Impact of Carbon Nanotubes on the Mechanical and Electrical Properties of Silicone

As the matrix for the study RTV-2 (Room Temperature Vulcanization) silicone of the Siliform series from Ag-Bet (Dobrzelin, Poland) was used. This is a two-component, sulfurized silicone rubber. According to the technology sheet enclosed with the product, the Shor A hardness of the material is 25 ShA, and the material integrity is up to 3.8 MPa with deformation up to 400%. The viscosity of the A component is 23Pa·s, and when mixed with the hardener (3% wt.), the viscosity is 19 Pa·s.

The useable time after mixing the silicone and hardener is 20 min and the time to full cure at room temperature after mixing the components is 5 h. The silicone used in this study is a typical material used for moulds used to cast components from gypsum or chemicallycured duroplastic resins. Thanks to its properties, silicone can also be used for gaskets used in low-pressure hydraulics and vacuum systems.

For the nanofiller, industrial-grade multi-walled nanotubes of 90% wt.%, 10 nm OD, manufactured by Bucky USA (Houston, TX, USA), were chosen. The MWCNTs were delivered in the form of powder. The manufacturer specifications were as follows: purity: 90 wt.%, OD: 10–30 nm, inner diameter (ID): 5–10 nm, length: 10–30 m, SSA: >200 m²/g, bulk density: 0.06 g/cm³, true density: ~2.1 g/cm³.

Four types of samples were made. The first type of specimen, reference specimens with an s0% designation, was made from pure silicone, without filler. The next three types were:

s4%–4% wt. nanotubes content,

The content of 8 wt.% of nanotubes in the silicone is the maximum amount that facilitates the components to be mixed with each other in a doughy fluid consistency. A content of 4 wt.% and 6 wt.% of nanotubes was used to estimate the effect of the percentage of nanotubes on the properties of the newly formed composite.

Due to the short useful time after mixing the silicone with the hardener, the measured masses of silicone and filler were mixed first. Samples of 20 g from each mixture were measured for analysis of the uncured materials viscosity testing. Next, curing agents were added to the mixture. After mixing, the materials were molded. The mixing process was performed as follows:

Mechanical mixing A agent with nanotubes.

When 4% filler was added to the silicone, it was still liquid. When 6% and 8% filler were added to the silicone, the viscous forces exceeded the bulk forces – it adhered to the surface with no tendency to run off at an observation time of about 1 min.

In order to investigate the effect of the addition of MWCNTs on the change of the complex viscosity, storage modulus and loss modulus of the uncured mixture, rheological tests were carried out using an oscillatory ARES rheometer (Rheometric Scientific Inc., TA Instruments, New Castle, DE, USA) in parallel plate geometry mode. Using the appropriate strain (0.10%) from the linear elastic range, a dynamic oscillatory stress-controlled rotational test was performed at room temperature with a frequency sweep from 0.1 Hz to 100 Hz.

To assess the quality of the produced and cross-linked silicone-based nanocomposites after adding MWCNTs, observations of the structure were performed using transmission optical microscopy. Samples were prepared using Ultramicrotome Leica EM UC6 (Leica Microsystems, Wetzlar, Germany) equipped with a chamber for low-temperature cutting, where the materials were cut with a diamond knife at the temperature −100°C into slides with 2 μm thickness. Afterwards, samples were placed on the surface of a microscopic glass and observed through a light transmission microscope PZO Biolar (Biolar, Warsaw, Poland).

Testing of mechanical properties consisted of measurements of Shor A hardness, modulus of elasticity and Poisson’s ratio.

For susceptible materials such as rubbers, silicones and polyurethanes, one of the most characteristic strength parameters is the Shor hardness A. Susceptible materials can change their properties due to time or external influences (Frigione et al., 2001). Therefore, the measurement of Shor A hardness has become a convenient and quick method to determine the state and physical changes of a material. According to Rodríguez-Prieto et al. (2021), O-rings used in nuclear power plants are tested using the Shora A method after delivery to the user, before installation and during operation. Thanks to the mean values determined from several measurements taken at different locations, it is possible to assess changes in the hardness of the material. The determined standard deviation and normal distribution, on the other hand, make it possible to become a parameter for assessing the homogeneity of the material (Limpert & Stahel, 2011; Liu et al., 2019).

Accordingly, it was assumed that each sample of cured material would be checked 30 times, manually with the use of the Rubber hardness tester Limit 4000. On this basis, the mean value and the standard deviation were determined. In addition to the information obtained about the effect of the nanotube content on the hardness of the nanocomposite, the standard deviation of the hardness at randomly selected locations was used as one parameter to assess the level of homogenization.

In the introduction, a number of publications on the effect of nanotubes on the tensile properties of nanocomposites in a matrix of different polymers were cited. In the present study, the strength properties were also investigated.

Flat panels of 3 mm thickness were made from each nanocomposite and silicone. After curing, three samples were cut from each panel according to the scheme illustrated in Figure 1.

Figure 1.

The tests were performed on an Adamel 2M testing machine from MTS (Eden Prairie, MN, USA) with a sensor type ISA 3125 No. 2066 with a range up to 500N. At four fixed loads F: 5N, 10N, 15N and 20N, the tension was stopped and the crosssectional area of the specimens was measured using an electronic caliper with a measurement accuracy of 0.02 mm. The change in specimen length ΔL was recorded using the testing machine software. From the measurements, strains Eq. (1), stresses Eq. (2) and Poisson’s ratios Eq. (3) were determined. 1 ε=ΔLL, \[\varepsilon =\frac{\Delta L}{L},\] 2 σ=Fa⋅b, \[\sigma =\frac{F}{a\cdot b},\] 3 v=Δbb⋅LΔL, \[v=\frac{\Delta b}{b}\cdot \frac{L}{\Delta L},\]

ε – strain,

A change in the Poisson’s ratio Eq. (3) affects the change in value and the reciprocal of the moduli: the bulk elasticity K – Helmholtz modulus for the three-dimensional case Eq. (4) and the form deformability G – Kirchhoff modulus Eq. (5). 4 K=E3⋅(1−2⋅v), \[K=\frac{E}{3\cdot \left( 1-2\cdot v \right)},\] 5 G=E2⋅(1+v). \[G=\frac{E}{2\cdot \left( 1+v \right)}.\]

For example, a decrease in the Poisson’s ratio results in a decrease in the bulk modulus K and an increase in the modulus of elasticity G. The modulus quotient Eq. (6) also decreases, which can be interpreted as a decrease in the material’s susceptibility to bulk deformation. This in turn makes the material more susceptible to volumetric deformation – it becomes more compressible (Baughman et al., 1998; Hall et al., 2008; Yeganeh-Haeri et al., 1992; Lakes, 1987): 6 3⋅K2⋅G=1+v1−v. \[\frac{3\cdot K}{2\cdot G}=\frac{1+v}{1-v}.\]

Given the above, it was assumed that Poisson’s ratios of the materials obtained would be determined. On this basis, the compressibility will be calculated according to the definition for solids Eq. (7): 7 B=1K. \[B=\frac{1}{K}.\]

In the publications cited in the Introduction, the properties of the nanocomposites were characterized by the specific conductivity, expressed in (s/m). In the present study, the current density was determined as a function of voltage Eq. (8). From the currentvoltage characteristics, it is possible to determine the level of percolation depending on the percentage of nanotubes in the matrix and the applied load/extension of the nanocomposite sample. Furthermore, this characterization makes it possible to determine whether the material satisfies Ohm’s law. 8 J=IS(F,x), \[J=\frac{I}{S\left( F,x \right)},\]

J – current density [ Amm2 ]\[\left[ \frac{A}{m{{m}^{2}}} \right]\]

I – measured current intensity [A],

S(F, x) – cross-sectional area of the specimen [mm²] depending on the applied load F and displacement x.

Electrical measurements were made for unloaded specimens and for four load levels. The tensile test was carried out analogously to the strength tests. For the electrical measurements, a Keithley 2614B Source Meter (Keithley Instruments, Cleveland, OH, USA) with Keithley KickStart Base software was used (Figure 2).

Figure 2.

The effect of the addition of MWCNTs on complex viscosity, storage and loss moduli is presented in Figure 3. Due to the high aspect ratio, MWCNTs are known as nanofillers that strongly influence the viscosity of thermoplastic thermosets nanocomposites (Nobile, 2011). From analyzing the results of complex viscosity (Figure 3a), it is concluded that the viscosity changes as a function of frequency for all materials. This means that there is a shear-thinning behavior with decreasing viscosity, even neat silicone does not behave like a Newtonian liquid. No significant change in complex viscosity was observed between the increase in MWCNTs content from 4 wt.% to 6 wt.% compared to pure silicone. Only for the content of 8 wt% of MWCNTs, the increase was notable. Similar to complex viscosity, storage and loss modulus increase with increasing content of MWCNTs. From analyzing the storage and loss modulus curves (Figure 3b and c), it is observed that nanocomposites with 6 wt.% and 8 wt.% of MWCNTs show elastic behavior due to higher storage modulus values. At lower concentrations of MWCNTs, a more viscous behavior of the material is observed.

Figure 3.

The optical observations of uncured silicone/MWCNT nanocomposites are presented in Figure 4a–d. Due to the significant content of MWCNTs in the produced materials (from 4 wt.% to 8 wt.%), numerous agglomerates of the nanofiller used were observed, which are represented by black areas in Figure 4b–d. The uniformity of observed agglomerates was not noticed, which is the result of a large number of MWCNTs, increased viscosity and strong interactions between the filler and the polymer.

Figure 4.

In addition, voids were observed in the structure of the nanocomposite with a content of 8 wt.% of MWCNTs, which was marked in yellow in Figure 4d. This is the effect of a significant increase in viscosity, which was confirmed in rheological tests (Figure 3a). In this case, for the obtained mixture’s viscosity, the applied vacuum was too low to get rid of air bubbles during the material manufacturing process.

The tests were carried out and as expected they showed an increase in hardness with increasing nanotube content. According to the graph (Figure 5), it can be assumed that the increase in Shor A hardness is linear as a function of the nanotube content. The average increase in hardness per one per cent of no nanotube content is approximately 2.7 ShA.

Figure 5.

The standard deviation with 30 hardness measurements on each sample, was less than 5% of the mean value which was within 1.02–1.30ShA. Taking the scatter measure as a means of measuring material homogenization, the measurement result can be considered satisfactory. The gaskets used in radioactive storage elements are characterized by similar homogenization as determined by the spread measure of the Shor A hardness measurement (Rodríguez-Prieto et al., 2021).

Strength tests were limited to the determination of two parameters: elastic modulus and Poisson’s ratio.

From Figure 6a it can be assumed that the stresses as a function of strain have a linear course. Consequently, the moduli of elasticity (Figure 6b) for the nanocomposites were determined from the relationship (9). 9 E=σε, \[E=\frac{\sigma }{\varepsilon },\]

σ – stress [MPa], ε – strain.

Figure 6.

As the nanotube content in the silicone increased, the elastic modulus of the nanocomposite increased. Between 4 wt% and 6 wt% nanotube content in the nanocomposite, there was a step increase in the elastic modulus. The standard deviation also increased as a function of the nanotube content in the nanocomposite (Nicolau-Kuklińska et al., 2017).

Measurements of the specimens, taken during the tensile test, allowed to determine the Poisson’s ratio for the composites and silicone (Figure 7) and then determine the compressibility (Figure 7). From the calculation results, it can be seen that as the nanotube content increases to a level of 6 wt.%, the Poisson’s ratio increases and thus the compressibility decreases. At a nanotube content of 8 wt.%, the Poisson’s ratio decreases slightly, and with it the compressibility increases.

Figure 7.

In publication (Li et al., 2004) the authors presented a new formula for the dependence of porosity on rock compressibility. According to this formula, rock compressibility increases as porosity increases, while the conventional empirical formula displays the reversal. Based on this approach, the compressibility of the 8% composite can be assumed to be due to porosity, as illustrated in Figure 4d.

A weak electrical conductivity is observed for the manufactured nanocomposites with a content of MWCNT of 4 wt.% and 6 wt.% (see Figures 8 and 9). The current density of these composites places them in the group of dielectrics. Moreover, after applying tensile loading to these specimens, vanishing of electrical conductivity is observed. In contrast to the above-discussed cases, the composite with a content of MWCNT of 8 wt% manifested significantly improved electrical conductivity at the level of semiconductors (see Figure 10). What is interesting, conductivity drops insignificantly when the specimen is subjected to tensile loading and the conductivity remains strong. This phenomenon can be interpreted as exceeding the electrical percolation threshold, resulting in the formation of a critical percolation cluster, enabling the conductivity of an electrical current. This fact is confirmed by the micrographs presented in Figure 4, where for the highest considered content of MWCNT, the agglomerates of the MWCNT nanofiller reveal a tendency of formation of such a critical percolation cluster. Also for previous cases, it can be concluded that there is aformation of percolation clusters, i.e. when the content of MWCNT was 4 wt.% and 6 wt.%, respectively, however, the applied tensile loading successfully interfered with the electrical conduction. In the case of 8 wt.% MWCNT composites, such loading slightly decreases the measured current density values, however, the conductivity of the composite was retained. This confirms the formation of a critical percolation threshold for a content of MWCNT between 6 wt.% and 8 wt% in the analyzed silicone/MWCNT nanocomposite.

Figure 8.

Figure 9.

Figure 10.