rubber material implies that high shear stress will be generated locally during compounding, which is beneficial to the delamination of layered silicates. Hence, compounding with the proper rubber formulation design allows the intercalation and delamination of silicate layers in rubber to become possible [11]. From the literature, the use of organosilicates as precursors for the nanocomposite formations has been extended into various elastomer systems including natural rubber (NR) [12-14], epoxidized natural rubber (ENR) [6, 15], styrene butadiene rubber (SBR) [9, 16], ethylene-propylene-diene rubber (EPDM) [2, 17], nitrile butadiene rubber (NBR) [18-20], polyurethane elastomer [10, 21-23] and silicone rubber [24, 25]. The compounding methods used in these cases can be divided into solution blending, melt mixing and latex compounding. It has been shown that solution blending is a reproducible method to intercalate the rubber molecules into the silicate galleries.
Solution blending of the polar NBR rubber and layered silicates has not yet been reported.
In order to a get more detailed expression on the reinforcing effect of
organosilicate in rubbers, NBR was selected to produce nanocomposites by solution blending with hydrophobic organosilicates. NBR, having high polarity and being one of the most important industrial raw materials, is worthy to be surveyed. The effect of the amount of organosilicate on the mechanical, thermal and barrier properties of these nanocomposites was investigated in the present study. In
addition, the thermodynamical behavior of NBR/layered silicate nanocomposites was also studied.
3-2 Experimental Methods and Analysis 3-2.1 Material preparation
Acrylonitrile-butadiene rubber (NBR, Nipol N32, acrylonitrile content 32%, Chapter 3: Mechanical, Thermal, and Barrier Properties
of NBR/Organosilicate Nanocomposites
Zeon Inc., Japan) was selected as the rubber matrix. Organosilicate intercalated by
dimethyl, dihydrogenatedtallow, quaternary ammonium (Cloisite 15A, Southern Clay Products Inc., USA) with aspect ratio 75-100, was used in this work. For comparison, conventional carbon black (CB, N-220, China Synthetic Rubber, Taiwan) and silica (Aerosil, Degussa, Germany) were also mixed with NBR in a two roll mixing mill directly.
The composition used in this study is given in Table 3A-1.
The mixing of the organosilicate with elastomer was facilitated by the use of solvent [6, 26]. NBR was dissolved in methyl ethyl ketone (MEK). The organosilicates were also swollen in MEK independently and then mixed with the NBR solution with vigorous stirring for 12hrs. The MEK in the resulting dispersion was then evaporated and the sample dried under vacuum for 2 days. After that, the resulting organosilicate-reinforced NBR was mixed with the additives and sulfur in a two-roll miller. The samples were then cured at 160 oC in an electrically heated hydraulic press to their respective cure times, t90. These values were derived from a Monsanto oscillating disc rheometer (MDR 2000) - cf. Table 3A-2.
3-2.2 Characterization
Tensile and tear tests were performed on dumb-bell shaped specimens, following the ASTM standard D 412 and D 624, respectively on a material testing system machine (Sintech, MTS, USA) at a cross-head speed of 50 cm/min. For each data point, five specimens were tested, and the average of the values was taken.
Hardness was directly measured by hardnessmeter (Teclock, Japan) according to ASTM D 2240 standards. Fourier-transformed infrared spectroscopy (FTIR) experiments were performed with a Perkin Elmer IR2000 spectrometer over a frequency range of 650-4000 cm-1. The samples for the FTIR study were prepared with cured NBR compounds using compression moulding. Thermogravimetric analysis (TGA) of each sample was carried out under nitrogen purge in a
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
Perkin-Elmer TGA-7. About 10 mg of cured sample was heated from 50 to 750 oC at 10 oC/min. Dynamic-mechanical thermal analysis (DMA7, Perkin Elmer) spectra were recorded on rectangular specimens (length x width x thickness = 6x 1x 0.25 cm3) in tensile mode at a frequency of 10 Hz. DMTA results, viz. storage modulus and mechanical loss factor (tan δ) were measured in the temperature range from -100 to 70 oC at a heating rate of 5 oC /min.
The XRD patterns of the rubber samples were obtained by a D5000
diffractometer (Siemens, München, Germany) using Ni-filtered CuKα radiation (λ = 0.1542 nm). The samples were scanned in step mode at a 1.5 °/min scan rate in the range of 2θ < 10°. The TEM used is a JEOL JEM 2000FX electron microscope operating with an acceleration voltage of 200 kV. Ultrathin sections of the cured samples were microtomed using Leica Ultracut Uct into about 100-nm thick slices with a diamond knife; subsequently, a layer of carbon was deposited onto these slices and placed on 400 mesh copper nets. The crosslinking density of an elastomer can be determined from equilibrium swelling or mechanical measurements. Swelling experiments were adopted in this thesis and carried out with cured samples by putting the samples in toluene at 25 oC for 48 hrs in order to achieve the equilibrium swelling condition. The uptake solvent percentage, Q, and volume fraction of NBR in the swollen gel, Vr, were calculated by the following equations:
Q = (Msw - Mi)/ Mi (1) Where Mi andMsw are the weight of the rubber sample before immersion in the solvent and the swollen state, respectively.
Vr = (1/Dsam) / [(1/ Dsam)+(Q/ Dsol)] (2) Where Dsam and Dsol are the densities of the rubber sampleand solvent (0.87 for toluene). The crosslinking density of the sample, ν, defined by the number of elastically active chains per unit volume, was calculated by the Flory-Rehner equation
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
[27]:
(3)
Where Vs is the molar volume of the swelling solvent (106.1 cm3/mol for toluene).
χ is the Flory-Huggins (rubber-toluene) interaction parameter and was taken as 0.435 for the NBR-toluene system in this calculation.
Crosslinking density has also been determined from equilibrium stress-strain measurements using the Mooney-Rivlin equation :
(4)
where σ is engineering stress, λis extention ratio, C1 and C2 are constants. By plotting σ/[2(λ-λ-2)] versus 1/λ and extrapolating to 1/λ= 0, a value of C1 can be obtained. By comparison with the theory of rubber elasticity, it has been proposed that C1= ½ NRT, where N is crosslink density, R the gas constant, and T absolute temperature.
Thermodynamical aspects of rubber elasticity are crucial to obtain a deeper understanding of mixing in NBR/silicate nanocomposites. The expansion of rubber in the presence of a solvent will significantly modify the conformational entropy (ΔS) and the elastic Gibbs free energy (ΔG). The elastic Gibbs free energy can be
determined from the Flory-Huggins equation [11, 14]:
ΔG = RT[ln(1- Vr) + Vr +χVr2] (4)
From the statistical theory of rubber elasticity, the conformational entropy ΔS can be obtained from ΔG = -TΔS, which assumes that no changes in the internal energy of the network occurs upon stretching.
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
The water and solvent (methanol) vapor transmission were measured at 40 oC and 50
% relative humidity, according to ASTM E96. To minimize the influence of thickness on the vapor transmission, the thickness of the sample is set to be 5mm by molding. The permeability, P, is calculated by the following equation:
P = 6237 × (W-W1) (5) Where W and W1 are the weights of the sample assembly before and after testing respectively. The relative permeability was defined as the measured permeability divided by the pure NBR permeability.
3-3 Results and Discussions
The dispersion image of layered silicates on the NBR matrix are shown and discussed in the section 2-3-A for the NBR-layered silicate nanocomposites. The mechanical properties and the chemical reaction mechanism of the NBR-layered silicate
nanocomposites are discussed in the section 2-3-B.
3-3-A. Layered silicates dispersion in NBR nanocomposites 3-3-A.1 curing characteristics
Table 3A-1 gives the composition of NBR nanocomposites used in this study.
Its curing characteristics are summarized in Table 3A-2. In the presence of
organosilicates, the initial scorch time, t2, (about 89 seccond) was shorter than that of pure NBR (106 second). This is because the acidic nature of organosilicate activates the formation of soluble zinc ions. The soluble zinc ions in turn induces the
decomposition of accelerator into free radicals in the earlier reaction stage. These free radicals cause the premature vulcanization and result in a decrease of the scorch time. The cure time of NBR nanocomposites, t90, increased with the content of the organosilicate. This result is different from the catalytic effect reported in some of Chapter 3: Mechanical, Thermal, and Barrier Properties
of NBR/Organosilicate Nanocomposites
the NR/organosilicate nanocomposite literature [15]. It can be attributed to the greater acidity of the organosilicates, which exhausts some decomposed accelerator free radicals in the following crosslinking reaction, retarding curing and influences the kinetics of the crosslinking reaction. Another possible cause [28] is resulted from the ZnO, which neutralizes the most active sites of the filler surface and has been proven in silica. Both ML and MH of these nanocomposites increase with the amount of organosilicate, resulting an increase in the stiffness of NBR matrix effectively.
The increasing stiffness, MH-ML, corresponds with the increasing hardness tested results as given in Table 3A-2.
3-3-A.2 Dispersion of the silicates
The efficiency of organosilicate in reinforcing the polymer matrix is primarily determined by the degree of its dispersion in the matrix and the extent of intercalation of organosilicate by polymer molecules. Figure 3A-1 shows the (001) diffraction peak of the organosilicate is shifted to lower diffraction angles in the nanocomposite as compared to that of the pure organosilicate. The d-spacings of the stacks were 3.53-4.26 nm for the NBR/organosilicate nanocomposites, whereas it is 3.15 nm for the pure organosilicate. In the NBR nanocomposites containing 3 parts of layered silicates has the largest interlayer distance, 4.26nm. At low silicate loading, the viscosity of the nanocomposite is not large enough to generate a sufficient shearing force for moving the majority of rubber molecules to intercalate into the layered silicates gallery. Hence, the increase in the d-spacing of layered silicate is limited.
At high silicate loading, the dispersion of layered silicates is difficult, but a high shear force has been resulted for moving large rubber molecules to intercalate into the layered silicates gallery. Therefore, a compromise was reached by the two factors at about 3 parts silicate loading. This indicates that an optimum loading of
organosilicate in the NBR case has been obtained. The peaks around 4.18-4.93o and Chapter 3: Mechanical, Thermal, and Barrier Properties
of NBR/Organosilicate Nanocomposites
6.0-7.0o result from higher order diffractions from the (002) and (003) diffraction planes, respectively. The multi-diffraction peaks reflect a well-defined layered silicate structure. In the case of 1 phr organosilicate in NBR, a lower but strong peak around 6.50o (1.36nm) was displayed. It has been attributed to the thermal
degradation and desoption of the organic materials in the gallery [29]. In this study, however, the NBR nanocomposites were prepared by solution blending without a high temperature mixing process. Thus, it is speculated to be an effect of the
organosilicate confinement (reaggregation) [10, 15]. The formation of a zinc-sulfur accelerator complex “extracts” the amine intercalant of the organosilicates, thus causing the collapse of the layers. Figure 3A-2 shows the TEM image of the NBR nanocomposite containing 7.5 phr organosilicate. Most of the layered silicates are well-distributed in the NBR matrix and a large portion of the organic-modified silicate layer particles appear to be interlated along with a few single delaminated platelets.
No visible differences in the morphology of any of the NBR/organosilicate samples can be found. The TEM result is in consistent with that of the XRD study, as shown in Figure 3A-1. Both TEM and XRD results thus confirm that the layers of the organosilicate particles have been intercalated successfully in the NBR
nanocomposites.
3-3-B. Mechanical, thermal and barrier properties of NBR nanocomposites 3-3-B.1 Mechanical properties
Figure 3B-1 shows stress-strain curves in tensile for the NBR nanocomposites containing different amounts of organosilicate. The tensile properties of these nanocomposites increase with the amount of organosilicate. At high strains, stress hardening behavior is observed for the NBR incorporating organosilicate. Table 3B-1 summarizes the ultimate properties of tensile strength, elongation at break,
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
modulus at 100, 300 and 500% elongation, and tear strength. The mechanical properties of the NBR nanocomposites, such as the tensile strength, elongation at break, modulus at different elongation and tear strength relative to the pure NBR are enhanced due to the presence of intercalated/exfoliated layered silicates. A more than six-fold increase in tensile strength, two-fold increase in M500, 39%
enhancement in elongation at break, and 267.8% enhancement in tear strength are obtained with the addition of 10 phr of the exfoliated organosilicate. These gains in strength and stiffness without loss in elongation in these nanocomposites are quite different from the behavior of conventional composites. The reinforcing effect is presumed to occur due to intercalated/exfoliated organosilicate layers that are covered by highly crosslinked rubber molecular chains with strong interfacial interactions in between [3, 9, 21].
Figure 3B-2 shows the tensile properties of NBR containing organosilicate, high strengthening super abrasion furnace carbon black (CB, N220) and silica (Aerosil).
The reinforcing effect on NBR by organosilicate appears more pronounced.
Particularly, the reinforcing efficiency in the tensile modulus is higher than that of NBR/CB compound under 600% strain. For instance, only 3 to 5 parts of
organosilicate are enough to obtain a comparable tensile strength to that of NBR containing 40 parts of Aerosil. In the conventional carbon black, the reinforcing effect was interpreted as bound rubber phenomenon involves physical adsorption, chemisorption, and mechanical interaction [28].
Infrared spectra of the polymer nanocomposites can be used to illustrate the occurrence of chemical reactions or strong interactions between polymer and
reinforcers [12, 21, 30-31]. Polymer/layered silicate nanocomposites are difficult to analyze because of the reflective properties of the layered silicate platelets. Figure 3B-3 shows FTIR spectra, over a frequency range of 950-1600 cm-1, of organosilicate
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
as received, cured NBR/organosilicate and pure NBR compounds, Absorption peaks are observed at 1010, 1041, 1129 (C=S not linked to N), 1200 (C=S not linked to N), and 1242 (C=S linked to N) cm-1, which are characteristic of cured pure NBR
compound. The peaks at 1010 and 1041 cm-1, attributed to sulfoxide stretching vibrations, shift to 1040 and 1077 cm-1 in the NBR/organosilicate nanocomposite, due to the interferece effect of Si-O stretching vibrations in layered silicates. The peaks at 1129, 1200, and 1242 cm-1, attributed to the accelerator reaction products
((cyclohexyl benzthiazyl sulphonamide (CBS) and dibenzthiazyl disulphide (DM)) and sulfur, disappear in the NBR nanocomposite. Whereas, a medium and broad absorption band at approximately 1217 cm-1 appears. This indicates that a strong interaction between layered silicates and the curative reaction products exists and results in a new absorption peak in the NBR nanocomposite. The interesting outcome is the appearance of a C-SO2-N bond peak around 1365 cm-1 in the NBR nanocomposite, which is absent in the spectrum of pure NBR and of the layered silicates. This may be the result of the possible reaction of an ammonium salt intercalant and zinc-sulfur-amine complexation between the organosilicate and highly polarized NBR during the vulcanization process. Schematic drawings of the
possible reaction mechanism are provided in Figure 3B-4. NBR molecular chains were therefore more easily intercalated into the intergallery space of the organosilicate during this process. Recalling the microstructure discussed above, the NBR
molecules intercalated into the organosilicate layers could also be considered an effect of bound rubber. This also indicates a strong interaction exists between the
intercalated NBR chains and organosilicate layers [6].
The effect of the incorporation of the organosilicate on the crosslinking density of NBR can be estimated by an application of the Flory-Rehner equation. Table 3B-2 summarizes the crosslinking density and thermodynamical characteristics of the
Chapter 3: Mechanical, Thermal, and Barrier Properties of NBR/Organosilicate Nanocomposites
NBR compounds. The crosslinking density is found to increase with the amount of organosilicates. The crosslinking density for the NBR/organosilicate nanocomposite is significantly higher than that of pure NBR. The results are in agreement with those vulcanization characteristics and the tensile properties, indicating a strong rubber/organosilicate interaction (bound rubber) in the nanocomposites. Table 4 lists the thermodynamical parameters, ΔG andΔS, of the NBR nanocomposites. A considerable increase in the free energy is observed in the NBR/organosilicate nanocomposites. These results can be attributed to better compatibility between the organosilicate and NBR rubber. The NBR molecules can penetrate into the galleries more easily and this results in intercalated/exfoliated structures.
3-3-B.2 Thermal properties and relaxation behavior
The thermal properties and relaxation behavior of NBR nanocomposites were analyzed by TGA and DMA, respectively. Table 3B-3 summarizes the degradation temperatures (Td5, Td50), storage moduli (E’), glass transition temperatures (Tg), and Tanδ values. The degradation temperature at 5% and 50% weight loss were both obtained in this study. The Td5 decreases with an increasing amount of
organosilicate. This is because the alkyl chain type organo-modifier, such as hydrogenated tallow begins to decompose above 200 oC [32]. Whereas, the resistance of nitrile butadiene rubber molecules to thermal decomposition was improved by the presence of intercalated/exfoliated organosilicate at temperature higher than 450 oC. For instance, the decomposition temperature for a ten parts loading of organosilicate on NBR nanocomposite (493 oC) was 25 oC higher than that of pure NBR (468 oC).
The onset storage modulus E’, given in Table 3B-3, increases as the amount of rganosilicate increases. Specifically, the nanocomposites containing 10 phr
organosilicate have the highest storage moduli which are 57% higher than that of pure Chapter 3: Mechanical, Thermal, and Barrier Properties
of NBR/Organosilicate Nanocomposites
NBR. These results can be attributed to the larger active surface area and stronger NBR/organosilicate interactions, which is in harmony with the mechanical results.
The glass transition temperature was obtained from the peak temperature of tan δ as depicted in Table 3B-3. Decreases in Tg and tan δ also occur as the organosilicate content increases. Similar results were reported and rationalized for ammonium salt intercalated in silicate layers that may act as a plasticizer or lubricant [16, 32].
3-3-B.3 Barrier properties
Figure 3B-5 highlights the barrier improvement of NBR-based nanocomposites against water and methanol vapor. The relative vapor permeability of the NBR rubber is reduced markedly by 85% (from 100% to 15%) for water and by 42% (from 100% to 58%) for methanol, whereas the permeability values for water and methanol are 967 and 5111 g/m2/day, respectively. This indicates the intercalated/exfoliated NBR/silicate nanocomposites improve the vapor barrier properties dramatically. It is noteworthy that the methanol permeability results agree with the crosslinking density results discussed above. Therefore, the permeability is controlled by the
microstructure of the nanocomposite and the interaction between NBR and organosilicates.
The dramatic enhancement of mechanical properties of NBR by forming nanocomposites through a solution blending process presented in this study is attractive than those of NBR nanocomposites obtained by melt mixing or latex blending process as reported in the literature [18-20]. For example, a six folds increase in the tensile strength without a loss in the ultimate elongation by the solution method has been achieved. Whereas, a three to four folds increase of tensile strength is observed with a reduction in the elongation by conventional melt mixing process.
Permeability test results show the same trends. This can be attributed to two factors.
The first is that the interlayer distance of layered silicate was expanded more by the Chapter 3: Mechanical, Thermal, and Barrier Properties
of NBR/Organosilicate Nanocomposites
solvation effect of MEK than by melt-mixing method. The agglomerated silicate particles are reduced significantly by shearing force in the blending stage, which also lead to better dispersion and miscibility of organosilicate in NBR matrix. Finally the large and coiled nitrile butadiene rubber molecular chains are also swollen and
stretched in the good solvent, which is advantageous for intercalation during the blending process.
3-4 Conclusions
NBR/organosilicate nanocomposites, with exfoliated and intercalated structures, were successfully prepared by the solution blending method. The hardness, tensile properties and tear strength of these nanocomposites increase substantially with the amount of incorporated organosilicates by containing a few weight percent of layered silicates, as compared to pure NBR. This reinforcing effect can be attributed to strong interactions between the intercalated NBR molecular chains and organosilicate
NBR/organosilicate nanocomposites, with exfoliated and intercalated structures, were successfully prepared by the solution blending method. The hardness, tensile properties and tear strength of these nanocomposites increase substantially with the amount of incorporated organosilicates by containing a few weight percent of layered silicates, as compared to pure NBR. This reinforcing effect can be attributed to strong interactions between the intercalated NBR molecular chains and organosilicate