**2. Synthesis, processing, and main properties of polymer coatings**

#### **2.1 Synthesis and processing**

The synthesis and processing of nanocomposite polymer coatings usually have at least two separated stages: the dispersion of nanofillers into the monomers, prepolymers, or polymers and the coating manufacture.

There are numerous different techniques to disperse the carbon nanofiller into monomers and polymers. In fact, there are numerous articles and reviews published. For this reason, in this chapter, they are only mentioned. It is wellknown that the improvement of properties, in special mechanical ones, on nanocomposites is strongly dependent on the dispersion quality together with the polymer-nanofiller interface, which relies on the chemical and physical interaction between functionalized nanofillers and polymer matrix. Good dispersion of nanoparticles is critical to achieve high-performance nanocomposite. The most common **processing techniques of nanocomposites** can be organized on three ways: direct mixing, in situ polymerization in the presence of nanoparticles, and solution mixing. One more processing way is the in situ synthesis of particles, which is usually based on in situ sol–gel process inside polymers, but it is only used for inorganic nanofillers.

#### *Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

Carbon nanofillers must be dispersed on polymer or prepolymers depending on the polymer nature. The dispersion of nanofillers on thermoplastic polymers often carries out during the polymer manufacturing process, as extrusion or calandering. However, the nanofiller dispersion on rubber and thermosetting polymers is usually carried out in a previous step of curing process into monomers or prepolymers. In this last case, different dispersion techniques can be also applied, based on the application of mechanical forces or an electric or magnetic field.

As it was mentioned above, the second step consists on manufacturing the own coating, applying different common **processing techniques of coatings**. Cold spray process is commonly used for processing polymer nanocomposite coatings, avoiding the thermal deterioration of substrate. Dispersion, emulsion, and latex in situ polymerizations are other applied manufacturing processes.

#### **2.2 Properties of nanocomposite polymer coatings**

Graphitic nanofillers are often used to improve the **mechanical properties** of polymer coatings. The poor tribological performance of polymer coatings can be improved by adequately addition of graphitic nanofillers into the matrix because graphite is a solid lubricant. Polymer coating containing graphene can present excellent tribological properties, with low friction coefficient and reduced wear rate [7]. The increment of graphene content gradually decreases both friction coefficient and wear rate of composite coating. Under high temperature, graphene-reinforced thermosetting coatings show better friction reduction and wear resistance than neat coating. The values of these properties are enhanced by the increase of graphene content. Meanwhile, the friction coefficient and the wear rate of the graphene/ composite coatings do not show a clear tendency with the increase of temperature. This behavior could be explained by the formation of a transfer film on the surface, which suppresses the huge heat and contact pressure [7]. CNT/polymer coatings can induce anti-friction, wear-proof, and self-lubrication performance [8], reducing the friction and improving the wear resistance. However, numerous factors affect their tribological behavior, such as the composition and properties of sliding pairs, such as their surface roughness and main mechanical properties (hardness, stiffness, and fracture toughness) and the sliding parameters, such as load, speed, temperature, and lubrication state, among others. This behavior is explained by the different involved mechanisms: bridge crack of CNT and lock the propagation of cracks, lubricant effect by dislodgement of individual graphene layers, strengthening of reinforced polymer matrix and dissipation of heat, and reducing the temperature induced wear [8]. It is worthy to note that there is an optimum carbon nanoparticle content to achieve the best tribological properties. However, this value depends on many factors such as aspect ratio of nanofiller, the dispersion degree and orientation of nanofillers, and the interactions with polymer matrix at interfaces.

The incorporation of carbon nanoparticles into polymer composites also increases their hardness. Increasing nanofiller content leads to improvement of hardness; however, the slope of the curve is reduced as the amount of graphitic nanofiller increases, which is attributed to agglomerations in the composite coating [9].

One of the most important applications of polymers reinforced with graphitic nanofillers is as **anticorrosive coatings**. The anticorrosive coatings can be classified in accordance to the protection mechanism against corrosion [10]: barrier protection, cathodic protection, anodic passivation, electrolytic inhibition, and active corrosion inhibition.

have extraordinary electrical and thermal conductivity and a unique combination of mechanical properties with great stiffness and high toughness [1–6]. They are composed of carbon, exhibiting low toxicity and environmental friendliness. For all these reasons, they are considered as multifunctional fillers of polymer matrix. In fact, polymer nanocomposites reinforced with carbon nanoparticles usually present enhanced mechanical, electrical, and thermal properties together with new perfor-

They can act as strain sensors due to their piezoresistive behavior, varying the electrical resistance of composite induced by the deformation of the electrical network formed by graphitic nanofillers. On the other hand, the nanofillers can be used as actuators, for example, as self-heater due to Joule's heating or as chemical absorbents. In this case, the matrix is a neat stimulus-responsive polymer, while the

carbon nanofillers provide the stimuli to induce the polymer response.

**2. Synthesis, processing, and main properties of polymer coatings**

least two separated stages: the dispersion of nanofillers into the monomers,

monomers and polymers. In fact, there are numerous articles and reviews published. For this reason, in this chapter, they are only mentioned. It is wellknown that the improvement of properties, in special mechanical ones, on nanocomposites is strongly dependent on the dispersion quality together with the polymer-nanofiller interface, which relies on the chemical and physical interaction between functionalized nanofillers and polymer matrix. Good dispersion of nanoparticles is critical to achieve high-performance nanocomposite. The most common **processing techniques of nanocomposites** can be organized on three ways: direct mixing, in situ polymerization in the presence of nanoparticles, and solution mixing. One more processing way is the in situ synthesis of particles, which is usually based on in situ sol–gel process inside polymers, but it is only used for

prepolymers, or polymers and the coating manufacture.

The synthesis and processing of nanocomposite polymer coatings usually have at

There are numerous different techniques to disperse the carbon nanofiller into

mance as smart materials.

*Summary of smart coatings with carbon nanoparticles.*

*21st Century Surface Science - a Handbook*

**Figure 1.**

**2.1 Synthesis and processing**

inorganic nanofillers.

**208**

Graphitic nanoparticles enhances the **barrier properties** of polymer coating due to the "torturous path effect" and "nano-barrier wall effect," which strongly depends on exfoliation, dispersion, and orientation degree of nanofiller, their aspect ratio, the polymer-nanofiller interface, and the crystallinity of thermoplastic polymer or cross-linking degree of thermosetting resins. The presence of nanofillers constructs tortuous paths, decreasing the diffusion coefficient. The orientation of graphene and their high surface area forms a zigzag diffusion pathway hindering the diffusion of corrosion species. In addition, their excellent electrical conductivity prevents the electrons form the cathodic site by providing an alternative path [11]. The functionalization of graphitic nanofillers with polar groups [12] enhances the ionic resistance of coating by the creation of negative charge on the graphitic nanofillers surface when exposed to alkaline and neutral environment, preventing the diffusion of chlorine and hydroxyl anions.

associated to the reduction of the electrical conductivity of neat nanofillers due to the partial breakage of some C-C structure during the functionalization and the surrounding isolating polymer layer formed over the functionalized nanofillers,

Nanoreinforced coatings are being studied to improve the **flame retardancy** (FR) of flammable substrates [20]. The addition of graphene usually reduces the total heat release (THR) because they reduce the release of deleterious gas during thermal decomposition, arising from the radical trapping and layered hindering effect. CNT also enhances the FR behavior due to the strengthening of carbonized layers [21]. Also, they act as excellent physical barrier, reducing significantly the

Among other functionalities, the concept of structural health monitoring (SHM) is of great interest in polymer coatings. It is based in an online inspection of the damage extent. In this context, a proper SHM technique must accomplish the following four levels, established by Rytter et al. [22]: (1) detection, (2) localization, (3) quantification of the damage, and (4) the estimation of remaining life, also

Nowadays, there are a lot of different SHM techniques such as lamb waves, fiber optics, and acoustic emission, among others. However, they usually involve complex mathematical and statistical tools and do not often give an overall information of the health of the structure [23, 24]. Therefore, the development of SHM tech-

As commented before, carbon nanoparticles present unique mechanical and, especially, electrical properties in comparison to other materials [25]. Therefore, their addition into an insulating media promotes the creation of electrical networks. This fact induces an enhancement of the electrical conductivity of several orders of

Here, it is important, firstly, to define the concept of **percolation threshold**. It is the critical volume fraction of nanoparticles in which an efficient electrical network is formed, allowing the current flow. It depends on several factors mainly related to the geometry of the nanofiller (including their 0D, 1D, or 2D nature) as well as their dispersion state, that is, their distribution inside the nanocomposite. The determination of the percolation threshold is a crucial factor that determines the minimum

Furthermore, the influence of the different parameters of the nanoparticle network in the percolation threshold has been widely studied in the last years. Li et al. [28] proposed a simple analytical model correlating the geometry, aspect ratio, and

magnitude, becoming the polymer coating electrically conductive [26, 27].

which hinders the direct contacts between electrical nanoparticles.

**3. Smart nanocomposite coatings: Self-sensing**

niques is now gaining a great deal of attention.

**3.1 Fundamentals of SHM with carbon nanoparticles**

content of nanofiller that is needed for electrical applications.

*Schematics of the four SHM levels established by Rytter et al.*

peak heat release rate (PHRR).

*Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

known as *prognosis* (**Figure 2**).

**Figure 2.**

**211**

In the last years, the development of **superhydrophobic** surfaces is being extensively researched. Superhydrophobic coatings have a wide range of applications in textile, automotive parts, construction, agriculture, optical, and maritime industry. It is well established that three factors are needed to create a superhydrophobic surface: low surface energy, microscaled roughness, and nanoand microscaled hierarchical surfaces [13]. The hierarchical multiscale surface can be achieved in coatings with high CNT content, forming CNT agglomerates at the microscale and CNT themselves at nanoscale. Polyvinylidine fluoride (PVDF) are commonly used as hydrophobic polymer coating. Water contact angle increases from 105° for neat PVDF to 170° with very high content of CNT [14]. Similar enhancement of hydrophobicity is reached by the addition of graphene nanoplatelets [15]. The superhydrophobicity is usually requested together with other functionalities, such as self-cleaning, anti-icing, and deicing, which will be addressed further on.

The incorporation of graphene into polymers allows increasing the **thermal stability** of polymer composite [16]. Higher specific area and aspect ratio of carbon nanofillers induce higher stability [17]. This enhancement is also accompanied with an important increase of the low thermal conductivity of polymers. Very high enhancements have been reported, from 0.2 W/mK for neat epoxy coating to 20 W/mK for composite with 30% graphene [16]. Graphene is more effective nanofiller than CNT in order to enhance the thermal conductivity of polymer composites. The thermal conductivity of graphene is attributed to phonons and electrons [18], justifying their excellent thermal conductivity (2000 W/mK). For example, copper is considered a good thermal conductor, whose conductivity is 400 W/mK. Here, the contribution of phonons is limited to 1–2% of total. The addition of nonfunctionalized graphene can induce an enhancement of 50% of thermal conductivity. However, when graphene is modified with functional groups, which is able to form covalent and non-covalent bonds with the polymer matrix, the increment can reach 100%. Moreover, a higher 300% of thermal conductivity increment is reached when graphene is modified with a titanium-coupling agent [19]. The researchers [18, 19] explained it by the reduction of interfacial thermal resistance between graphene and polymer due to the surface modification. For the same reason, higher dispersion degree of graphene into polymer matrix also enhances the thermal conductivity of composite.

It is well-known that the **electrical conductivity** is also increased by the addition of graphitic nanofillers. Higher specific area and higher aspect ratio of nanofillers decrease the nanoparticles which lead to reach an important enhancement of electrical conductivity, from 10<sup>8</sup> to 10<sup>10</sup> S/m for the isolating polymers to 0.01–10 S/m for nanocomposites. Here, the functionalization of nanofillers usually implies a low enhancement of electrical conductivity of composites. This is

#### *Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

Graphitic nanoparticles enhances the **barrier properties** of polymer coating due

to the "torturous path effect" and "nano-barrier wall effect," which strongly depends on exfoliation, dispersion, and orientation degree of nanofiller, their aspect ratio, the polymer-nanofiller interface, and the crystallinity of thermoplastic polymer or cross-linking degree of thermosetting resins. The presence of nanofillers constructs tortuous paths, decreasing the diffusion coefficient. The orientation of graphene and their high surface area forms a zigzag diffusion pathway hindering the diffusion of corrosion species. In addition, their excellent electrical conductivity prevents the electrons form the cathodic site by providing an alternative path [11]. The functionalization of graphitic nanofillers with polar groups [12] enhances the ionic resistance of coating by the creation of negative charge on the graphitic nanofillers surface when exposed to alkaline and neutral environment, preventing

In the last years, the development of **superhydrophobic** surfaces is being extensively researched. Superhydrophobic coatings have a wide range of applications in textile, automotive parts, construction, agriculture, optical, and maritime

superhydrophobic surface: low surface energy, microscaled roughness, and nanoand microscaled hierarchical surfaces [13]. The hierarchical multiscale surface can be achieved in coatings with high CNT content, forming CNT agglomerates at the microscale and CNT themselves at nanoscale. Polyvinylidine fluoride (PVDF) are commonly used as hydrophobic polymer coating. Water contact angle increases from 105° for neat PVDF to 170° with very high content of CNT [14]. Similar enhancement of hydrophobicity is reached by the addition of graphene

nanoplatelets [15]. The superhydrophobicity is usually requested together with other functionalities, such as self-cleaning, anti-icing, and deicing, which will be

The incorporation of graphene into polymers allows increasing the **thermal stability** of polymer composite [16]. Higher specific area and aspect ratio of carbon nanofillers induce higher stability [17]. This enhancement is also accompanied with an important increase of the low thermal conductivity of polymers. Very high enhancements have been reported, from 0.2 W/mK for neat epoxy coating to 20 W/mK for composite with 30% graphene [16]. Graphene is more effective nanofiller than CNT in order to enhance the thermal conductivity of polymer composites. The thermal conductivity of graphene is attributed to phonons and electrons [18], justifying their excellent thermal conductivity (2000 W/mK). For example, copper is considered a good thermal conductor, whose conductivity is 400 W/mK. Here, the contribution of phonons is limited to 1–2% of total. The addition of nonfunctionalized graphene can induce an enhancement of 50% of thermal conductivity. However, when graphene is modified with functional groups, which is able to form covalent and non-covalent bonds with the polymer matrix, the increment can reach 100%. Moreover, a higher 300% of thermal conductivity increment is reached when graphene is modified with a titanium-coupling agent [19]. The researchers [18, 19] explained it by the reduction of interfacial thermal resistance between graphene and polymer due to the surface modification. For the same reason, higher dispersion degree of graphene into polymer matrix also

It is well-known that the **electrical conductivity** is also increased by the addition of graphitic nanofillers. Higher specific area and higher aspect ratio of nanofillers decrease the nanoparticles which lead to reach an important enhancement of electrical conductivity, from 10<sup>8</sup> to 10<sup>10</sup> S/m for the isolating polymers to 0.01–10 S/m for nanocomposites. Here, the functionalization of nanofillers usually

implies a low enhancement of electrical conductivity of composites. This is

industry. It is well established that three factors are needed to create a

the diffusion of chlorine and hydroxyl anions.

*21st Century Surface Science - a Handbook*

enhances the thermal conductivity of composite.

addressed further on.

**210**

associated to the reduction of the electrical conductivity of neat nanofillers due to the partial breakage of some C-C structure during the functionalization and the surrounding isolating polymer layer formed over the functionalized nanofillers, which hinders the direct contacts between electrical nanoparticles.

Nanoreinforced coatings are being studied to improve the **flame retardancy** (FR) of flammable substrates [20]. The addition of graphene usually reduces the total heat release (THR) because they reduce the release of deleterious gas during thermal decomposition, arising from the radical trapping and layered hindering effect. CNT also enhances the FR behavior due to the strengthening of carbonized layers [21]. Also, they act as excellent physical barrier, reducing significantly the peak heat release rate (PHRR).

### **3. Smart nanocomposite coatings: Self-sensing**

Among other functionalities, the concept of structural health monitoring (SHM) is of great interest in polymer coatings. It is based in an online inspection of the damage extent. In this context, a proper SHM technique must accomplish the following four levels, established by Rytter et al. [22]: (1) detection, (2) localization, (3) quantification of the damage, and (4) the estimation of remaining life, also known as *prognosis* (**Figure 2**).

Nowadays, there are a lot of different SHM techniques such as lamb waves, fiber optics, and acoustic emission, among others. However, they usually involve complex mathematical and statistical tools and do not often give an overall information of the health of the structure [23, 24]. Therefore, the development of SHM techniques is now gaining a great deal of attention.

#### **3.1 Fundamentals of SHM with carbon nanoparticles**

As commented before, carbon nanoparticles present unique mechanical and, especially, electrical properties in comparison to other materials [25]. Therefore, their addition into an insulating media promotes the creation of electrical networks. This fact induces an enhancement of the electrical conductivity of several orders of magnitude, becoming the polymer coating electrically conductive [26, 27].

Here, it is important, firstly, to define the concept of **percolation threshold**. It is the critical volume fraction of nanoparticles in which an efficient electrical network is formed, allowing the current flow. It depends on several factors mainly related to the geometry of the nanofiller (including their 0D, 1D, or 2D nature) as well as their dispersion state, that is, their distribution inside the nanocomposite. The determination of the percolation threshold is a crucial factor that determines the minimum content of nanofiller that is needed for electrical applications.

Furthermore, the influence of the different parameters of the nanoparticle network in the percolation threshold has been widely studied in the last years. Li et al. [28] proposed a simple analytical model correlating the geometry, aspect ratio, and

**Figure 2.** *Schematics of the four SHM levels established by Rytter et al.*

dispersion state with the value of the percolation threshold. They concluded that the lower the aspect ratio and the higher the degree of agglomeration of nanoparticles, the higher the percolation threshold. In this context, carbon black (CB) reinforced polymers show very high values of percolation threshold [29] due to their low aspect ratio Bauhofer et al. [30] did an extensive review of percolation threshold in carbon nanotube (CNT)-based polymer nanocomposites by analyzing the effect of nanofiller geometry and dispersion technique. It was observed that the most aggressive dispersion procedures, such as ultrasonication, although leading to the most homogeneous distribution of nanoparticles, lead to a very significant breakage of the CNTs. This prevalent reduction of the aspect ratio leads to increasing values of percolation threshold.

higher the tunneling distance at the initial situation, that is, when no strain is applied, the higher will be the electrical resistance variation associated to the variation in the tunneling distance when applying strain, as can be observed in **Figure 3**. Here, it is necessary to define the concept of gauge factor (GF), as the variation

> *GF* <sup>¼</sup> *<sup>Δ</sup>R=R***<sup>0</sup>** *ε*

Therefore, in order to achieve the highest GFs, it would be necessary to work with volume fractions of nanofiller near the percolation threshold, as the distance between adjacent nanoparticles will be the highest possible to form an efficient electrical network and, thus, the variation of the tunneling distance will be the highest. This has been observed in both GNP and CNT nanocomposites, where the

However, in this sense, there are significant differences among the different nanoparticles. For example, GNP-based nanocomposites have shown a more accused exponential behavior of the electrical resistance with applied strain [35] than CNT-based ones [36]. It means that the values of GF at low strain levels are

This accused exponential behavior of GNP nanocomposites can be explained accordingly to the different interactions that take place inside the electrical network. In fact, the tunneling area of these 2D particles is, generally, much higher than in the case of CNTs, and it leads to the fact that the value of the interparticle distance can be much higher for an efficient tunneling transport. Therefore, as explained before, the higher the tunneling distance, the more accused the exponential correlation between the electrical resistance and the applied strain will be.

Moreover, there is also a correlation between the exponential behavior and the sensitivity of the nanocomposite, and, thus, GNP-based ones show higher GF values than CNT nanocomposites (from 12–15 to 2–4 at low strain levels, respectively, for

nanocomposites manufactured following similar techniques) [35, 36].

*Variation of the tunneling resistance as a function of the tunneling distance between nanofillers.*

contents near the percolation threshold achieved the highest GFs [34, 35].

much lower than at high strain levels.

*Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

**Figure 3.**

**213**

(2)

of the normalized resistance *ΔR=R***<sup>0</sup>** divided by the applied strain, *ε*:

However, the effect of dispersion procedure varies depending on the nanofiller. In this case, ultrasonication has proved to be a good dispersion technique for graphene nanoplatelet (GNP)-based nanocomposites. GNPs are formed by several layers of graphene, and it is widely used as reinforcement in polymer nanocomposites because of the lower cost. Here, ultrasonication promotes the exfoliation of the graphene layers [31]. Therefore, these exfoliating mechanisms induce a reduction of the percolation threshold due to an increase of the aspect ratio of the nanofillers. The combination of an ultrasonication stage with three roll milling can promote the creation of GNP nanocomposites with very low percolation threshold due to the combination of exfoliation stretching effects.

The concept of SHM with nanoparticles, therefore, is based in the monitoring of the changes of the electrical network when subjected to strain or damage. However, for a better understanding of these SHM techniques, it is important to know which are the main conducting mechanisms in the electrical network. Here, three different mechanisms can be identified: the intrinsic conductivity of the nanofiller, the contact between adjacent nanoparticles, and the tunneling transport that takes places between two neighboring particles that are not in intimate contact. Among them, the tunneling transport plays a dominant role in the electrical network of the nanocomposite [32]. It is explained because the associated tunneling resistance is several orders of magnitude higher than the intrinsic and contact resistance. Therefore, the variations of the electrical network when subjected to strain or damage will be ruled by the variation of the tunneling distance between nanoparticles.

In this regard, Simmons [33] established a linear-exponential correlation between the tunneling resistance and the interparticle distance, also known as tunneling distance. It means that the higher the separation between neighboring nanoparticles, the higher the electrical resistance is. More specifically, when subjected to strain, there is a variation of the electrical resistance that is correlated to an increase of the tunneling distance between adjacent nanoparticles:

$$R\_{tumel} = \frac{h^2 t}{A e^2 \sqrt{2m\rho}} \exp\left(\frac{4\pi t}{h} \sqrt{2m\Phi}\right) \tag{1}$$

where *Rtunnel* is the tunneling resistance; *t* is the tunneling distance; h is the Planck's constant; m and *e* are the electron mass and charge; *A* is the area in which electrical transport takes places or tunneling area; and *φ* is the height barrier of the matrix.

#### **3.2 Sensitivity of polymer-based nanocomposites**

In this context, the concept of percolation threshold that has been discussed before plays a key role. In fact, the linear-exponential dependence means that the

#### *Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

dispersion state with the value of the percolation threshold. They concluded that the lower the aspect ratio and the higher the degree of agglomeration of nanoparticles, the higher the percolation threshold. In this context, carbon black (CB) reinforced polymers show very high values of percolation threshold [29] due to their low aspect ratio Bauhofer et al. [30] did an extensive review of percolation threshold in carbon nanotube (CNT)-based polymer nanocomposites by analyzing the effect of nanofiller geometry and dispersion technique. It was observed that the most aggressive dispersion procedures, such as ultrasonication, although leading to the most homogeneous distribution of nanoparticles, lead to a very significant breakage of the CNTs. This prevalent reduction of the aspect ratio leads to increasing values

However, the effect of dispersion procedure varies depending on the nanofiller. In this case, ultrasonication has proved to be a good dispersion technique for graphene nanoplatelet (GNP)-based nanocomposites. GNPs are formed by several layers of graphene, and it is widely used as reinforcement in polymer nanocomposites because of the lower cost. Here, ultrasonication promotes the exfoliation of the graphene layers [31]. Therefore, these exfoliating mechanisms induce a reduction of the percolation threshold due to an increase of the aspect ratio of the nanofillers. The combination of an ultrasonication stage with three roll milling can promote the creation of GNP nanocomposites with very low percolation threshold due to the combination of

The concept of SHM with nanoparticles, therefore, is based in the monitoring

of the changes of the electrical network when subjected to strain or damage. However, for a better understanding of these SHM techniques, it is important to know which are the main conducting mechanisms in the electrical network. Here, three different mechanisms can be identified: the intrinsic conductivity of the nanofiller, the contact between adjacent nanoparticles, and the tunneling transport that takes places between two neighboring particles that are not in intimate contact. Among them, the tunneling transport plays a dominant role in the electrical network of the nanocomposite [32]. It is explained because the associated tunneling resistance is several orders of magnitude higher than the intrinsic and contact resistance. Therefore, the variations of the electrical network when subjected to strain or damage will be ruled by the variation of the tunneling

In this regard, Simmons [33] established a linear-exponential correlation between the tunneling resistance and the interparticle distance, also known as

> *t Ae***<sup>2</sup>** ffiffiffiffiffiffiffiffiffiffi **<sup>2</sup>***m<sup>φ</sup>* <sup>p</sup> *exp*

where *Rtunnel* is the tunneling resistance; *t* is the tunneling distance; h is the Planck's constant; m and *e* are the electron mass and charge; *A* is the area in which electrical transport takes places or tunneling area; and *φ* is the height barrier of the matrix.

In this context, the concept of percolation threshold that has been discussed before plays a key role. In fact, the linear-exponential dependence means that the

**4***πt h*

ffiffiffiffiffiffiffiffiffiffi **<sup>2</sup>***m<sup>Φ</sup>* <sup>p</sup> � �

(1)

tunneling distance. It means that the higher the separation between neighboring nanoparticles, the higher the electrical resistance is. More specifically, when subjected to strain, there is a variation of the electrical resistance that is correlated to an increase of the tunneling distance between

*Rtunnel* <sup>¼</sup> *<sup>h</sup>***<sup>2</sup>**

**3.2 Sensitivity of polymer-based nanocomposites**

of percolation threshold.

*21st Century Surface Science - a Handbook*

exfoliation stretching effects.

distance between nanoparticles.

adjacent nanoparticles:

**212**

higher the tunneling distance at the initial situation, that is, when no strain is applied, the higher will be the electrical resistance variation associated to the variation in the tunneling distance when applying strain, as can be observed in **Figure 3**.

Here, it is necessary to define the concept of gauge factor (GF), as the variation of the normalized resistance *ΔR=R***<sup>0</sup>** divided by the applied strain, *ε*:

$$\mathbf{GF} = \frac{\Delta \mathbf{R} / \mathbf{R\_0}}{\mathbf{e}} \tag{2}$$

Therefore, in order to achieve the highest GFs, it would be necessary to work with volume fractions of nanofiller near the percolation threshold, as the distance between adjacent nanoparticles will be the highest possible to form an efficient electrical network and, thus, the variation of the tunneling distance will be the highest. This has been observed in both GNP and CNT nanocomposites, where the contents near the percolation threshold achieved the highest GFs [34, 35].

However, in this sense, there are significant differences among the different nanoparticles. For example, GNP-based nanocomposites have shown a more accused exponential behavior of the electrical resistance with applied strain [35] than CNT-based ones [36]. It means that the values of GF at low strain levels are much lower than at high strain levels.

This accused exponential behavior of GNP nanocomposites can be explained accordingly to the different interactions that take place inside the electrical network. In fact, the tunneling area of these 2D particles is, generally, much higher than in the case of CNTs, and it leads to the fact that the value of the interparticle distance can be much higher for an efficient tunneling transport. Therefore, as explained before, the higher the tunneling distance, the more accused the exponential correlation between the electrical resistance and the applied strain will be.

Moreover, there is also a correlation between the exponential behavior and the sensitivity of the nanocomposite, and, thus, GNP-based ones show higher GF values than CNT nanocomposites (from 12–15 to 2–4 at low strain levels, respectively, for nanocomposites manufactured following similar techniques) [35, 36].

**Figure 3.** *Variation of the tunneling resistance as a function of the tunneling distance between nanofillers.*

#### **3.3 SHM in nanocomposite coatings**

The enormous potential of the nanoparticles for SHM applications has been widely exploited in the development of surface sensors and smart coatings. More specifically, their use as substitutes of strain gages is gaining attention nowadays. much higher. Therefore, the amount of nanofiller necessary for SHM purposes is

Furthermore, apart from flexible devices such as strain gauges, wearable electronics, or human motion sensors, their SHM capabilities in other polymer coating based on thermosetting polymers have been widely demonstrated. More specifically, GNP-based coatings have proved to be very sensitive at low strain levels to both tensile and compressive loading, as well as a good repeatability under cyclic loads [42]. In addition, CNT-based ones have also demonstrated good sensing

Although the interest as strain sensing devices is very significant, their crack sensing capabilities can be even of more interest. Here, the crack detection is based in a sudden breakage of the electrical pathways due to the presence of the crack itself. It will be reflected in a sharp increase of the electrical resistivity of the coating, and, thus, the electrical resistance during the measurement will increase as

In this regard, electrical impedance tomography is gaining a great deal of attention as an SHM technique. It is based in a mapping of the electrical conductivity of a structure based in the electrical resistance measurements on its surface. Therefore, by using this technique, it will be possible to detect, locate, and even quantify

*(a) Schematics of electrode disposition and (b) mapping of variation of the electrical resistance correlated to an*

much higher in polymer coatings.

*Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

properties with a high linearity [43].

**Figure 5.**

**215**

*induced damage [43].*

well, as shown in the schematics of **Figure 4**.

Basically, a strain gauge is a device for indicating the strain level of a structure at the point of attachment. To date, the most used are based in conventional metallic foils in which the strain is obtained from the electrical resistance variation due to the deformation of the metallic foil when subjected to this strain level. GF values of conventional strain gauges are around 2 and usually show a very linear dependence of electrical resistance change with applied strain.

The research in strain-sensing devices with carbon-based nanocomposites is mainly focused in the development of highly stretchable sensors. In this regard, graphene, carbon nanotubes, and other carbon nanoparticles, such as carbon black, have been widely explored, among others. The addition of these nanoparticles to polymers with high strain capabilities such as fluoroelastomers [37], thermoplastic polyurethane [38], or vulcanized silicone [39] has demonstrated excellent sensing capabilities. More specifically, they present enormous potential for human motion sensing or wearable electronics [37] as the GF at high strain levels (>20%) can be in the range of 400–4000 depending on the content and morphology of the carbon nanofiller. Here, a highly accused exponential behavior is observed at higher strain levels due to the prevalence of tunneling mechanisms in the carbon nanoparticle network. In addition, they can be also used as pressure sensors, with excellent sensing capabilities in comparison to others [40] as they are able to detect very small pressure changes due to the strain induced fail that they promote.

Here, it can be stated that the 2D nature of the electrical network in a nanocomposite coating promotes an increase of the percolation threshold when compared to a bulk nanocomposites, where a 3D uniform disposition of the nanofillers is supposed to be [41]. Moreover, the cross-sectional area of the coatings is obviously much lower than 3D nanocomposites, so the electrical resistance is

**Figure 4.** *Schematics of the effect of a crack in the electrical network.*

#### *Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

**3.3 SHM in nanocomposite coatings**

*21st Century Surface Science - a Handbook*

**Figure 4.**

**214**

*Schematics of the effect of a crack in the electrical network.*

of electrical resistance change with applied strain.

The enormous potential of the nanoparticles for SHM applications has been widely exploited in the development of surface sensors and smart coatings. More specifically, their use as substitutes of strain gages is gaining attention nowadays. Basically, a strain gauge is a device for indicating the strain level of a structure at the point of attachment. To date, the most used are based in conventional metallic foils in which the strain is obtained from the electrical resistance variation due to the deformation of the metallic foil when subjected to this strain level. GF values of conventional strain gauges are around 2 and usually show a very linear dependence

The research in strain-sensing devices with carbon-based nanocomposites is mainly focused in the development of highly stretchable sensors. In this regard, graphene, carbon nanotubes, and other carbon nanoparticles, such as carbon black, have been widely explored, among others. The addition of these nanoparticles to polymers with high strain capabilities such as fluoroelastomers [37], thermoplastic polyurethane [38], or vulcanized silicone [39] has demonstrated excellent sensing capabilities. More specifically, they present enormous potential for human motion sensing or wearable electronics [37] as the GF at high strain levels (>20%) can be in the range of 400–4000 depending on the content and morphology of the carbon nanofiller. Here, a highly accused exponential behavior is observed at higher strain levels due to the prevalence of tunneling mechanisms in the carbon nanoparticle network. In addition, they can be also used as pressure sensors, with excellent sensing capabilities in comparison to others [40] as they are able to detect very

small pressure changes due to the strain induced fail that they promote. Here, it can be stated that the 2D nature of the electrical network in a nanocomposite coating promotes an increase of the percolation threshold when compared to a bulk nanocomposites, where a 3D uniform disposition of the

nanofillers is supposed to be [41]. Moreover, the cross-sectional area of the coatings is obviously much lower than 3D nanocomposites, so the electrical resistance is

much higher. Therefore, the amount of nanofiller necessary for SHM purposes is much higher in polymer coatings.

Furthermore, apart from flexible devices such as strain gauges, wearable electronics, or human motion sensors, their SHM capabilities in other polymer coating based on thermosetting polymers have been widely demonstrated. More specifically, GNP-based coatings have proved to be very sensitive at low strain levels to both tensile and compressive loading, as well as a good repeatability under cyclic loads [42]. In addition, CNT-based ones have also demonstrated good sensing properties with a high linearity [43].

Although the interest as strain sensing devices is very significant, their crack sensing capabilities can be even of more interest. Here, the crack detection is based in a sudden breakage of the electrical pathways due to the presence of the crack itself. It will be reflected in a sharp increase of the electrical resistivity of the coating, and, thus, the electrical resistance during the measurement will increase as well, as shown in the schematics of **Figure 4**.

In this regard, electrical impedance tomography is gaining a great deal of attention as an SHM technique. It is based in a mapping of the electrical conductivity of a structure based in the electrical resistance measurements on its surface. Therefore, by using this technique, it will be possible to detect, locate, and even quantify

superficial defects by analyzing the changes of its surface resistivity. In this context, their effectiveness has been widely demonstrated in polymer coatings for the detection of superficial cracks [43], where the electrical resistance measurement between adjacent channels can easily detect an artificial damage (**Figure 5**), as well as in sensing skins for spatial pressure mapping, where the strain induced by the applied pressure is monitored [44]. Here, the main issue is correlated to the positioning of the electrodes and the data processing, which usually involves the use of complex mathematical tools. However, the results for SHM applications are very promising and give a new functionality to nanoreinforced polymer coatings.

where *Q* is the heat generated, *i* is the current flow, *R* is the electrical resistance,

The first thing that can be analyzed from Eq. (1) is that higher current intensity would lead to higher heat generated and, consequently, higher contents of carbon nanoparticles will be desired for this purpose in order to increase the temperature

Apart from the heat generated, there is an important fact regarding these percolated electrically conductive networks, which is the homogeneous distribution of heat through the coating. Two important effects must be taken into account for this aspect: (1) thermal conductivity of polymers which is particularly low, thus making heat transfer through the coating more difficult, and (2) homogeneous presence of the carbon nanoparticles through the polymer matrix, which is not always easily

Both CNTs and GNPs show extremely good thermal conductivity individually. Nevertheless, in spite of their similar intrinsic thermal conductivity, the morphology of GNPs makes them more interesting for this purpose, even when compared to SWCNT [50]. Even at the same content of both types of nanoparticles, Zakaria et al. found that the thermal conductivity was higher for GNP nanocomposites than the MWCNT ones. In fact, although for electrical properties, higher contents of GNPs are usually required to meet similar properties to the ones found for MWCNT, in that case, at only 3 wt.% of nanoreinforcement, GNP nanocomposites showed an increase of 126.4% in thermal conductivity while 3 wt.% of MWCNT only increased this property by 60.2% [51]. In fact, experimental values of thermal conductivity are usually lower than those predicted theoretically, and it has been attributed mainly to waviness, dispersion, alignment, interfacial resistance, and contact

Proper exfoliation of GNPs causes an important increment on thermal conductivity related to an increase of the aspect ratio. Chu et al. [53] proposed a model to calculate the thermal conductivity of nanocomposites based on randomly oriented nanoparticles which takes into account geometrical aspects of the nanoparticles (aspect ratio) as well as differences in the intrinsic thermal conductivity of the nanofiller in each direction. In the case of GNPs, these aspects will be strongly related to their exfoliation and dispersion in the polymer matrix. On the other hand, the waviness of the nanoreinforcement may reduce the effective aspect ratio of the nanofillers which lead to propose few layers GNPs as an optimal solution instead of

These self-heating coatings do not require extremely high thermal conductivities, but they should be high enough to ensure good heat transfer through the whole

The formation of aggregates is very common in this type of materials, and this may cause that at very low carbon nanoparticle loadings, some resin areas are free of nanoreinforcement, which leads to nonuniform heating of the samples. In that

individual monolayers that tend to roll up easier during dispersion stage.

surface for the purposes mentioned above.

reached or to reduce the voltage required. Although all common carbon nanostructures can be used for this purpose (carbon black, carbon nanotubes, graphene nanoplatelets, or even graphite flakes), the importance of reaching high intensity values usually gives the best results for CNT-filled materials [47]. In fact, very high CNT amounts can be found in the literature in order to increase the electrical conductivity and, consequently, the current flowing at lower voltages applied. This is the case of the study based on ABS as matrix where CNT was added up to 15 wt.% in order to allow reaching temperatures over 200°C when voltages of only 12 V were applied [48] or the research carried out by Chu et al. where similar results in terms of temperature and voltage at contents of 7.5 wt.% of CNT in PDMS were found [49]. The interest in the use of low voltages is based on the use of

batteries commonly installed in cars and trucks, among others.

and t is the time the current is applied.

*Smart Coatings with Carbon Nanoparticles DOI: http://dx.doi.org/10.5772/intechopen.92967*

reached.

resistance [52].

**217**
