Transparent nanocomposite coatings based on epoxy and layered double hydroxide: Nonisothermal cure kinetics and viscoelastic behavior assessments

Progress in Organic Coatings 113 (2017) 126–135

Contents lists available at ScienceDirect

Progress in Organic Coatings journal homepage: www.elsevier.com/locate/porgcoat

Transparent nanocomposite coatings based on epoxy and layered double hydroxide: Nonisothermal cure kinetics and viscoelastic behavior assessments

MARK

⁎

Hadi Rastina, Mohammad Reza Saebb, , Milad Nonahala, Meisam Shabanianc, Henri Vahabid, Krzysztof Formelae, Xavier Gabrionf, Farzad Seidig, Payam Zarrintaja, Morteza Ganjaee Sarih, Pascal Laheurtei a

School of Chemical Engineering, College of Engineering, University of Tehran, P.O. Box: 11155-4563, Tehran, Iran Department of Resin and Additives, Institute for Color Science and Technology, P.O. Box 16765-654, Tehran, Iran Faculty of Chemistry and Petrochemical Engineering, Standard Research Institute (SRI), P.O. Box 31745-139, Karaj, Iran d Université de Lorraine, Laboratoire MOPS E.A. 4423, Metz, F-57070, France e Department of Polymer Technology, Faculty of Chemistry, Gdansk University of Technology, Gdansk, Poland f FEMTO-ST Institute, Department of Applied Mechanics, UMR 6176, University Bourgogne-Franche-Comté, CNRS/ENSMM/UFC/UTBM, F-25000, Besançon, France g Department of Materials Science and Engineering, School of Molecular Science and Engineering, Vidyasirimedhi Institute of Science and Technology, 21210, Rayong, Thailand h Department of Nanomaterials and Nanocoatings, Institute for Color Science and Technology, P.O. Box 16765-654, Tehran, Iran i Universite de Lorraine, Laboratoire LEM3 UMR 7239, Metz, F-57045, France b c

A R T I C L E I N F O

A B S T R A C T

Keywords: Nonisothermal cure kinetics Epoxy Layered double hydroxide Dynamic mechanical analysis Transparent coatings

Layered double hydroxide (LDH) has a particular place in clay family because of its flame retardant action. The nanoplatelet-like structure of LDH makes possible development of polymer composites with cationic or anionic nature structures in which macromolecules are positioned in between nanoplatelet galleries. In this work, neat epoxy and its transparent nanocomposite coatings with sodium dodecylbenzene sulfonate (SDBS)-modified LDHs; Mg-Al and Zn-Al LDHs, were prepared and their cure kinetics and viscoelastic behavior were tracked through nonisothermal calorimetric and dynamic mechanical analyses. The higher progression of crosslinking in the epoxy network was observed for epoxy/Zn-Al LDH nanocomposites, while activation energy of cure reaction took a higher value for Mg-Al LDH-incorporated systems. Moreover, epoxy/Mg-Al LDH system revealed higher value of storage modulus and glass transition temperature thanks to larger galleries of Mg-Al nanoplatelets. Network formation in the presence of SDBS-modified Zn-Al LDH nanoplatelets was facilitated due to the action of Zn metal as an adduct with a lone-pair of oxygen atom of epoxy leading to an enhanced epoxy ring-opening. Viscoelastic behavior of transparent coatings containing Zn-Al LDH and Mg-Al LDH was studied through temperature-sweep test at various frequencies to compare the results of calorimetric and thermo-mechanical analyses.

1. Introduction Polymer composites comprising well-dispersed layered inorganic nanostructures have taken a particular place in academia and the industry alike due to their action on electrical, thermal, rheological, and mechanical properties at low loading levels [1–4]. Among these layered inorganic nanostructure fillers are layered double hydroxides (LDHs) with their unique characteristics such as high aspect ratio, tunable charge density and high value of anion exchange capacity [5,6]. The LDH has been recognized as anionic clay with brucite-like structure in

⁎

which anions were found to be positioned in between clay galleries neutralizing the positive charge of cations. Facile synthesis route of LDH with various metal cations, intercalated anions, and desirable surface modification [7,8] makes it a strong candidate for a broad range of application ranging from water treatment [9,10] and hydrothermal reactor [11] to drug and gene delivery [12] and biosensor[13]. Accordingly, coprecipitation, electrogeneration, polyol route and sol-gel techniques were appeared as efficient routes in tailoring the structure of LDH for having diverse chemical compositions, sizes and shapes [5]. The majority of the commonly used polymers from high

Corresponding author. E-mail addresses: [email protected], [email protected] (M.R. Saeb).

http://dx.doi.org/10.1016/j.porgcoat.2017.09.003 Received 19 August 2017; Received in revised form 2 September 2017; Accepted 5 September 2017 0300-9440/ © 2017 Elsevier B.V. All rights reserved.

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

Fig. 1. Schematic of chemical structure of the Mg-Al LDH and Zn-Al LDH clays used in this work.

flammability because of their petroleum-based nature. Incorporation of flame retardant agents into the polymeric matrices could to a large extent make them suitable for applications where heat generation was significant. Recent studies suggest that LDH can be employed as a halogen-free flame retardant in polymers [14–18]. A series of LDH/ polymer nanocomposites have accordingly been developed in view of flame retardant action of LDH [19–21]. However, hydrophilic nature of such nanoplatelet-like structures necessitates modification of LDH to enhance the degree of compatibility of LDH with hydrophobic polymeric matrices [22,23]. The presence of voluminous organic anions embedded in between interlayer spaces could to an acceptable extent solve this challenge by exfoliation of LDH nanosheets [15,24]. For instance, Wang et al. [25] reported an improved flame resistance properties for polypropylene upon introduction of organomodified magnesium–aluminum (Mg-Al) LDH, as evidenced by drop in specific heat release rate, heat release capacity and total heat release. Epoxy thermosetting resins are well-known for their superior young modulus, low shrinkage after curing and excellent corrosion resistance [26]. Therefore, they have been widely exploited in adhesives, coatings and composites applications [27–29]. On the other hand, high flammability and brittleness to some extent narrowed application of epoxy in engineering fields [30]. Zammarano et al. [31] reported an improved flame resistance performance for epoxy composites comprising LDH compared to counterparts having montmorillonite clay because of intrinsic properties of LDH. Becker et al. [32] revealed an improved performance of epoxy-based nanocomposites containing various loadings of LDH comparing their burning rate with neat epoxy. They reported that the highest mechanical properties was with composites containing 1 wt.% LDH. It can be concluded that though LDH enhanced flame resistance of epoxy, it could deteriorate the mechanical and crosslinking reactions, which necessitates careful formulation of epoxy with LDH. There is no need to emphasize for experts in the field of thermosetting coatings that almost all properties of epoxy-based composite coatings are pertinent to promotion of network formation in the course of curing reaction [33–35]. Analysis of the cure kinetics offers worthwhile evidence about chemical reactions taking place during cure process towards understanding of structure-property relationship [36]. In a previous study, the role of graphene oxide (GO) nanoplatelets modified with aliphatic amines on cure kinetics and fracture behavior of epoxy resin was discussed [37]. It was observed that modification of

GO increases the crosslinking density by association of amine functional groups attached to the GO platelet surface to epoxy ring opening reactions. Similar trends were observed when using nanofillers having different geometrical shapes in epoxy like multi-walled carbon nanotubes modified with amine precursor [38,39] and Fe3O4 magnetic nanoparticles modified with β-cyclodextrin [40]. Two kinds of LDH, i.e. Mg-Al and zinc–aluminum (Zn-Al) LDH were employed here to prepare epoxy-based transparent nanocomposite coatings. It was known that both kinds of LDH can improve thermal stability and flame retardancy of polymers [21,41–43]. The cure kinetics and viscoelastic behavior of transparent epoxy films containing Mg-Al and Zn-Al nanoplatelets modified with sodium dodecylbenzene sulfonate (SDBS) was comprehensively discussed in this work. Cure kinetics was studied by nonisothermal calorimetric analyses under different heating rates using an in-house code that finds parameters of model-free and model-fitting equations. Analyses were interpreted based on activation energy as a function of the extent of cure in the presence and absence of Mg-Al and Zn-Al LDHs. Viscoelastic properties of transparent coatings were analyzed through mechanical and thermomechanical tests to find glass transition, heat of cure, and activation energies of neat and LDH-incorporated epoxy coatings. The role of MgAl and Zn-Al LDH on the aforementioned properties was uncovered through a comparative view of results as well as mechanistic description of network formation in the presence of nanoplatelets.

2. Experimental 2.1. Materials Epoxy resin with commercial name D3415 was purchased from Sigma-Aldrich Co. (USA). Cycloaliphatic polyamine resin with a viscosity of 55,000 mPa.s (Epikure F205) was provided by Hexion Chemical Co. (USA) and used as amino curing agent. The curing agent was mixed with epoxy with weight ratio of 2:3 with respect to epoxy resin. To synthesize Mg-Al LDH and Zn-Al LDH, all ingredients including magnesium nitrate, zinc nitrate, aluminum nitrate, sodium hydroxide, sodium dodecylbenzene sulfonate (SDBS) were provided by Merck Co. (USA). The ttetrahydrofuran (THF) used for dissolving epoxy and hardener in film formation was provided by Merck (USA) and used as received.P 127

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

2.2. Preparation of transparent epoxy-based nanocomposite coatings

2.3. Measurements

Detailed procedure applied in synthesis of SDBS-modified LDH nanoplatelets of Mg and Zn type can be found in previous works of this group [8,18]. Fig. 1 demonstrates chemical structure of Mg-Al and ZnAl LDH nanoplatelets used for preparation of transparent epoxy nanocomposite films. In order to prepare epoxy-based nanocomposite coatings, 0.1 g of SDBS-modified LDH was added to 142.8 g of thinned epoxy (S.C. = 70 wt.%) to ensure a 0.1 wt.% mixture is obtained. The addition process was performed under high shear mechanical agitation and controlled manually to reduce powder waste by the mixer and the container walls. The mechanical mixing continued for 20 min before the dispersion underwent a complimentary ultrasonic process to accomplish dispersion of the nanoplatelets and enhance the stability of the epoxy mixture. Nevertheless, the agitation process caused formation of some small bubbles. To erase the bubbles, the resulting mixtures were placed in vacuum oven for 2 h. Consequently, to cure the epoxy mixtures, the polyamine hardener was added to the resulting mixtures at room temperature at appropriate ratios. To calculate epoxy to hardener ratio one needs to consider that EEW of the epoxy is 170.2 g/ eq. and the average AHEW of the hardener is 105 g/eq. To obtain a oneto-one stoichiometric network, 170.2 g of pure epoxy was mixed by 105 g of the hardener. The pot-life of such mixture was around 30 min and the mixture was useless after elapsing this time. In the next step, the inferred mixtures were applied on glass substrates by the aid of an elcometer film applicator at 120 μm thickness. The applied films were then put inside an oven adjusted at 70 °C for 2 h and left inside the unplugged oven for 72 h to ensure that the reactions are done and the films are fully cured. The free-stand films were also obtained by peeling the cured films applied on glass. To study viscoelastic behavior of epoxy/LDH nanocomposite coatings, thin films of samples having thickness of ca. 100–150 μm were prepared using solven method. Fig. 2 gives a vivid view of the transparency of the prepared nanocompsoite coatings containing Mg-Al and Zn-Al LDH nanoplatelets modified with SDBS precursor.

Differential scanning calorimetry (DSC) was performed on a Q2000 DSC (TA Instruments USA) to study cure kinetic behavior of the neat and nanocomposites of epoxy resin comprising LDH over the heating range of 0–300 °C. Inspired by previous studies, 5–10 mg of the prepared nanocomposites was placed in an aluminum pan and heated nonisothermally at different heating rates of 5, 10, 15, and 20 °C.min−1. Dynamic mechanical tests were carried out on a BOSE electroforce 3200 with a load capacity of 450N to investigate the dynamic mechanical or viscoelastic properties of the prepared samples over the range of −60 °C–150 °C with heating rate of 3 °C. min−1 under strain amplitude of 0.01%–0.5%. Prior to the test, samples were dried out at 80 °C under nitrogen purge for 12 h to remove their moisture. Three different frequencies of 0.1, 1, and 10 Hz were recognized to be informative for the sake of comparison of viscoelastic properties of nanocomposite coatings with that of neat epoxy. Tensile tests were performed using a Dynamics Mechanical Analysor (DMA) (Bose Electroforce 3200) at ambient temperature. Tests were performed on 3 specimens with a displacement control at a speed of 5 μm/s. The strain was measured with laser extensometer EIR LE05, Young's moduli was determined by linear regression of the stress–strain curves on a range of strain varied between 0 and 0.4%. 3. Results and discussion 3.1. XRD characterization of the synthesized LDH XRD patterns of the SDBS-modified Zn-AL and Mg-Al LDH are compared in Fig. 3. From the XRD patterns it can be recognized that SDBS-modified LDH nanolayers were successfully synthesized through one-step reactions. XRD pattern of the unmodified LDH layers showed maximum basal diffraction (003) at 2Ө = 11.8° [44], while for both synthesized LDH nanolayers the maximum basal diffraction was lower than that of 11.8°. The XRD patterns showed the maximum basal diffraction (003) at 2Ө = 3.7° and 2Ө = 2.9° for Zn–Al and SDBS Mg–Al LDH, respectively. Thus, these patterns obviously confirm successful one-step synthesis of both of SDBS-modified LDH. It should be mentioned that the interlayer distance between Mg-Al LDH was larger in Fig. 2. The appearance of neat epoxy and LDH-incorporated epoxy nanocomposite films for the sake of transparency comparison.

128

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

unimodal peak of curing reaction confirmed the assumption of singlestep kinetic reaction. In order to study the effect of LDH on curing reaction, cure kinetics of the prepared samples were investigated by isoconversional method. The degree of conversion over the curing reaction for thermosetting resin can be simply described according to following relation [33]:

α=

ΔHT ΔH∞

(1)

where ΔHT and ΔH∞ refer to heat of reaction released at certain temperature T and total released heat over the whole reaction, respectively. According to the literature, the rate of curing reaction (dα/dt) is dependent on temperature (k(T)) and reaction model (f(α)) function, as follows: Fig. 3. XRD patterns of the SDBS-modified Zn-Al and Mg-Al LDH.

dα = k (T ) f (α ) dt

view of its lower 2Ө = 2.9° compared to that of Zn-Al with 2Ө = 3.7°. Thus, one may expect lower disturbance by Mg-Al LDH when curing accomplishes in nanocomposites. This needs to be evaluated through DSC analyses discussed later.

(2)

In the case of epoxy resin, reaction rate constant and reaction model can simply be described by the following equations [36]:

E k (T ) = A exp ⎛− a ⎞ ⎝ RT ⎠

(3)

3.2. Cure kinetics investigations

f (α ) = α m (1 − α )n

(4)

Fig. 4 shows DSC thermograms of the prepared neat epoxy and its nanocomposites at four heating rates. The shift in the peak of the curves towards higher temperatures upon increasing the heating rate from 5 to 20 °C.min−1 was somehow expected in view of more kinetic energy being considered for the system at high temperatures [39]. Moreover,

in which A indicates the frequency factor; R is universal gas constant; T is absolute temperature; Ea is the activation energy; m and n are reaction exponents. Based on the Friedman model, activation energy of cure reaction can be determined from the slope of ln(dα/dt) vs. 1/T, as follows:

Fig. 4. DSC thermograms of prepared samples at four different heating rate (β) of 5, 10, 15, and 20 °C.min−1.

129

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

Fig. 5. (a) Plots of ln(dα/dt) vs. 1/T for epoxy resin and (b) resulting activation energy for all samples obtained by Friedman model. Fig. 6. Mechanism proposed for facilitated crosslinking observed for epoxy/Zn-Al LDH systems.

dα E ln ⎛ ⎞ = ln(Af (α )) − a RT ⎝ dt ⎠

for epoxy/Mg-Al LDH system. Thus, there should be another factor contributing to facile curing reaction in epoxy/Zn-Al LDH nanocomposite. It can be concluded that the lower activation energy in epoxy/Zn-Al LDH compared to epoxy/Mg-Al LDH is due to Lewis acid action of Zn. The Zn metal forms an adduct with a lone-pair of oxygen atom of epoxy to assists in ring opening reaction (Fig. 6). Thus, the positive effect of larger galleries in the case of epoxy/Mg-Al LDH system that could lead to an improved interaction between the epoxy and hardener molecules could be somewhat overshadowed by the catalytic effect caused by the Zn. To provide with further evidence, Kissinger–Akahira–Sunose (KAS) isoconversional model was employed to detect conversion-dependent activation energy, as follows [45]:

(5)

Fig. 5 illustrates the evolution of activation energy for all the prepared samples over the whole curing reaction interval. As seen, activation energy has changed over the whole reaction, signifying the complex mechanism of curing reaction. For neat epoxy system, activation energy follows a slight descending order upon increasing the extent of cure. This is due to the autocatalytic nature of the epoxy curing. Addition of inorganic filler causes a significant physical hindrance, which increases the activation energy value, as seen in epoxy/ Mg-Al LDH system. However, introduction of modified Zn-Al LDH into the epoxy resin compensates slightly for the physical hindrance; so that the activation energy takes values very close to the blank epoxy. The chemical structure of the both Mg-Al and Zn-Al LDH are similarly modified with SDBS, but the difference in crosslinking of epoxy in the presence of LDH takes its origin in the nature of Zn and Mg divalent metals acting in an opposite direction. Bearing in mind larger galleries detected for Mg-Al LDH (Fig. 3), one may expect a facile crosslinking

β d ⎛−1n ⎛ T 2i ⎞ ⎞ E ⎝ i ⎠⎠ ⎝ = α 1 R d T

( ) α, i

130

(6)

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

Fig. 7. (a) Plots of ln(β/Tα2) vs. 1/Tαfor epoxy resin and (b) resulting activation energy for all samples obtained by KAS model.

d (1 − α ) ⎤ dα E E 1−α ⎞ − a = (n − m)1n ⎛ Value I = 1n ⎛ ⎞ + a − 1n ⎡ RT dt RT ′ ⎝ α ⎠ ⎝ dt ⎠ ⎦ ⎣ (7)

Table 1 Kinetic parameters of autocatalytic model obtained from KAS isoconversional method. Sample name

Epoxy Epoxy/Zn-Al LDH Epoxy/Mg-Al LDH

Friedman

KAS

ΔH(kJ/ mol)

m

n

lnA (s−1)

m

n

lnA (s−1)

307 325

0.20 0.14

1.46 1.59

4.77 8.92

0.11 0.06

1.56 1.69

6.95 10.75

370

0.12

1.71

11.67

0.06

1.78

13.07

d (1 − α ) ⎤ dα E E Value II = 1n ⎛ ⎞ + a + 1n ⎡ + a RT dt RT ′ ⎝ dt ⎠ ⎦ ⎣ = (n + m)1n(α − α 2) + 21n A

(8)

In fact, the slope of Value I against ln[(1-α)/α] gives (n − m), while the slope and intercept of Value II against ln(α − α2) result in (n + m) and 2lnA. Table 1 summarizes the values of kinetic parameters calculated for the neat epoxy and its nanocomposites on the basis of Friedman and KAS models. According to the statistics, the overall cure reaction order, (m + n) takes a value above unity for all the studied samples, which is a signature of the complexity of crosslinking mechanism. It is accepted that in such a situation, the crosslinking mechanism is nearly multistage. It is to be highlighted that the heat of cure, ΔH values for both epoxy/Mg-Al LDH and epoxy/Zn-Al LDH systems are much higher than that of neat epoxy. Such an increase would be the result of contribution of metal hydroxides existing in the both types of SDBS-modified LDH nanoplatelets with catalytic effects towards epoxy ring opening. By contrast, values of m for nanocomposites are significantly lower than that of neat epoxy, demonstrating less autocatalytic reaction between epoxy and amine hardener in the presence of LDH. The higher value of heat of cure, 370 kJmol−1 observed for epoxy/Mg-Al LDH compared to 325 kJmol−1 for epoxy/Zn-Al LDH suggests that crosslinking has been continued for a longer time interval in the case of former because of its stronger basic action. On the other hand, the values of LnA and Ea corresponding to the epoxy/Zn-Al LDH are lower than the equivalents detected by modeling approach for the epoxy/Mg-Al LDH, corroborating the proficiency of curing reaction in the presence of Zn metal hydroxides, which has been described mechanistically in Fig. 6. Thus, kinetically speaking, introduction of Mg-Al LDH into the epoxy decreased the pace of curing compared to the Zn-Al LDH for which epoxy ring opening was facilitated.

Fig. 8. Storage modulus as a function of temperature for neat epoxy and LDH-incorporated nanocomposite coatings.

The slope of the plot ln(β/Tα2) vs. 1/Tα gives activation energy, as represented in Fig. 7. As seen, the evolution of activation energies over the time span of crosslinking corresponding to blank epoxy and its nanocomposites follow a similar pattern, which put emphasis on the reliability of the trend captured by Friedman model. By substituting Eqs. (3) and (4) into Eq. (2) and some simple mathematical manipulations, m and n values can be obtained from plot of Value I against ln[(1-α)/α] and Value II against ln(α-α2), as follows [46]:

3.3. Viscoelastic behavior assessment Viscoelastic behavior of neat epoxy and composite coatings containing Zn-Al and Mg-Al LDH was studied by temperature-sweep dynamic mechanical measurements at various frequencies ranging from 0.1 Hz to 10 Hz. In this manner, the effect of LDH nanoplatelets on viscoelastic properties of the prepared epoxy-based nanocomposites was uncovered. The frequency of test was changed to calculate the 131

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

epoxy/Mg-Al LDH shows such a transition at higher temperatures. This observation indicates that glass transition temperature (Tg) of coatings containing Mg-Al LDH is meaningfully higher in comparison with the other two. This can be related to positive effects of the filler on the curing reactions, which results in enhanced crosslinking highlighted in previous section. Right after the transition area, there is a rubbery plateau which is magnified in Fig. 8. Higher Eꞌ values for Mg-Al filled epoxy at this region (which shows almost no plateau preferably at terminal zone) also endorse the speculation of increasing crosslink density in the presence of Mg-Al LDH. Fig. 9 shows alteration of loss modulus as a function of temperature for coatings. The peaks in the plots are indicative of the highest damping appeared as Tg, where epoxy/Mg-Al LDH transparent coating shows the highest Tg among studied samples. The very same behavior in both storage and loss moduli (Figs. 9 and 10) was observed different frequencies so that the diagrams were skipped and the quantitative outcomes were presented. Then, the Tg was determined using DMA outcomes at different frequencies for the filled and unfilled epoxies. Fig. 10 shows storage modulus of transparent coatings as a function of temperature at different frequencies. The plots show the evolution of storage modulus as a function of temperature for three different frequencies of solicitation. The storage modulus of ca. 1.9 GPa at 25 °C and 0.1 Hz was increased at low temperature of −20 °C–2.3 GPa and decreased again down to 1.5 GPa at about 70 °C for all studied frequencies. A large decrease was observed when the glass transition region was swept above 90 °C. In this region situated between 80 °C and 100 °C, the rigidity of the sample dropped significantly, as signaled by a storage modulus of ca. 0.2 GPa at around 100 °C. A very similar trend was recognized at higher frequencies of 1 and 10 Hz, but the stiffness increased as the frequency rose. At higher rates of deformation, polymer chains could find sufficient time for relaxation leading to

Fig. 9. Loss modulus diagram as a function of temperature for epoxy, epoxy/Zn-Al LDH and epoxy/Mg-Al LDH.

activation energy and ΔH of three types of transparent nanocomposite coatings. Fig. 8 shows the storage modulus alteration as a function of temperature at deformation rate of 1 Hz for epoxy, epoxy/Zn-Al LDH and epoxy/Mg-Al LDH systems. It is clear that unfilled epoxy shows higher storage modulus (around 15% more) at the glassy region compared to nanocomposite coatings. The glassy region showing an almost two distinct transitions has been intensified for the nanocomposite coatings. This can be attributed to the structural inhomogeneities, even for neat epoxy, as a consequence of assorted chains. The inhomogeneities for the filled samples are primarily originating from the particle flocculation and agglomeration. Across the glass transition area, the modulus starts to rapidly plummet for all the samples, but

Fig. 10. Storage modulus as a function of temperature for (a) epoxy, (b) epoxy/Mg-Al LDH, and (c) epoxy/Zn-Al LDH nanocomposite coatings at different frequencies.

132

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

Fig. 11. Loss modulus as a function of temperature for (a) (a) Epoxy, (b) Epoxy/Mg-Al LDH, and (c) Epoxy/Zn-Al LDH measured at 3 frequencies.

higher storage modulus and shift in Tg towards higher temperatures. The nanocomposite epoxy-based coatings follow the same trend as presented in Fig. 10b and Fig. 10c, except the fact that a noisier trend at glassy region that appeared at low frequencies. This is most probably due to the local flocculation of nanoplatelets that cases different responses to be detected based on forces sensed by the load cell of the device. As the frequency increases, the time needed for breaching the aggregates becomes much larger compared to the relaxation time of the chain and its effect fades away. Fig. 11 shows the loss modulus plots for the unfilled and LDH-incorporated transparent epoxy coatings at different frequencies. The value of Tg for the Epoxy system was ca. 90 °C at frequency of 0.1 (Fig. 11a). An increase of about 2.5 °C in Tg was measured between 0.1 and 10 Hz. The second peak at about 140–145 °C was certainly assigned to the fusion or polymerization temperature. The value of Tg was ca. 120 °C for epoxy/Mg-Al LDH at the frequency of 0.1 (Fig. 11b). A rise of 5 °C in Tg was observed for this sample when the frequency changed between 0.1–10 Hz. The Tg of epoxy/Zn-Al LDH nanocomposite coating was ca. 90 °C at 1 Hz (Fig. 11c). These outcomes are clearly representative of the fact that Epoxy/Mg-Al LDH became stiffer and

Fig. 12. Variation of Tg as measured by the tanδ peak for epoxy/Mg-Al LDH.

Table 2 The Tg at different frequencies for different materials (identified on the peak of loss modulus and the peak of tanδ), calculation of ΔH on the Tg measured on the peak of tan δ. Parameter

ΔH(kJ/mol) tanδ 0.1–10 Hz

Tg (°C) Loss 0.1 Hz

Tg (°C) tanδ

Tg (°C) Loss 1 Hz

Tg (°C) tanδ

Tg (°C) Loss 10 Hz

Tg (°C) tanδ

Epoxy Epoxy/Mg-Al-LDH Epoxy/Zn-AL-LDH

412 584 351

90.8 ± 1.3 121 ± 3.3 90.0 ± 0.1

97.3 ± 0.9 129 ± 2 96.3 ± 0.9

94.3 ± 0.8 125 ± 3.2 93.3 ± 1.2

101 ± 1 135 ± 1 101.3 ± 0.9

97.8 ± 1.3 129 ± 2.5 95.8 ± 1.2

110.7 ± 0.9 140 ± 0.5 112.3 ± 3.7

133

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

the curing results. The activation energy obtained for this sample by DSC measurement was in the same direction higher than that of epoxy/ Zn-Al-LDH sample. It was suggested mechanistically that Zn interacts with epoxide ring and interferes the ring-opening reactions. Apparently, this may look in the favor of curing reactions; however, such interference can act as an inhibitor that prevents amine reactions with the epoxy ring. The outcome of such process is that in the epoxy/Mg-Al LDH although more energy was needed, but the main curing reactions were accomplished, while in the other nanocomposite such reactions were possibly prohibited. From this perspective, higher Tg obtained for Mg-Al incorporated film could be expected. Altogether, the results are expressing that the Mg-Al LDH nanoplatelets do not worsen the crosslinking reactions, in turn they may enhance them through a special mechanism illustrated in Fig. 6. Thus, the higher viscoelastic properties achieved for this sample are a direct result of the enhanced crosslinking. This investigation is followed and accomplished by measuring quasistatic mechanical properties by the aid of tensile testing. Fig. 13 demonstrates the stress-strain curves for Epoxy, epoxy/Mg-Al-LDH and epoxy/Zn-Al-LDH nanocomposite coatings. To quantitatively compare the samples, the elastic modulus, maximum stress and maximum strain of the samples underwent DMA measurements were extracted from the plots (Table 3). It is crystal clear that elastic modulus of epoxy/Mg-AlLDH takes the highest value among the studied samples, which was expected since this sample showed a significantly higher Tg based on DMA measurements. The maximum elongation had its minimum value for epoxy/Mg-Al LDH coating corroborating the high brittleness of this sample. This outcome was somehow seen in DMA as its cross-linking density was higher compared to epoxy/Zn-Al LDH and blank epoxy samples. The maximum stress for this sample was the least most probably during the accumulation of the nanoplatelets inside the epoxy matrix. Nevertheless, such speculation must be verified by electron microscopy techniques, which is the subject of future studies. It is not surprising that epoxy/Zn-Al LDH coating shows more flexibility in view of its more elongation at break, even more than the blank epoxy. Again, this may be logically credited from the speculation that Zn interferes to the ring-opening reactions and prohibits amine interactions and epoxide groups; consequently, crosslink density declines. The obtained Tg from DMA confirms such conclusion as well. Altogether, one can say that addition of Mg-Al LDH nanoplatelets to the epoxy increases the stiffness and decreases flexibility; while Z-Al LDH assists in improving the flexibility even through it is not able to enhance crosslink density and accordingly the stiffness.

Fig. 13. Stress-strain curves for epoxy, epoxy/Mg-Al-LDH and epoxy/Zn-Al-LDH.

Table 3 Quasi-static properties of epoxy, epoxy/Mg-Al-LDH and epoxy/Zn-Al-LDH. Sample

E (MPa)

σmax (MPa)

εmax (%)

Epoxy Epoxy/Mg-Al-LDH Epoxy/Zn-AL-LDH

1960 ± 53 2110 ± 75 2100 ± 150

35.5 ± 1 27.5 ± 1.5 36.6 ± 1.1

2.24 ± 0.13 1.52 ± 0.12 2.28 ± 0.12

shows meaningfully an increased Tg. Such evidences are in good agreement with kinetics study demonstrating that Mg-Al LDH has more accelerating effect on epoxide ring opening in the presence of amine groups compared to the Zn-Al LDH. It can be inferred that in a more crosslinked network and; therefore, an improved dynamic mechanical properties could be expected from Epoxy/Mg-Al LDH coatings. To give more light on the idea that epoxy/Mg-Al LDH coating has a higher crosslinking density, the activation energy and ΔH of nanocomposite coatings were calculated using Goertzen and Kessler approaches [47]. As shown by Goertzen and Kessler, the activation energy of the glass transition relaxation can be estimated by a proportional relation between the slopes of a plot of logarithmic frequency (Hz) versus the reciprocal of absolute Tg (K) removed from tanδ peak. The equation is given in Eq. (9) and the plot of ln(f) versus 1/Tg is given in Fig. 12, which is satisfactorily linear. The horizontal lines represent the standard deviation of the measured Tg.

ΔH = −R

d (1n(f )) d

( ) 1 Tg

4. Conclusion

= (−8.134*10−3)(slope ) In this work, epoxy-based nanocomposite coatings comprising two different kinds of LDH (Mg-Al and Zn-Al) were prepared. Cure kinetic and mechanical behavior of the prepared nanocomposite coatings were captured in terms of nonisothermal DSC and DMA measurements. Investigation on cure kinetics of epoxy/LDH composites shows that MgAl LDH could positively participate in curing reaction, leading to a network having more crosslink density. Moreover, activation energy of Mg-Al LDH, obtained by Friedman and KAS models, was more than the blank epoxy and Zn-Al LDH. The results of DMA supported the ones obtained through nonisothermal calorimetric analyses, corroborating continued crosslinking due to the more basic action of Mg hydroxides, while facilitated crosslinking was credited from Zn-Al LDH incorporated systems thanks to Lewis action of Zn metal contributing to opening of epoxy rings. Evidently, it was found that not only glass transition temperature of epoxy/Mg-Al LDH was considerably higher than the blank epoxy, but also heat of cure reaction, as a fingerprint of crosslink density in thermosetting systems, was escalated upon introduction of Mg-Al LDH. Furthermore, it was found that epoxy/Mg-Al LDH nanocomposite coatings show more stiffness compared to epoxy/ Zn-Al LDH, as detected via quasi-static mechanical tests; so that yield strength and elongation at break of such sample decreased due to the

(9)

For all the samples, the calculation was done according to Eq. (9). For example, the activation energy calculated for epoxy/Mg-Al LDH is represented through Eq. (10).

ΔH = −R

d (1n(f )) d

( ) 1 Tg

= (−8.134*10−3)(−70.348*103) =

584kJ mol

(10)

For the sake of deeper understanding and comparison, the outcomes from the calculations on activation energy along with Tgs obtained from the DMA analyses are given in Table 2. The results given in Table 2 underlines that epoxy/Mg-Al-LDH coating possesses a significantly higher Tg across the whole frequency range, while almost no difference exists between the Tgs measured for the epoxy and epoxy/Zn-Al-LDH samples. Moreover, the calculated ΔH for the epoxy/Mg-Al LDH coating was meaningfully higher compared to the other two, as evidenced by cure kinetic study (Table 1). The results declare that epoxy/Mg-Al LDH epoxy nanocomposite coating needs more energy to enable the matrix chains to gyrate inside their free volumes. To interpret such behavior one should take a deeper look at 134

Progress in Organic Coatings 113 (2017) 126–135

H. Rastin et al.

[24] J. Hu, M. Gan, L. Ma, Z. Li, J. Yan, J. Zhang, Surf. Coat. Technol. 240 (2014) 55. [25] D.-Y. Wang, A. Leuteritz, B. Kutlu, M.A. der Landwehr, D. Jehnichen, U. Wagenknecht, G. Heinrich, J. Alloys Compd. 509 (2011) 3497. [26] H. Gu, C. Ma, J. Gu, J. Guo, X. Yan, J. Huang, Q. Zhang, Z. Guo, J. Mater. Chem. C 4 (2016) 5890. [27] Y. Dong, L. Ma, Q. Zhou, J. Coat. Technol. Res. 10 (2013) 909. [28] M. Jouyandeh, O. Moini Jazani, A.H. Navarchian, M.R. Saeb, J. Reinf. Plast. Compos. 35 (2016) 1685. [29] B. Kaffashi, A. Kaveh, O. Moini Jazani, M.R. Saeb, Polym. Bull. 68 (2012) 2187. [30] T.-Y. Tsai, S.-W. Lu, F.-S. Li, J. Phys. Chem. Solids 69 (2008) 1386. [31] M. Zammarano, M. Franceschi, S. Bellayer, J.W. Gilman, S. Meriani, Polymer 46 (2005) 9314. [32] C.M. Becker, A.D. Gabbardo, F. Wypych, S.C. Amico, Compos. Part A: Appl. Sci. Manuf. 42 (2011) 196. [33] S. Vyazovkin, A.K. Burnham, J.M. Criado, L.A. Pérez-Maqueda, C. Popescu, N. Sbirrazzuoli, Thermochim. Acta 520 (2011) 1. [34] M.R. Saeb, F. Najafi, E. Bakhshandeh, H.A. Khonakdar, M. Mostafaiyan, F. Simon, C. Scheffler, E. Mäder, Chem. Eng. J. 259 (2015) 117. [35] M.R. Saeb, E. Bakhshandeh, H.A. Khonakdar, E. Mäder, C. Scheffler, G. Heinrich, Sci. World J. 2013 (2013). [36] M.R. Saeb, M. Ghaffari, H. Rastin, H.A. Khonakdar, F. Simon, F. Najafi, V. Goodarzi, P. Vijayan, P.D. Puglia, F.H. Asl, K. Formela, RSC Adv. 7 (2017) 2218. [37] M. Nonahal, M.R. Saeb, S. Hassan Jafari, H. Rastin, H.A. Khonakdar, F. Najafi, F. Simon. Polym. Compos., http://dx.doi.org/10.1002/pc.24415. [38] P. Vijayan, P.D. Puglia, H. Rastin, M.R. Saeb, B. Shojaei, K. Formela, Prog. Org. Coat. 108 (2017) 75. [39] M.R. Saeb, H. Rastin, M. Nonahal, M. Ghaffari, A. Jannesari, K. Formela, J. Appl. Polym. Sci. 134 (2017) 45221. [40] M.R. Saeb, H. Rastin, M. Shabanian, M. Ghaffari, G. Bahlakeh, Prog. Org. Coat. 110 (2017) 172. [41] C. Nyambo, D. Wang, C.A. Wilkie, Polym. Adv. Technol. 20 (2009) 332. [42] D.-Y. Wang, A. Leuteritz, Y.-Z. Wang, U. Wagenknecht, G. Heinrich, Polym. Degrad. Stab. 95 (2010) 2474. [43] L. Wang, S. Su, D. Chen, C.A. Wilkie, Polym. Degrad. Stab. 94 (2009) 770. [44] D.-Y. Wang, F.R. Costa, A. Vyalikh, A. Leuteritz, U. Scheler, D. Jehnichen, U. Wagenknecht, L. Häussler, G. Heinrich, Chem. Mater. 21 (2009) 4490. [45] M. Abdalla, D. Dean, P. Robinson, E. Nyairo, Polymer 49 (2008) 3310. [46] T. Zhou, M. Gu, Y. Jin, J. Wang, Polymer 46 (2005) 6174. [47] W.K. Goertzen, M. Kessler, Compos. Part B: Eng. 38 (2007) 1.

more crosslink density, which resulted in a more brittle structure. References [1] F. Kovanda, E. Jindova, B. Doušová, D. Koloušek, J. Pleštil, Z. Sedláková, Acta Geodyn. Geomater. 6 (2009) 111. [2] S. Lv, Y. Yuan, W. Shi, Prog. Org. Coat. 65 (2009) 425. [3] T.S. Kermaniyan, H. Garmabi, M.R. Saeb, R. Boldt, J. Vinyl Addit. Technol. 23 (2017) 35. [4] A.E. Jahromi, H.R.E. Jahromi, F. Hemmati, M.R. Saeb, V. Goodarzi, K. Formela, Compos. Part B: Eng. 90 (2016) 478. [5] Q. Wang, D. O’Hare, Chem. Rev. 112 (2012) 4124. [6] A. Raeisi, M. Shabanian, K. Faghihi, Prog. Org. Coat. 111 (2017) 402. [7] M. Shabanian, H. Ardeshir, S. Haji-Ali, H. Moghanian, M. Hajibeygi, K. Faghihi, H.A. Khonakdar, H. Salimi, Appl. Clay Sci. 123 (2016) 285. [8] M. Hajibeygi, M. Shabanian, H.A. Khonakdar, Appl. Clay Sci. 114 (2015) 256. [9] M. Zubair, M. Daud, G. McKay, F. Shehzad, M.A. Al-Harthi, Appl. Clay Sci. 143 (2017) 279. [10] K. Yang, L.-g. Yan, Y.-m Yang, S.-j. Yu, R.-r. Shan, H.-q. Yu, B.-c. Zhu, B. Du, Sep. Purif. Technol. 124 (2014) 36. [11] Q. Wang, S.V.Y. Tang, E. Lester, D. O'Hare, Nanoscale 5 (2013) 114. [12] B. Li, J. He, D.G. Evans, X. Duan, Appl. Clay Sci. 27 (2004) 199. [13] D. Shan, S. Cosnier, C. Mousty, Anal. Chem. 75 (2003) 3872. [14] M. Ardanuy, J. Velasco, Appl. Clay Sci. 51 (2011) 341. [15] E.N. Kalali, X. Wang, D.-Y. Wang, J. Mater. Chem. A 3 (2015) 6819. [16] M. Hajibeygi, M. Shabanian, M. Omidi-Ghallemohamadi, Appl. Clay Sci. 139 (2017) 9. [17] M. Hajibeygi, M. Shabanian, H. Moghanian, H. Khonakdar, L. Häußler, RSC Adv. 5 (2015) 53726. [18] M. Shabanian, N. Basaki, H.A. Khonakdar, S.H. Jafari, K. Hedayati, U. Wagenknecht, Appl. Clay Sci. 90 (2014) 101. [19] C.M. Becker, T.A. Dick, J.T. Ramos, F. Wypych, S.C. Amico, The effect of the addition of Mg-Al LDH intercalated with dodecyl sulfate on the fire retardancy properties of epoxy, Macromol. Symp. 319 (2012) 129 Wiley Online Library. [20] W. Chen, B. Qu, J. Mater. Chem. 14 (2004) 1705. [21] F.R. Costa, U. Wagenknecht, G. Heinrich, Polym. Degrad. Stab. 92 (2007) 1813. [22] Y.-N. Chan, T.-Y. Juang, Y.-L. Liao, S.A. Dai, J.-J. Lin, Polymer 49 (2008) 4796. [23] N. Muksing, R. Magaraphan, S. Coiai, E. Passaglia, Express Polym. Lett. 5 (2011).

135