*3.3. Self-Strain Sensing Properties*

The prepared hybrid film was cut into a rectangular strip (30 × 10 mm) for evaluation of the piezoresistive response and sensing performance. The strain-monitoring capability of the hybrid film sensor was tested in a flexural test. Electrical resistance and mechanical strain during the test were measured simultaneously by a digital Multimeter (Keithley 2450) and strain gauge, respectively. The hybrid film sensor was attached to the center of an Al (Al6064-T6) test specimen (dimensions: 200 × 19 × 2 mm) using epoxy to make perfect bonding between the specimen and hybrid film sensor. When the load was applied on the Al specimen, the hybrid film bonded through high strength epoxy and the metallic strain gauge experienced the same strain. Two copper electrodes were adhered to the hybrid film sensor at a distance of 25 mm using silver paste to minimize the contact

resistance. The electrical resistance in monitoring tests was measured by the two-point method due to the simplicity of the method regarding scalability to real applications [21]. Resistivity and strain data were recorded by the digital data acquisition system (cDAQ-9174 NI) through the Lab VIEW software. In this work, a four-point-bending test was conducted to study the piezoresistive behavior of the hybrid film sensor. The spans between the two inner points and two outer points are 60 mm and 120 mm, respectively. A schematic diagram and experimental setup of the four-point-bending test are shown in Figure 6.

**Figure 6.** Schematic diagram and experimental setup of the four-point-bending test.

The addition of GNPs increases the conductivity of the hybrid film as described in Section 3.2. Four-point-bending tests were performed to monitor the electrical resistance change of the hybrid film with different GNP contents induced by the strain. Gauge factor is an important parameter which can be used to describe the sensitivity of the strain sensor. It is defined as the ratio of the normalized electrical resistance and strain induced in the sensor as follows.

$$\text{GF} = \frac{\Delta R / R\_0}{\varepsilon} \tag{1}$$

where ΔR is the resistance change with strain, *R*<sup>0</sup> is the initial resistance prior to straining, ε is the applied strain.

Representative normalized resistance–strain curves of the experimental results are plotted in Figure 7 for various GNP contents ranging from 0 to 50 wt.%. It can be observed that the normalized resistance behaves in positive piezoresistive trend, i.e., the normalized resistance change monotonic increases with the increase of the strain. Moreover, the normalized resistance of the hybrid film is increasing with the increase of GNP content. For the increase of the resistance curve, an evident change occurs around at the strain of 0.2%. The whole curve can be divided into two stages. In stage 1 (strain range 0–0.2%), the increase of the resistance tends to be linear with a small slope. In stage II (strain range 0.2–1%), the resistance change exhibits a linear relationship with a large slope. The slope of the curve represents the gauge factor of the hybrid film which can be used to characterize the strain sensitivity of the hybrid film sensor. The gauge factors of the hybrid films with different GNP contents for stage I and II are listed in Table 3. It can be observed that the gauge factor is increasing with the increase of the GNP content as shown in Figure 8. Furthermore, gauge factor in stage I is larger than that of stage II. As the GNP content increases from 0 wt.% to 50 wt.%, the gauge factor increases from 1.16 to 2.34 in stage I, and increases from 1.54 to 3.56 in stage II. The mechanism corresponding to the increase of the resistance in the two stages can be explained as follows. The resistance of the hybrid film can be attributed to three main aspects, namely, contact resistance, tunneling resistance and intrinsic resistance. In stage I, the gauge factor is mainly affected by intrinsic resistance, the relative displacements of MWCNT and GNP are small, the wavy carbon nanotubes are straightened under strain due to its large flexibility, and a smaller gauge factor is acquired. However, in stage II, the normalized resistance change (ΔR/R0) of the hybrid film is mainly relied on the contact and tunneling resistances of adjacent nanomaterial sheets. The conductivity between neighboring flakes is determined by their overlap area and the contact resistance [37]. The assumption in the sensitivity change of MWCNT/GNP hybrid films can be further explained by the schematic diagram shown in Figure 9. Once a mechanical strain is applied to the hybrid film, the overlap area between neighboring flakes becomes smaller and the gap distance becomes larger, which results in an increase of the tunneling pathway between adjacent nanoplatelets so the tunneling resistance increases. In the process of mechanical loading, the tunneling resistance instead of the contact resistance becomes the dominant factor of the resistance. In addition, the more the GNP content, the more easily the conductive path gets disrupted by external strains, which results in higher strain sensitivity. Similar results were reported by Lu et al. [38]. They found that the sensitivity of the GNP/epoxy sensor was varied along with the applied strain and can be separated to three strain regions (0–0.2%), (0.2–0.6%) and (0.6–1.2%), respectively. The gauge factors of the GNP/epoxy sensor with 1.58 vol.% of GNP corresponding to these three strain regions were 2.53, 3.77 and 4.69, respectively.


**Table 3.** Gauge factor for different strain stage.

**Figure 7.** Normalized resistance change increases with the increase of the strain.

**Figure 8.** Gauge factor for different strain stages.

To investigate the stability, reversibility and reliability of the hybrid film sensor, the specimens were subjected to 200 cyclic loading–unloading tests. This test aimed to monitor the electric resistance response of the hybrid film under cyclic loading. The dynamic responses of the normalized resistance change and mechanical strain of hybrid films with 0 wt.% (GNP-0) and 50 wt.% (GNP-50) of GNP are plotted in Figure 10a,b, respectively. It can be observed that there is no obvious change during the 200 cycling tests for the hybrid film sensors. This demonstrates that the high durability and stability of the hybrid film sensor. Some researchers [39,40] also did the cycle loading–unloading tests for stability of the GNP/epoxy sensors, they are stable under certain cycles, but the cycles they tested were as low as 50 or even several cycles.

**Figure 9.** Schematic representation of microstructure changes in hybrid films subjected to mechanical strain. (**a**) MWCNT film; (**b**) Stretching of MWCNT film under flexural strain; (**c**) MWCNT/GNP hybrid film; (**d**) Stretching of MWCNT/GNP hybrid film under flexural strain.

**Figure 10.** Normalized resistance change and mechanical strain of the hybrid film under cyclic loading-unloading test (**a**) 0 wt.% GNP-0 (**b**) 50 wt.% GNP-50.

### **4. Conclusions**

MWCNT/GNP hybrid films were prepared with the aid of surfactant Triton X-100 and sonication through vacuum filtration process. SEM images show that MWCNTs and GNPs are successfully deposited to form densely packed film with layered structure. The effect of GNP content ranging from 0 to 50 wt.% on the mechanical and electrical properties of the hybrid films were characterized using the tensile test and Hall effect measurements, respectively. It can be observed that both the tensile strength and fracture strain are decreasing with the increase of GNP content. The electrical conductivity is increasing from 47.7 S/cm to 192.6 S/cm as the GNP loading increases from 0 to 50 wt.%. A series experimental tests were conducted to study the piezoresistive behavior and the strain-sensing capability of the hybrid film. The gauge factor defined as the ratio of relative change in resistance to applied strain was used to characterize the sensitivity of the strain sensor. There are two different linear strain-sensing stages (0–0.2% and 0.2%–1%) in the resistance of the hybrid film with applied strain. The gauge factor increases from 1.164 to 2.236 as the GNP loading increases from 0 to 50 wt.% in the strain-sensing range 0–0.2%. Moreover, the repeatability and stability of the strain sensitivity of the hybrid film were conformed through the cyclic loading and unloading tests. From the results obtained, it is demonstrated that the MWCNT/GNP hybrid film is very suitable for strain sensing.

**Author Contributions:** Conceptualization, J.R.H. and S.-C.H.; methodology, J.R.H. and X.X.Y.; validation, S.-C.H. and X.X.Y.; investigation, J.R.H. and M.N.Z.; writing: original draft preparation, J.R.H.; writing: review and editing S.-C.H.

**Funding:** This research was funded by Ministry of Science and Technology of the R.O.C, grant number MOST 104-2221-E155-057-MY3.

**Acknowledgments:** The authors would like to thank Mr. Yuan-Ming Liang and Wei-Chun Hsu at Yuan Ze University for their assistance of the experimental setup and tests.

**Conflicts of Interest:** The authors declare no conflict of interest.
