diluent, in which case, the pot life is slightly longer than 1 h. The last hardener tested was tris (dimethylaminomethyl) phenol (TDMAP), a tertiary amine (Fig. 1), which is also used as an accelerator for epoxy curing agents [27]. Fig. 8 shows that TDMAP added DGEBA-diluent mixtures have pot lives, which are slightly longer than those with TMDA and shorter than those containing MXDA; the pot lives are shorter than 1 h, except for the mixture with RDDF diluent. The test results in Fig. 8 reveal that the addition of 10% of monofunctional RDMF did not change the pot lives of the DGEBAhardener-diluent mixtures significantly, with the exception of CTMDA. Similarly, it was reported that when the monofunctional glycidyl ether of eugenol (GE) was used at 5% as a reactive diluent for the diglycidyl ether of diphenolate n-pentyl ester, the gel time was increased only by 11% [17]. It was therefore necessary to use GE at a concentration of 15% and above to obtain significant delays. On the other hand, for mixtures with CTMDA, the pot life is delayed by more than 4 h for both RDMF and RDDF additions; however, this mixture already has a pot life of 169 min without any diluents. Fig. 8 also shows that difunctional RDDF is more effective in delaying pot life when compared to RDMF. On the contrary, trifunctional RDTF either slightly increased, or decreased the pot life, the decrease being most probably owing to its higher functionality. When resin-hardener systems without diluent are compared, the longest pot life is observed for the DGEBA resin, followed by DGEBAF, with DGEBF having the shortest pot life for both the hardeners TDMAP and TETA (Fig. 9). Fig. 9 also shows that TDMAP increases pot lives to a greater extent than TETA for the three resin types tested, showing the importance of having six functionalities of TETA. For the mixtures with RDMF diluent, it can be observed that in general, a higher diluent content leads to a longer pot life. For a diluent content of 15%, the same trend is observed as for the diluent-free sample, with a small discrepancy between DGEBAF and DGEBF resins-TDMAP systems; the latter shows a slightly longer pot life than the former. For the diluent contents 5 and 10%, DGEBAF has the longest pot lives for both hardener types; followed by DGEBA, whereas, DGEBF had the shortest pot life. Figs. 10 and 11 show, respectively, the effect of the type of reactive diluent on the pot lives of the different epoxy resins tested for
140
aliphatic amines > cycloaliphatic amines > aromatic amines: 120
100
POT LIFE (min)
Another cycloaliphatic amine used is isophorone diamine (IPDA), which increases the pot life to a much greater extent than MXDA, as shown in Fig. 8. The increased pot life of IPDA is partly due to steric hindrance generated by the aromatic ring (Fig. 1) [29] but is also due to the reduced chain flexibility resulting from the rigid aromatic ring [26]. Accelerators can be used to reduce the delay in pot life. In this study, 10% benzyl alcohol is utilized to accelerate the crosslinking reaction [13,29], whereby the pot life is decreased to less than 1 h (Fig. 8). Trimethyl hexamethylene diamine, a cyanoethylated mixture of isomers (CTMDA), an aliphatic amine, was also used as a hardener. Cyanoethylation of amines consists in the replacement of active hydrogen atoms by CN groups, which reduces their electron donating ability. Therefore, the opening of the oxirane group of the epoxy resin is inhibited and the system is rendered less reactive [30]. Although trimethyl hexamethylene diamine is a hardener with a fast-pot-life, as can be seen in Fig. 8, cyanoethylation seems to increase the pot life to a large extent, making it the slowest among the hardeners tested in this study. To shorten the pot life, benzyl alcohol is added as an accelerator; thus, the addition of
80
60
40
20
DGEBA+TDMAP
DGEBAF+TDMAP
DGEBF+TDMAP
DGEBA+TETA
DGEBAF+TETA
DGEBF+TETA
0 0
5
10
15
DILUENT RDMF RATE (%) Fig. 9. Effects of epoxy and hardener types on variation of pot lives of mixtures with RDMF rate.
376
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
90 80
POT LIFE (min)
70
2.
60 50
DGEBA
40
DGEBAF
30
DGEBF
20
3.
10 0 0%
10% RDMF
10% RDDF
10% RDTF
Fig. 10. The effect of 10% diluent addition on the pot lives of epoxy-TDMAP mixtures.
4. 80 70
5.
POT LIFE (min)
60 50 DGEBA
40
DGEBAF
30
DGEBF
20
6.
10 0 0%
10% RDMF
10% RDDF
10% RDTF
Fig. 11. The effect of 10% diluent addition on the pot lives of epoxy-TETA mixtures.
the hardeners TDMAP and TETA. In both cases, the amount of diluent is the same (10%). Fig. 10 reveals that for the hardener TDMAP, the addition of any of the diluents used in this study prolongs the pot life for the three epoxy types with only one exception; 10% RDTF addition slightly shortens the pot life of DGEBA resin. Among the systems containing epoxies with hardener TDMAP, the longest pot life is obtained for the DGEBA-RDDF system. When the diluent types are compared, it can be seen that the diluents RDMF and RDDF delay hardening to a greater extent than the diluent RDTF. A comparison of Figs. 10 and 11 reveals that for diluent free systems, pot lives are longer when the hardener TDMAP was employed than when the hardener TETA was used, due to the six functionality of TETA as mentioned above. For the TETA added systems, the addition of 10% RDMF increases the pot life to a greater extent than the other diluents for the three resins tested, as demonstrated in Fig. 11. On the other hand, unlike systems containing hardener TDMAP, the additions of RDDF into DGEBAF, and RDTF into DGEBF, result in a slight decrease in pot lives for systems with TETA as the hardener. Our test results thus demonstrate that pot lives are influenced by the nature of the hardeners and diluents used, as well as by the type of epoxy resins as expected. 4. Conclusions The following conclusions can be drawn from this study: 1. Among the epoxy-reactive diluent mixtures, the highest viscosities were obtained with DGEBA and the lowest with the DGEBF, showing a trend similar to the viscosities of the neat resins. The addition of 5% diluent caused a significant decrease in viscosities; however, increasing the diluent content further did not change the viscosity as much as the first case. The low-
est viscosities are observed for mixtures with RDMF, and RDTF showed the highest viscosities in general, in concordance with the viscosities of the diluents used. The flow properties of mortars were correlated with the viscosities of the epoxy- (with or without) diluent mixtures included in these mortars. The addition of hardeners into the epoxy mortars improved the flow rates; the increase in flow changed from 100% to a maximum of 280%, by just altering the type of hardener used. The viscosity of the hardener affected the flow of epoxy mortars; as expected, the lower the viscosity of a hardener, higher was the flow of the mortar. However, not only the viscosity of a hardener, but also its shape influenced the flow behavior of epoxy mortars. Inclusion of a diluent into mortars improved the workability, and the flow increased directly with the diluent content. Among the six hardeners tested, triethylene tetraamine (TETA) led to the shortest pot life, most probably because TETA is an aliphatic amine having a functionality of six. Aliphatic amines showed shorter pot lives than cycloaliphatic amines in general. When the type of epoxy resins used is considered in epoxyTDMAP or epoxy-TETA systems, the longest pot life belongs to DGEBA, DGEBAF follows it and DGEBF has the shortest pot life. The addition of RDMF diluent in the epoxy-hardener systems cause delay in pot life for all epoxy types tested, and the higher the RDMF content the longer is the pot life in general. For epoxy-TDMAP system, addition of a diluent at 10% rate increased the pot lives for all epoxy and diluent types used, except DGEBA-RDTF system, in which the pot life is slightly reduced, probably due to the high functionality of RDTF. On the other hand, for TETA added mortars, RDMF is found as the most effective diluent in delaying the pot lives of all epoxy types tested.
Acknowledgement The authors wish to thank Ms. Nazmiye Parlak, the director of R&D Department of SIKA Construction Chemicals, Turkey. References [1] D. Van Gemert, A. Beeldens, Evolution in modeling microstructure formation in polymer-cement concrete, in: M.H. Ozkul, H.N. Atahan, U.A. Dogan, B. Pekmezci, O. Sengul (Eds.), Proceedings of 7th Asian Congress on Polymers in Concrete (ASPIC), Istanbul, 2012, pp. 59–73. [2] J.R.M. d’Almeida, S.N. Monteiro, The influence of the amount of hardener on the tensile mechanical behavior of an epoxy system, Polym. Adv. Technol. 9 (1998) 216–221. [3] S. Montserrat, G. Andreu, P. Cortes, Y. Calventus, P. Colomer, J.M. Hutchinson, J. Malek, Addition of a reactive diluent to a catalyzed epoxy-anhydride system. 1. Influence on the cure kinetics, J. Appl. Polym. Sci. 61 (1996) 1663–1674. [4] J.W. Qian, Y.M. Miao, L. Zhang, H.L. Chen, Influence of viscosity slope coefficient of CA and its blends in dilute solutions on permeation flux of their films for MeOH/MTBE mixture, J. Membr. Sci. 203 (1–2) (2002) 167–173. [5] S.J Shaw, Additives and modifiers for epoxy resins, in Chemistry and Technology of Epoxy Resins, in: B. Ellis (Ed.), Springer Science + Business Media, Dordrecht 1993, pp. 117–143. [6] J. Guo, W. Guo, R. Xie, Y. Li, Y. Yang, Y., Influence of the additives on the workability and mechanic properties of epoxy mortar, Adv. Mater. Res. 168– 170 (2011) 2003–2007. [7] F. Mustatä, I. Bicu, C.N. Cascaval, Rheological and thermal behaviour of an epoxy resin modified with reactive diluents, J. Polym. Eng. 17 (6) (1997) 491–506. [8] M. Villanueva, I. Fraga, J.A. Rodriguez-Anon, J. Proupin-Castineiras, Study of the influence of a reactive diluent on the rheological properties of an epoxydiamine system, J. Therm. Anal. Calorim. 98 (2009) 521–525. [9] H. Harani, S. Fellahi, M. Bakar, Toughening of epoxy resin using Hydroxylterminated polyesters, J. Appl. Polym. Sci. 71 (1999) 29–38. [10] P.L. Teh, M. Mariatti, H.M. Akil, Effect of ethanol as diluent on the properties of mineral silica filled epoxy composites, J. Compos. Mater. 42 (2008) 129–142.
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377 [11] M.R. Loos, L.A.F. Coelho, S.H. Pezzin, S.C. Amico, The Effect of Acetone Addition on the Properties of Epoxy, Polímer.: Ciênc. Tecnol. 18 (1) (2008) 76–80. [12] E. Ozeren, Effects of Epoxy Resin and Hardener Type on Durability and Mechanical Properties of Epoxy Mixes, M. S. Thesis, Istanbul Technical University, Polym. Sci. Technol. Program, 2005. [13] B.L. Burton, Acceleration of Amine-Cured Epoxy Resin Systems, Huntsman Corp, 2013. [14] Chemical Book, http://www.chemicalbook.com/ChemicalProductProperty_ EN_CB51 23061.htm (Accessed 20.02.2017). [15] Sigma Aldrich Product Catalog, http://www.sigmaaldrich.com/catalog/ product/aldrich/475726?lang=en®ion=TR (Accessed 20.02.2017). [16] ISO 10364:2007(E) Structural adhesives – Determination of the pot life (working life) of multi-component adhesives ISO (the International Organization for Standardization), International Standard. [17] A. Maiorana, L. Yue, I. Manas-Zloczower, R. Gross, Structure–property relationships of a bio-based reactive diluent in a bio-based epoxy resin, J. Appl. Polym. Sci. 43635 (2016) (1 of 8). [18] L. Grunberg, A.H. Nissan, Mixture law for viscosity, Nature (London) 164 (1949) 799–800. [19] V. Dumitrescu, O. Pântea, Viscosities of binary mixtures of toluene with butan1-ol and 2-methylpropan-2-ol, J. Serb. Chem. Soc. 70 (11) 1313–1323. [20] G.P. Dubey, P. Kaur, Thermodynamic, transport and excess properties of 2butoxy ethanol +1-alkanol (C6, C8, C10) at different temperatures, J. Chem. Thermodyn. 64 (2013) 239–248. [21] M.A. Boyle, C.J. Martin, J.D. Neuner, in: ASM Handbook, Vol. 21, Composites, ASM International, Material Park, Ohio, 2001, pp. 79–89.
377
[22] I.K. Hong, Y.S. Yoon, S.B. Lee, Selection of thinner for epoxy type resins for neon transformer housing, J. Ind. Eng. Chem. 18 (2012) 1997–2003. [23] L.B. Kandyrin, P.V. Surikov, V.N. Kuleznev, Influence of the viscosity of filled epoxy resins on their spreading on a horizontal surface, J. Eng. Phys. Thermophys. 76 (3) (2003) 568–571. [24] A. Bouvet, E. Ghorbel, R. Bennacer, The mini-conical slump flow test: analysis and numerical study, Cem. Concr. Res. 40 (10) (2010) 1517–1523. [25] J.E. Wallevik, Relationship between the Bingham parameters and slump, Cem. Concr. Res. 36 (2006) 1214–1221. [26] F.F. De Nograro, P. Guerrero, M.A. Corcuera, I. Mondragon, Effect of chemical structure of hardener on curing evolution and on the dynamic mechanical behavior of epoxy resins, J. Appl. Polym. Sci. 56 (1995) 177–192. [27] W. Brostow, S.H. Goodman, J. Wahrmund, Epoxies, H. Dodiuk, S.H. Goodman (Eds.) in: Handbook of Thermoset Plastics, Chap. 8, Elsevier Science, ProQuest Ebook Central, 2013, pp. 191–252. [28] H.Q. Pham, M.J. Marks, Epoxy resins, in: Kirk Othmer (Ed.), Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 10, John Wiley and Sons Inc., 1999, pp. 347–471. [29] W.R. Ashcroft, Curing agents for epoxy resins, in: B. Ellis (Ed.), Chemistry and Technology of Epoxy Resins, Chap, 2, Springer Science + Business Media, Dordrecht, 1993, pp. 37–71. [30] G.S. Mukherjee, Evaluation of processing temperature in the production of fibre reinforced epoxy composites, J. Therm. Anal. Calorim. 108 (2012) 947– 950.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effects of epoxy, hardener, and diluent types on the workability of epoxy mixtures Eren Ozeren Ozgul, M. Hulusi Ozkul ⇑ Istanbul Technical University, Faculty of Civil Engineering, Maslak 34469, Istanbul, Turkey
h i g h l i g h t s The addition of 5% diluent caused a decrease in viscosities between 52 and 72%. Glycidylether C12-C14 alcohol is the most effective diluent in reducing the viscosities of epoxy-diluent mixtures. There is a correlation between the flow of the mortars and the viscosities of the epoxy-diluent mixtures. Altering the type of the hardener used in epoxy mortars may increase the flow rate up to 2.8-fold. The higher the functionality of a hardener the shorter is the pot life of epoxy-hardener-diluent mixtures.
a r t i c l e
i n f o
Article history: Received 21 August 2017 Received in revised form 28 September 2017 Accepted 2 October 2017
Keywords: Polymer concrete Epoxy mortar Viscosity Workability Flow Pot life
a b s t r a c t Three types of epoxy resins, namely, diglycidyl ether of bisphenol A and F, and a third obtained by mixing the two resins, and three types of reactive diluents, based on glycidylether, were used to form epoxy mixtures. The effect of diluent content up to 15 wt% on the viscosities of the three epoxies was investigated. The flow properties of mortars prepared with epoxy, hardener, and diluent, as a binder, were compared with respect to the viscosity and molecular weight of the epoxies and diluents, as well as the viscosity and shape of the hardeners used. Six types of hardeners, including four aliphatic and two based on cycloaliphatic amine, were used in the production of mortars. Furthermore, the effects of the type of epoxy, content, type and functionality of diluents, and the functionality and molecular structure of hardeners, on the pot lives of the mixtures were examined. In this study, it is found that not only the viscosities of resin and diluents but also viscosities and molecular structures of hardener and accelerator are effective on the workability properties of epoxy mortars. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction The term ‘‘Concrete-Polymer Composite” (C-PC) is commonly used to define concretes containing a polymer as a binder, or as part of a binder [1]. The former type of concrete, known as a ‘‘polymer concrete” (PC), contains a low viscosity resin as the polymer and does not include any cement. The latter type, named as ‘‘polymer cement concrete” (PCC), includes a polymer, usually in the form of latex or emulsion, which serves as a modifier and also as a part of the Portland cement binding system. A third type of C-PC, namely, ‘‘polymer impregnated concrete” (PIC), can be added to this list. Epoxy resins are widely used in civil engineering applications. In view of their high resistance to abrasion, epoxy mortars and concretes can be used as surface coatings in bridges, ⇑ Corresponding author. E-mail addresses: [email protected] (E. Ozeren Ozgul), [email protected] (M.H. Ozkul). https://doi.org/10.1016/j.conbuildmat.2017.10.008 0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
pavements, and in industrial floors. Due to their outstanding durability, epoxy resins can be used in industrial buildings, underground applications, and in constructions on seashores. Another important area of application of epoxy resins is in repairing reinforced concrete structures for either filling concrete cracks or patching damaged concrete, due to the excellent adhesion of epoxy to concrete surfaces. In addition, epoxy resins also find applications in corrosion repairing. However, epoxies can pose certain health hazards and care should be taken while using them. Diglycidyl ether of bisphenol A (DGEBA), based on epichlorohydrin, is the most widely used epoxy (Fig. 1), whose viscosity depends on the number of repeating units (n) and the molecular weight. The diglycidyl ether of bisphenol F (DGEBF) has a similar molecular structure to DGEBA, but with the difference that the two methyl groups attached to the carbon located between the benzene groups are replaced by H atoms. F-type epoxies have lower molecular weight than A-type and show lower viscosities.
370
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
For this reason, A and F types are mixed to lower the viscosity of the mixture and also to prevent the crystallization of DGEBA. In the pre-polymerized state, epoxy resins are in the liquid state and a crosslinking process is necessary to convert them into a thermoset polymer. A and F type epoxy resins contain epoxy groups at the end of the molecule (two functionalities) consisting of an oxygen atom bonded to two carbon atoms so as to form a threemembered ring. The strained character of the CAO bond and the high electron affinity of O make the epoxy group highly unstable [2], which means that a number of systems can open the ring to attach to the epoxy resin molecules. Molecules which can react with epoxy groups are known both as hardeners and curing or crosslinking agents. Hardener molecules enable the resin molecules to connect to each other to form a three-dimensional network. There are many types of hardeners, but the most important ones are classified into aliphatic and aromatic amines, anhydrides, and polyamides. Usually, hardener molecules are much smaller than resin molecules and therefore, have low viscosities. The addition of a hardener into a resin lowers the viscosity of the mixture to different extents depending on the viscosity and shape of the hardener and its stoichiometric amount used in the mixture. In some cases, it may be necessary to modify the viscosity of an epoxy resin by using diluents to improve its processability [3]. Diluents reduce the viscosity of epoxy mixtures by weakening the interaction between the resin molecules [4]. There are two main types of diluents, namely, non-reactive and reactive. Non-reactive diluents, which include aromatic hydrocarbons such as toluene or xylene, dibutyl phthalate, styrene and various phenolic compounds [5], do not participate in the reaction between the epoxy and the hardener. Reactive diluents can be classified into epoxy-based and those which are not based on epoxides. Among the reactive diluents, those with epoxy groups significantly influence the properties of epoxy resins. They can chemically bind to the resin and form a part of the network following crosslinking. For this reason, they are less volatile and therefore, more environmentally friendly than non-reactive diluents. Nonvolatile reactive diluents have been used to increase the workability of epoxy mortars, where a fivefold increase in fluidity was reported when 15% of a diluent was added [6]. Epoxidized soybean oil (20% ratio), was added as a reactive diluent to modify DGEBA, and the rheological properties were measured using a rheometer [7]. Test results showed that when the concentration of the oil in the mixture increased, the consistency index (zero shear viscosity) decreased, indicating improved flow properties. In another study, the reactive diluent vinyl cyclohexane dioxide at concentrations up to 25% was used in epoxy-amine systems and a reduction in both apparent viscosity and yield stress was reported [8]. To improve the impact strength of epoxy resins, polyol based reactive diluents were added [9], which, besides increasing the toughness, decreased viscosity, extended pot life and improved wetting properties. Ethanol was used as a diluent in an epoxy system, which allowed the maximum silica filler content to be increased from 40% to 60% [10]. Loos et al. [11] performed rheological measurements on DGEBA resin, using acetone as a diluent and showed that the addition of 10% acetone reduced the viscosity by 50%, which resulted in a better dispersion of nanofillers. In this study, three types of epoxy resins, six amine type hardeners (four aliphatic and two cycloaliphatic), and three reactive diluents (glycidyl ether-based) were used [12]. In the first part of the study, the viscosity of epoxy-diluent mixtures and workability properties of epoxy-diluent-hardener-sand mortars were investigated and the results were presented in this paper. In the second part of the study, the hardened state properties (compressive and flexural strengths, adherence strength to the concrete surface and
chemical resistance) of the same mortars have been explored and the results will be presented in a forthcoming paper.
2. Experimental 2.1. Materials 2.1.1. Epoxy resins The epoxy resins used in this study are diglycidyl ether of bisphenol A (DGEBA) (Fig. 1) and diglycidyl ether of bisphenol F (DGEBF), both of which are commercial products. As a third type of epoxy resin, the mixture of DGEBA and DGEBF (70% from the former and 30% from the latter) (DGEBAF), which is sold commercially, was used. Since DGEBF has lower viscosity relative to DGEBA, the two resins are blended to lower the viscosity; in this way each resin prevents the crystallization of the other one at the room temperature [13]. General information on the epoxy resins is given in Table 1.
2.1.2. Hardeners Six hardeners are used in this study and their properties are given in Table 2. For the hardeners trimethylhexamethylene diamine, cyanoethylated mixture of isomers (trimethylhexane -1,6-diamine) (CTMDA) and isophorone diamine (IPDA), the long pot-life times necessitated the use of benzyl alcohol (density 1.04 g/ml and viscosity@25 °C 8 cp.) as an accelerator (Fig. 1).
2.1.3. Reactive diluents The derivatives of glycidyl ether are the most widely used reactive diluents. Three types of glycidyl ether-based reactive diluents are employed, namely, a monofunctional reactive diluent (RDMF) (glycidylether C12-C14 alcohol), a difunctional reactive diluent (RDDF) (1,6 Hexanediol diglycidylether), and a trifunctional reactive diluent (RDTF) (trimethylopropane glycidylether) (Table 3 and Fig. 1).
2.1.4. Aggregate The aggregate mixture used in the study consists of 82% silica sand (density: 2.631 g/cm3) in size in the range 0.075–4 mm and 18% of fine silica (density: 2.67 g/cm3) with size <100 lm.
2.1.5. Resin mixtures In the mortar formulations, silica sand, fine silica and epoxy resin binders were used. Epoxy resin binder consists of an epoxy, a reactive diluent (if exists), stoichiometric amount hardener for both epoxy and diluent, and an accelerator (if exists). When a diluent is used, the amount of epoxy is reduced with the same amount of diluent. The quantities of hardeners were determined from stoichiometry calculations and the amounts of each hardener calculated for DGEBA are listed in Table 2; adjustments from these values were made for the other epoxies and diluents. For hardeners CTMDA or IPDA, pot–life times were very long, and hence, benzyl alcohol was used in 30 and 10% with respect to epoxy (and diluent if exists) as the accelerator. The resin binder/aggregate ratio was determined to have at least a flow rate of 100 ± 10%. In all mortar mixtures, epoxy resin binder, sand and fine silica contents were 15.1%, 69.6% and 15.3% by mass, respectively.
2.2. Testing procedures The tests were carried out at a temperature and R.H of 23 ± 2 °C and 60 ± 5%, respectively and all the materials used were conditioned for 24 h before testing.
2.2.1. Viscosity measurement Viscosity measurements were carried out using a Brookfield viscometer on mixtures of resin and diluent.
2.2.2. Flow measurement The flow of fresh mortar was determined using a cone of height 60 mm, a bottom surface diameter of 100 mm, and a top surface diameter of 70 mm.
2.2.3. Pot life Pot life is the time required for the epoxy resin-hardener-diluent mixtures to become rigid. For the mixtures containing CTMDA or IPDA, an accelerator (benzyl alcohol) was used as an additive. Pot life tests were carried out on 100 g of resin (with or without diluent) mixed with a stoichiometric amount of the hardener at room temperature in accordance with ISO 10364: 2007(E) [16]. The pot life of the mixture was measured from the amount of time between when it was first mixed, up to the point when the mixture could not be spread by hand.
371
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
Epoxy resin (DGEBA)
Hardeners
Meta-xylenediamine (MXDA) Trimethyl hexamethylene diamine (TMDA)
Triethylene tetraamine (TETA) Isophorone diamine (IPDA)
Accelerator (Benzyl alcohol)
Tris (dimethylaminomethyl) phenol (TDMAP) Reactive diluents
Hexanediol diglycidylether [14]
Trimethylopropane glycidylether [15]
C12-C14 Aliphatic glycidylether [14] Fig. 1. Molecular structures of epoxy, hardeners, accelerator and diluents used. (See above-mentioned references for further information).
Table 1 Properties of resins used in the study. Resin
Density (g/ml)
Viscosity (cp.)
Epoxy eq. weight (g/val)
Diglycidyl ether of bisphenol A (DGEBA) Diglycidyl ether of bisphenol F (DGEBF) A mixture of %70 DGEBA and %30 DGEBF (DGEBAF)
1.16 1.19 1.17
16700 5850 7500
194.9 171.2 182.5
3. Test results and discussion 3.1. Viscosity Fig. 2 shows the variation of viscosity of resins (DGEBA, DGEBF and DGEBAF) containing different amounts of the RDMF diluent. As
expected, viscosities drop with increase in diluent substitution ratio. The highest viscosities are observed for the DGEBA mixtures and the lowest for DGEBF-containing mixtures, similar to viscosities of the neat resins. The addition of 5% of diluent caused a 52– 72% decrease in viscosities with respect to the diluent-free samples; however, further increase in diluent rates to 10 and 15% does
372
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
Table 2 Properties of hardeners and stoichiometric quantities for 100 g of DGEBA. Hardener
Type
Viscosity (cp)
Density (g/ml)
Mol. Weight
Quantity (g)
Trimethyl hexamethylene diamine (TMDA) Meta-xylenediamine (MXDA) Triethylene tetraamine (TETA) Trimethyl hexamethylene diamine (cyanoethylated mixture of isomers) (CTMDA) Isophorone diamine (IPDA) Tris (dimethylaminomethyl) phenol (TDMAP)
Aliphatic Aliphatic Aliphatic Aliphatic Cycloaliphatic Accelerated cycloaliphatic
6 8 26 30 16 150
0.87 1.05 0.98 0.92 0.92 1.01
158.3 136.2 146.2 316.4 170.3 265.4
40 34 13 37.2 42.6 20
Table 3 Properties of reactive diluents used in the study. Reactive diluent
Density (g/ml)
Viscosity (cp.)
Epoxy eq. weight (g/val)
Glycidylether C12-C14 alcohol (monofunctional) (RDMF) 1,6 Hexanediol diglycidylether (difunctional) (RDDF) Trimethylopropane glycidylether (trifunctional) (RDTF)
0.89 1.07 1.15
10 23 149
289.4 142.9 142.0
4000
18000
3500 16000
VISCOSITY (cp)
12000 10000 DGEBA+RDMF DGEBF+RDMF DGEBAF+RDMF
8000 6000
VISCOSITY (cp)
3000 14000
2500 2000
DGEBA DGEBAF DGEBF
1500 1000 500 0
4000
RDMF
RDDF
RDTF
DILUENT TYPE
2000 0
Fig. 3. The effect of 10% diluent addition on the viscosities of epoxies.
0
5
10
15
20
DILUENT RATE (%) Fig. 2. Variation of viscosity of epoxies with diluent (RDMF) rate.
not proportionally decrease viscosities. Instead, the viscosities are reduced by 81–88% and 91–95%, respectively. Similar to our observations, it was reported that the addition of 5% glycidyl ether of eugenol reduced the viscosity of the diglycidyl ether of diphenolate n-pentyl ester (epoxy) by 38%, whereas, the viscosity was reduced by 70% and 100% when the diluent concentration was increased to 10% and 15%, respectively [17]. In addition, the differences between the viscosities of the mixtures decrease for higher diluent rates. For a diluent content of 5%, the difference between the maximum and minimum viscosity is 25.8% of the initial difference between the viscosities of the resins without diluents; however, this difference is reduced to 3.3% when the diluent content is increased to 15%. For liquid mixtures, many equations have been proposed to predict the viscosity of a mixture in terms of its components. One of them is the Grunberg and Nissan [18] equation, which is as follows:
ln gmix ¼ x1 lng1 þ x2 lng2 þ x1 x2 G12
ð1Þ
where g is the viscosity, x is the molar fraction of the components in the mixture and G12 is a temperature dependent constant. When the viscosities of the epoxies and diluents as given in Table 1 and 3, and the test results obtained after addition of diluents as shown in Fig. 2 are inserted in Eq. (1), negative G12 values are obtained for all
the binary systems tested, indicating that dispersive forces are predominant [19,20]. Under the assumption G12 = 0, Eq. (1) overestimates the viscosity of the resin-diluent mixtures; however, the trend in the viscosities of the mixtures can be estimated as shown in Fig. 2. The viscosities of resin mixtures with 10% added diluents are compared in Fig. 3. This figure shows that RDMF is the most effective diluent for reducing the viscosities of mixtures followed by RDDF. The highest viscosities are observed for mixtures containing RDTF for the three resins tested. The epoxy-diluent mixtures in increasing order of viscosity in terms of the diluent used are as follows:
glycidylether C12 C14 alcoholðRDMFÞ < 1; 6 Hexanediol diglycidyletherðRDDFÞ < trimethylopropane glycidyletherðRDTFÞ The molecular weight of a diluent also influences the viscosity of an epoxy-diluent mixture with the lower molecular weight leading to lower viscosity [21]. The molecular weights of the diluents used in this study are as follows (Table 3).
1; 6 Hexanediol diglycidylether ð285:7 gÞ < glycidylether C12 C14 alcohol ð289:4 gÞ < trimethylopropane glycidylether ð426:1 gÞ The viscosities of these diluents in increasing order are (Table 3):
373
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
glycidylether C12 C14 alcohol < 1;6 Hexanediol diglycidylether < trimethylopropane glycidylether Although the molecular weight of hexanediol diglycidylether is slightly smaller than that of C12-C14 glycidylether, the latter diluent lowers the viscosity of the resins to a greater extent than the former, indicating that the viscosity of a diluent is more important than its molecular weight in reducing the viscosity of the epoxy mixture. Hong et al. [22] tested four different reactive diluents in bisphenol-A epoxy resin with the aim to reduce the viscosity. Mixtures of the epoxy with 10% added diluents showed the following trend in viscosities (the mixtures are denoted by the names of the diluent used):
butyl glycidyl ether < styrene oxides < phenyl glycidyl ether < aliphatic glycidyl ether
to the viscosity and decreases exponentially, with an exponent of 1/8 [23]. Most of the studies relating the flow to rheological parameters were carried out on Portland cement based mixtures. Bouvet et al. [24] concluded from flow tests that for cement pastes, a lower viscosity led to a higher spread and proposed an expression to calculate the final spread as a function of the rheological parameters such as yield stress and viscosity. However, there are other studies showing that there is only a weak relationship between the spread and the viscosity [25]. In the slump flow test, a cone is first filled with a mixture, after which, it is removed vertically, while the diameter of the spread material is measured. A comparison of the flow of different mixtures is done by comparing the change in flow area (A) with respect to the initial area (A0). Flow rate:
w¼
If these diluents are arranged in order of increasing molecular weights, the only modifications in the above trend are in the positions of the butyl glycidyl ether and styrene oxides. Further comparison reveals that the viscosities of the diluents follow the same trend as the viscosities of the corresponding epoxy-diluent mixtures, showing that the diluent viscosity may be a more important parameter than the molecular weight in reducing the viscosity of the epoxy resin. When the different types of resins are compared for each diluent (Fig. 3), the highest viscosities are observed for DGEBA mixtures, with DGEBF having the lowest viscosities, which again shows a trend similar to the viscosities of the pure resins. The change in viscosities with the diluent content is given in Fig. 4 for DGEBA epoxy-diluent mixtures. As expected, the viscosities of the mixtures decrease with increase in diluents rate for the three reactive diluents used. A comparison of the diluents shows that RDMF is the most efficient with the lowest viscosity, except for the case of 5% diluent, where the viscosity is slightly greater than that of RDDF. Test results given in Fig. 4 reveal a 66–81% reduction in the viscosity of DGEBA for 5% diluent addition, between 78 and 88% for 10%, and finally, from 83% up to 95% when the diluent amount was 15%. The lowest viscosities are obtained for mixtures with RDMF, followed by RDDF, while mixtures with RDTF show the highest viscosities; this trend is similar to that observed for the viscosities of the individual diluents. 3.2. Flow
A A0 A0
The flow rates of epoxies (all the three types used in this study)hardeners TETA or TDMAP-with or without reactive diluents (all the three types used)-sand mortars just after the mixing process are given in Fig. 5 together with the viscosities of the same mixtures without the addition of any hardener or sand (from Figs. 2 and 3). Fig. 5 shows that there is a correlation between the flows of the mortars and the viscosities of the epoxies-with or without added diluent mixtures. Fig. 5 also reveals that the addition of TDMAP into the mortars leads to a greater improvement in the flow rate as compared to TETA. The flow properties of mortars prepared with DGEBA epoxy resin with 10% of the three reactive diluents together with the six different hardeners at different rates (Table 2), are displayed in Fig. 6. The viscosity reducing effect (diluent effect) of the hardeners on the flow behavior of the mortars can also be seen in the same figure. By only altering the type of the hardener used, a 100–280% (a 2.8-fold increase) change in the flow rates of the plain mortars without diluents can be achieved. Epoxy mortars, with CTMDA and IPDA as hardeners, with addition of 30% and 10% of accelerator BA, respectively, are also shown in the same figure. Fig. 6 reveals that the highest flows are observed for mixtures with CTMDA as the hardener and 30% benzyl alcohol (BA) as the accelerator, followed by IPDA hardener with 10% BA, demonstrating the utility of BA as a diluent in view of its low viscosity of 5.47 cp. Following mortars containing added accelerator, those containing hardeners MXDA and TMDA, which have the lowest viscosities among all the hardeners tested, show the highest flow rates among the remaining mortars. The flow rates of CTMDA and IPDA-added
Many studies have attempted to relate the slump or spread of mortars and concretes to their rheological parameters, measured using rheometers. Flow tests have been carried out on epoxy resin-diluent mixtures filled with sand and sand flour and it was reported that the spread of the mortars is inversely proportional
500
400
FLOW RATE (%)
VISCOSITY (cp)
20000 15000 10000
ð2Þ
300 R² = 0.6642
200
100
5000 RDTF RDDF
0
0
5
10
R² = 0.6736
0 0
5000
15
DILUENT RATE (%) Fig. 4. The effect of diluent type and rate on the viscosity of DGEBA.
10000
15000
20000
VISCOSITY (cp)
RDMF TDMAP
TETA
Log. (TDMAP)
Log. (TETA)
Fig. 5. The relationship between flow rates of mortars and viscosities of the epoxies used with or without diluent.
374
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
600
FLOW RATE (%)
500 400 300 200 100 0 NO DILUENT
RDMF
RDDF
RDTF
DILUENT TYPE TMDA
MXDA
TETA
CTMDA
CTMDA+AC
IPDA
IPDA+AC
TDMAP
Fig. 6. Variation of flow rate of DGEBA mortars with stoichiometric amount of hardener and 10% diluent.
mortars are close to each other, although the latter hardener has a lower viscosity, indicating that besides viscosity, the molecular interactions between the hardeners, resin, and diluents (if any diluent exists) are also important. The mortars with TDMAP come next, and finally, the lowest flow rates are obtained for TETA-containing mortars, which is in contradiction with the rule that a lower viscosity leads to a higher flow. The logarithmic additive rule of Eq. (1) was applied to the epoxy resin-hardener binary mixtures without accelerators (assuming G12 = 0). The values for viscosities and molecular fractions obtained from Tables 1 and 2, were inserted into Eq. (1) to predict the viscosities of the resin-hardener mixtures and the following sequence was obtained:
TMDA ðMW : 326 cpÞ < MXDA ð385 cpÞ < IPDA ð540 cpÞ < CTMDA ð2289 cpÞ < TETA ð3161 cpÞ < TDMAP ð5727 cpÞ where the mixtures have been named according to the name of the hardener used. When the flow rates of the mortars are considered, the following sequence is found.
than the one with TETA, although the viscosity of TETA is lower than TDMAP, which is once again due to the latter having pendant methyl groups. In Fig. 6, the effects of diluents on the flow rates of mortars are compared for each hardener type; in general, the addition of a diluent improved the workability. However, for the hardeners TETA and TDMAP, the addition of the diluent RDMF, and for CTMDA and IPDA, the addition of RDDF results in the highest observed flow. For the remaining two hardeners (TMDA and MXDA), RDMF and RDDF lead to similar performances and show higher workability than RDTF. The addition of 30% and 10% BA as an accelerator in CTMDA and IPDA-added mixtures increases the flows of the mortars for all the diluent types used. The viscosity of RDTF is the highest among the tested diluents and therefore, the lowest flows are obtained for RDTF-added mortars among all the hardeners tested. The effects of resin type, as well as hardeners TETA and TDMAP on the flow rates of plain and RDMF-added mortars are compared in Fig. 7. The mixtures with TETA exhibit lower flows than those with TDMAP for all resin types and diluent rates, which, as explained above, is due to the presence of pendant methyl groups in TDMAP. The addition of 5% of diluent into TETA-containing mortars enhances the flow by about 100%, but further increase of the diluent by 5% results in almost no change. On the other hand, increasing the diluent further to 15% causes the flow to increase by 200% with respect to the plain mortar. Similarly, the viscosity and shear yield properties of DGEBA with diamine-based hardener and vinyl cyclohexane dioxide diluent were investigated and it was found that addition 5% of diluent significantly increased the rheological properties [8]. However, further increasing the diluent concentration by 5% had no effect on the rheological properties. In contrast, when the diluent rate increased from 10% to 25%, the workability of the mixture once again increased considerably. The effect of resin type is not significant for mixtures with TETA as seen in Fig. 7, and close level flows were obtained for each diluent rate. Fig. 7 also shows the variation of workability of TDMAP mixtures. It can be seen that for plain mixtures, a lower viscosity of the resin results in a higher flow of the mortar. However, mixtures containing an added diluent do not follow this trend; the
MXDA TMDA < CTMDA < IPDA < TDMAP < TETA
400
FLOW RATE (%)
When the sequence of mixtures in increasing order of viscosities is compared with that arranged in increasing order of flows of the mortars prepared with the same mixtures, it is seen that the mixtures with the two lowest viscosities have similar flows (Fig. 6). For the other mixtures, the mixture which had the third lowest viscosity, is found to be in the fourth position if arranged in increasing order of flow; the same trend is observed for the fifth in the viscosity list, which is the sixth in the flow rate list. In the epoxy mixtures with added hardener, the viscosity of the molecules is not the only parameter affecting the viscosity of the mixtures, but the shape of the hardener is also found to be important. For example, the pendant methyl groups on the diluents hinder the interactions between adjacent resin molecules, which reduces the flow strength [26]; this effect is similar to that due to solvent molecules occupying the intermolecular space of a resin resulting in weaker inter-chain interactions [4]. When the flowaugmenting properties of the hardeners are compared (Fig. 6), although according to Grunberg Eq. (1) (assuming G12 = 0), the mixture with CTMDA should have a higher viscosity than IPDA, these two mixtures have similar flow behavior, which may be due to the presence of pendant methyl groups on both the hardeners (Fig. 1). Similarly, the mortar with TDMAP has a higher flow
500
300
200
100
DGEBA+TDMAP
DGEBF+TDMAP
DGEBAF+TDMAP
DGEBA+TETA
DGEBF+TETA
DGEBAF+TETA
0 0
5
10
15
DILUENT RDMF RATE (%) Fig. 7. Effects of epoxy and hardener types on variation of flow rate of mortars with RDMF rate.
375
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
POT LIFE (min)
250 200 150 100 50 0
NO DILUENT
10% RDMF
10% RDDF
10% RDTF
Fig. 8. Pot lives of DGEBA mixtures prepared with stoichiometric amounts of hardeners and 10% diluents.
mortar prepared with the resin DGEBF, has the worst workability, even when having the lowest viscosity, showing that the interaction between the epoxy type, hardener and diluent also influences the workability. 3.3. Pot life Fig. 8 shows the effects of different hardeners, in stoichiometric amounts, on the pot lives of DGEBA and mixtures prepared by adding 10% of different diluents. Fig. 8 reveals that the hardener triethylene tetra-amine (TETA), being an aliphatic amine and having a functionality of six (Fig. 1), exhibits the shortest pot life. The addition of 10% of a diluent only slightly increases the pot life. However, for all the diluents tested, the pot life remains under 1 h. Another aliphatic polyamine hardener tested is trimethyl hexamethylene diamine (TMDA), which shows a slightly longer pot life than TETA (Fig. 8); a similar behavior was reported by Brostow et al. [27]. Owing to the presence of an aromatic ring in its backbone, meta-xylenediamine (MXDA) (Fig. 1) behaves as a cycloaliphatic amine and shows properties, which are in between aliphatic and aromatic amines. Fig. 8 reveals that the mixtures with MXDA have longer pot lives as compared to those prepared with aliphatic amines. It is known that the reactivity of amines toward aromatic glycidyl ether resins is related to their nucleophilic nature [28] and decreases in the order:
30% of benzyl alcohol reduces the pot life to less than 1 h, except for the mixture containing RDTF as the diluent, in which case, the pot life is slightly longer than 1 h. The last hardener tested was tris (dimethylaminomethyl) phenol (TDMAP), a tertiary amine (Fig. 1), which is also used as an accelerator for epoxy curing agents [27]. Fig. 8 shows that TDMAP added DGEBA-diluent mixtures have pot lives, which are slightly longer than those with TMDA and shorter than those containing MXDA; the pot lives are shorter than 1 h, except for the mixture with RDDF diluent. The test results in Fig. 8 reveal that the addition of 10% of monofunctional RDMF did not change the pot lives of the DGEBAhardener-diluent mixtures significantly, with the exception of CTMDA. Similarly, it was reported that when the monofunctional glycidyl ether of eugenol (GE) was used at 5% as a reactive diluent for the diglycidyl ether of diphenolate n-pentyl ester, the gel time was increased only by 11% [17]. It was therefore necessary to use GE at a concentration of 15% and above to obtain significant delays. On the other hand, for mixtures with CTMDA, the pot life is delayed by more than 4 h for both RDMF and RDDF additions; however, this mixture already has a pot life of 169 min without any diluents. Fig. 8 also shows that difunctional RDDF is more effective in delaying pot life when compared to RDMF. On the contrary, trifunctional RDTF either slightly increased, or decreased the pot life, the decrease being most probably owing to its higher functionality. When resin-hardener systems without diluent are compared, the longest pot life is observed for the DGEBA resin, followed by DGEBAF, with DGEBF having the shortest pot life for both the hardeners TDMAP and TETA (Fig. 9). Fig. 9 also shows that TDMAP increases pot lives to a greater extent than TETA for the three resin types tested, showing the importance of having six functionalities of TETA. For the mixtures with RDMF diluent, it can be observed that in general, a higher diluent content leads to a longer pot life. For a diluent content of 15%, the same trend is observed as for the diluent-free sample, with a small discrepancy between DGEBAF and DGEBF resins-TDMAP systems; the latter shows a slightly longer pot life than the former. For the diluent contents 5 and 10%, DGEBAF has the longest pot lives for both hardener types; followed by DGEBA, whereas, DGEBF had the shortest pot life. Figs. 10 and 11 show, respectively, the effect of the type of reactive diluent on the pot lives of the different epoxy resins tested for
140
aliphatic amines > cycloaliphatic amines > aromatic amines: 120
100
POT LIFE (min)
Another cycloaliphatic amine used is isophorone diamine (IPDA), which increases the pot life to a much greater extent than MXDA, as shown in Fig. 8. The increased pot life of IPDA is partly due to steric hindrance generated by the aromatic ring (Fig. 1) [29] but is also due to the reduced chain flexibility resulting from the rigid aromatic ring [26]. Accelerators can be used to reduce the delay in pot life. In this study, 10% benzyl alcohol is utilized to accelerate the crosslinking reaction [13,29], whereby the pot life is decreased to less than 1 h (Fig. 8). Trimethyl hexamethylene diamine, a cyanoethylated mixture of isomers (CTMDA), an aliphatic amine, was also used as a hardener. Cyanoethylation of amines consists in the replacement of active hydrogen atoms by CN groups, which reduces their electron donating ability. Therefore, the opening of the oxirane group of the epoxy resin is inhibited and the system is rendered less reactive [30]. Although trimethyl hexamethylene diamine is a hardener with a fast-pot-life, as can be seen in Fig. 8, cyanoethylation seems to increase the pot life to a large extent, making it the slowest among the hardeners tested in this study. To shorten the pot life, benzyl alcohol is added as an accelerator; thus, the addition of
80
60
40
20
DGEBA+TDMAP
DGEBAF+TDMAP
DGEBF+TDMAP
DGEBA+TETA
DGEBAF+TETA
DGEBF+TETA
0 0
5
10
15
DILUENT RDMF RATE (%) Fig. 9. Effects of epoxy and hardener types on variation of pot lives of mixtures with RDMF rate.
376
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377
90 80
POT LIFE (min)
70
2.
60 50
DGEBA
40
DGEBAF
30
DGEBF
20
3.
10 0 0%
10% RDMF
10% RDDF
10% RDTF
Fig. 10. The effect of 10% diluent addition on the pot lives of epoxy-TDMAP mixtures.
4. 80 70
5.
POT LIFE (min)
60 50 DGEBA
40
DGEBAF
30
DGEBF
20
6.
10 0 0%
10% RDMF
10% RDDF
10% RDTF
Fig. 11. The effect of 10% diluent addition on the pot lives of epoxy-TETA mixtures.
the hardeners TDMAP and TETA. In both cases, the amount of diluent is the same (10%). Fig. 10 reveals that for the hardener TDMAP, the addition of any of the diluents used in this study prolongs the pot life for the three epoxy types with only one exception; 10% RDTF addition slightly shortens the pot life of DGEBA resin. Among the systems containing epoxies with hardener TDMAP, the longest pot life is obtained for the DGEBA-RDDF system. When the diluent types are compared, it can be seen that the diluents RDMF and RDDF delay hardening to a greater extent than the diluent RDTF. A comparison of Figs. 10 and 11 reveals that for diluent free systems, pot lives are longer when the hardener TDMAP was employed than when the hardener TETA was used, due to the six functionality of TETA as mentioned above. For the TETA added systems, the addition of 10% RDMF increases the pot life to a greater extent than the other diluents for the three resins tested, as demonstrated in Fig. 11. On the other hand, unlike systems containing hardener TDMAP, the additions of RDDF into DGEBAF, and RDTF into DGEBF, result in a slight decrease in pot lives for systems with TETA as the hardener. Our test results thus demonstrate that pot lives are influenced by the nature of the hardeners and diluents used, as well as by the type of epoxy resins as expected. 4. Conclusions The following conclusions can be drawn from this study: 1. Among the epoxy-reactive diluent mixtures, the highest viscosities were obtained with DGEBA and the lowest with the DGEBF, showing a trend similar to the viscosities of the neat resins. The addition of 5% diluent caused a significant decrease in viscosities; however, increasing the diluent content further did not change the viscosity as much as the first case. The low-
est viscosities are observed for mixtures with RDMF, and RDTF showed the highest viscosities in general, in concordance with the viscosities of the diluents used. The flow properties of mortars were correlated with the viscosities of the epoxy- (with or without) diluent mixtures included in these mortars. The addition of hardeners into the epoxy mortars improved the flow rates; the increase in flow changed from 100% to a maximum of 280%, by just altering the type of hardener used. The viscosity of the hardener affected the flow of epoxy mortars; as expected, the lower the viscosity of a hardener, higher was the flow of the mortar. However, not only the viscosity of a hardener, but also its shape influenced the flow behavior of epoxy mortars. Inclusion of a diluent into mortars improved the workability, and the flow increased directly with the diluent content. Among the six hardeners tested, triethylene tetraamine (TETA) led to the shortest pot life, most probably because TETA is an aliphatic amine having a functionality of six. Aliphatic amines showed shorter pot lives than cycloaliphatic amines in general. When the type of epoxy resins used is considered in epoxyTDMAP or epoxy-TETA systems, the longest pot life belongs to DGEBA, DGEBAF follows it and DGEBF has the shortest pot life. The addition of RDMF diluent in the epoxy-hardener systems cause delay in pot life for all epoxy types tested, and the higher the RDMF content the longer is the pot life in general. For epoxy-TDMAP system, addition of a diluent at 10% rate increased the pot lives for all epoxy and diluent types used, except DGEBA-RDTF system, in which the pot life is slightly reduced, probably due to the high functionality of RDTF. On the other hand, for TETA added mortars, RDMF is found as the most effective diluent in delaying the pot lives of all epoxy types tested.
Acknowledgement The authors wish to thank Ms. Nazmiye Parlak, the director of R&D Department of SIKA Construction Chemicals, Turkey. References [1] D. Van Gemert, A. Beeldens, Evolution in modeling microstructure formation in polymer-cement concrete, in: M.H. Ozkul, H.N. Atahan, U.A. Dogan, B. Pekmezci, O. Sengul (Eds.), Proceedings of 7th Asian Congress on Polymers in Concrete (ASPIC), Istanbul, 2012, pp. 59–73. [2] J.R.M. d’Almeida, S.N. Monteiro, The influence of the amount of hardener on the tensile mechanical behavior of an epoxy system, Polym. Adv. Technol. 9 (1998) 216–221. [3] S. Montserrat, G. Andreu, P. Cortes, Y. Calventus, P. Colomer, J.M. Hutchinson, J. Malek, Addition of a reactive diluent to a catalyzed epoxy-anhydride system. 1. Influence on the cure kinetics, J. Appl. Polym. Sci. 61 (1996) 1663–1674. [4] J.W. Qian, Y.M. Miao, L. Zhang, H.L. Chen, Influence of viscosity slope coefficient of CA and its blends in dilute solutions on permeation flux of their films for MeOH/MTBE mixture, J. Membr. Sci. 203 (1–2) (2002) 167–173. [5] S.J Shaw, Additives and modifiers for epoxy resins, in Chemistry and Technology of Epoxy Resins, in: B. Ellis (Ed.), Springer Science + Business Media, Dordrecht 1993, pp. 117–143. [6] J. Guo, W. Guo, R. Xie, Y. Li, Y. Yang, Y., Influence of the additives on the workability and mechanic properties of epoxy mortar, Adv. Mater. Res. 168– 170 (2011) 2003–2007. [7] F. Mustatä, I. Bicu, C.N. Cascaval, Rheological and thermal behaviour of an epoxy resin modified with reactive diluents, J. Polym. Eng. 17 (6) (1997) 491–506. [8] M. Villanueva, I. Fraga, J.A. Rodriguez-Anon, J. Proupin-Castineiras, Study of the influence of a reactive diluent on the rheological properties of an epoxydiamine system, J. Therm. Anal. Calorim. 98 (2009) 521–525. [9] H. Harani, S. Fellahi, M. Bakar, Toughening of epoxy resin using Hydroxylterminated polyesters, J. Appl. Polym. Sci. 71 (1999) 29–38. [10] P.L. Teh, M. Mariatti, H.M. Akil, Effect of ethanol as diluent on the properties of mineral silica filled epoxy composites, J. Compos. Mater. 42 (2008) 129–142.
E. Ozeren Ozgul, M.H. Ozkul / Construction and Building Materials 158 (2018) 369–377 [11] M.R. Loos, L.A.F. Coelho, S.H. Pezzin, S.C. Amico, The Effect of Acetone Addition on the Properties of Epoxy, Polímer.: Ciênc. Tecnol. 18 (1) (2008) 76–80. [12] E. Ozeren, Effects of Epoxy Resin and Hardener Type on Durability and Mechanical Properties of Epoxy Mixes, M. S. Thesis, Istanbul Technical University, Polym. Sci. Technol. Program, 2005. [13] B.L. Burton, Acceleration of Amine-Cured Epoxy Resin Systems, Huntsman Corp, 2013. [14] Chemical Book, http://www.chemicalbook.com/ChemicalProductProperty_ EN_CB51 23061.htm (Accessed 20.02.2017). [15] Sigma Aldrich Product Catalog, http://www.sigmaaldrich.com/catalog/ product/aldrich/475726?lang=en®ion=TR (Accessed 20.02.2017). [16] ISO 10364:2007(E) Structural adhesives – Determination of the pot life (working life) of multi-component adhesives ISO (the International Organization for Standardization), International Standard. [17] A. Maiorana, L. Yue, I. Manas-Zloczower, R. Gross, Structure–property relationships of a bio-based reactive diluent in a bio-based epoxy resin, J. Appl. Polym. Sci. 43635 (2016) (1 of 8). [18] L. Grunberg, A.H. Nissan, Mixture law for viscosity, Nature (London) 164 (1949) 799–800. [19] V. Dumitrescu, O. Pântea, Viscosities of binary mixtures of toluene with butan1-ol and 2-methylpropan-2-ol, J. Serb. Chem. Soc. 70 (11) 1313–1323. [20] G.P. Dubey, P. Kaur, Thermodynamic, transport and excess properties of 2butoxy ethanol +1-alkanol (C6, C8, C10) at different temperatures, J. Chem. Thermodyn. 64 (2013) 239–248. [21] M.A. Boyle, C.J. Martin, J.D. Neuner, in: ASM Handbook, Vol. 21, Composites, ASM International, Material Park, Ohio, 2001, pp. 79–89.
377
[22] I.K. Hong, Y.S. Yoon, S.B. Lee, Selection of thinner for epoxy type resins for neon transformer housing, J. Ind. Eng. Chem. 18 (2012) 1997–2003. [23] L.B. Kandyrin, P.V. Surikov, V.N. Kuleznev, Influence of the viscosity of filled epoxy resins on their spreading on a horizontal surface, J. Eng. Phys. Thermophys. 76 (3) (2003) 568–571. [24] A. Bouvet, E. Ghorbel, R. Bennacer, The mini-conical slump flow test: analysis and numerical study, Cem. Concr. Res. 40 (10) (2010) 1517–1523. [25] J.E. Wallevik, Relationship between the Bingham parameters and slump, Cem. Concr. Res. 36 (2006) 1214–1221. [26] F.F. De Nograro, P. Guerrero, M.A. Corcuera, I. Mondragon, Effect of chemical structure of hardener on curing evolution and on the dynamic mechanical behavior of epoxy resins, J. Appl. Polym. Sci. 56 (1995) 177–192. [27] W. Brostow, S.H. Goodman, J. Wahrmund, Epoxies, H. Dodiuk, S.H. Goodman (Eds.) in: Handbook of Thermoset Plastics, Chap. 8, Elsevier Science, ProQuest Ebook Central, 2013, pp. 191–252. [28] H.Q. Pham, M.J. Marks, Epoxy resins, in: Kirk Othmer (Ed.), Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 10, John Wiley and Sons Inc., 1999, pp. 347–471. [29] W.R. Ashcroft, Curing agents for epoxy resins, in: B. Ellis (Ed.), Chemistry and Technology of Epoxy Resins, Chap, 2, Springer Science + Business Media, Dordrecht, 1993, pp. 37–71. [30] G.S. Mukherjee, Evaluation of processing temperature in the production of fibre reinforced epoxy composites, J. Therm. Anal. Calorim. 108 (2012) 947– 950.