First of all, for the morphological study of the fabricated Ta2O5:Si:Graphite tricomposite nanoflower structure, we carried out scanning electron microscopy (SEM) at various magnification i.e, for 5 µm, 0.5 µm, 200 nm, and 100 nm. The SEM images depicted in figure 3 (a-d) are Ta2O5:Si:Graphite tricomposite nanoflower structure for f=0.5 with above mentioned magnifications. In Fig. 1 (a-d), it may be noticeable that for lower magnification values i.e. from 5 µm, 0.5 µm, the bunches of nanoflower were clearly visible which confirms the presence of flower-like structures. As the magnification is increased i.e. for 100 nm shown in Fig. 3 (c), clear image of the petals of flower can be seen. In Fig. 3 (d), the dimension of the petals are clearly visible. Initially when f=0, only Ta2O5 was there, the nanoflower structure of graphite and Ta2O5 were formed. As the value of ‘f’ increases i.e. for f=0.1 to f=0.9 sufficient amount of Si was there so Ta2O5:Si:Graphite tricomposite nanoflower structure was formed. Finally, when f=1, there was no Ta2O5 and the nanoflowers of only Si and graphite were formed. Just for simplicity we have given the SEM images of different magnification for f=0.5.
Now, to get the crystallographic and structural information of the fabricated nanostructure, X-ray diffraction (XRD) was performed. The XRD patterns of Ta2O5:Si:Graphite tricomposite nanoflower structure is shown in figure 4. The diffraction pattern of Ta2O5:Si:Graphite tricomposite nanostructure fabricated at room temperature for all the values of 'f ' from 0 to 1 is clearly mentioned.
Further, to obtain the information for the optical properties of fabricated Ta2O5:Si:Graphite tricomposite nanoflower structure, FTIR was carried out. In Fig. 5, it may be noticeable that the trend of the FTIR spectrum for all the fabricated tricomposite nanostructure were similar but the intensity of peaks increases with varying value of 'f '. The Intensity of FTIR peak increases as ‘f’ increases from 0 to 0.8. On further increasing the value of ‘f’, FTIR peak starts decreasing. In figure 5, typical band at 3586 cm−1, 3259 cm−1, 2872 cm−1, 2505 cm−1, 2207 cm−1, 1295 cm−1, 1067 cm−1, 769 cm− 1, 462 cm− 1, have been mentioned. The broad band at 3535 cm−1, 3259 cm−1, 2872 cm1, 2505 cm−1, 2207 cm−1, indicated the stretching vibration of hydroxyl group and interactions has been taken place by Ta2O5. Similarly band at 1067 cm−1, 769 cm− 1, 462 cm−1 was attributed to bending vibration of Ta and O. The peak at 1295 cm−1 is due to Si vibrations either by oxygen or Ta interactions. The information of the interactions observed from the bands of FTIR spectrum indicates the overall interactions within the fabricated Ta2O5:Si:Graphite tricomposite nanostructure which give significant knowledge about their optical properties.
The more detail information about the optical properties of Ta2O5:Si:Graphite tricomposite nanostructure for all the fabricated values of ‘f’ from 0 to 1 can be obtained using photoluminescence (PL) investigations. The PL emission spectra of Ta2O5:Si:Graphite tricomposite nanostructure were recorded and shown in figure 6 (a). It may be noticeable from the figure that for f=0, PL intensity is also lowest. As the value of ‘f’ increases from f=0.2 to f=0.4 PL intensity also increases but as the value of ‘f’ further slightly increased from 0.4 to 0.5, PL intensity starts decreasing. On further increasing the value of ‘f’ from 0.5 to 0.8, PL intensity becomes constant. From the trend of the PL intensity for f=0 to 0.8, it is noticeable that maximum PL intensity is observed when f=0.4. The pictorial representation given in Fig. 6 (b) shows the reason of the obtained trend of the PL intensity and the variation of defects levels for varying values of ‘f’ is also explained. The photoluminescence is defined by the number of emitted photons due to excitations of light and the associated electrons which are coming back from conduction band to valence band during recombination. In Fig. 6 (b) it is defined that when f=0, PLI is lowest because the number of defects level (metastable states) would also be low. But for higher values of 'f' PLI is more so there would be more defects level so electrons reaches to more number of metastable states and therefore it takes more time in returning to valence band from the conduction band consequently emits more number of photons before recombination. The highest PL intensity is obtained when f=0.4 therefore it may be inferred that maximum defects levels are found at f=0.4. Similarly as PL intensity decreases the defects levels get altered in the same fashion.
Now, experiments were performed for the Ta2O5:Si:Graphite tricomposite nanostructure for all the values of 'f' towards refractive index sensing applications. The absorption spectra of the prepared solution for f=0.4 in refractive index range of 1.33 to 1.38 is shown in Fig. 7 (a). The corresponding variation of peak absorption of Fig. 7 (a) is plotted in Fig. 7 (b). In accordance with the absorption spectra of Fig. 7(a), peak absorption wavelength is observed to increase with an increase in the refractive index of the solution i.e., a red–shift in the peak absorption wavelength is obtained which is clearly shown in Fig. 7 (b). This is termed as the calibration curve for the refractive index sensor which governs the non–linear trend of peak absorption wavelength with refractive index. It is worthwhile to mention that the amount of wavelength shift in the absorption spectrum is very sensitive to the surrounding refractive index. For medium with lower refractive index, more energy is required to collectively excite the surface electrons to generate LSPR signal. Thus, the peak absorption wavelength values confine towards lower wavelength regime. As the refractive index of the medium is increased, comparatively smaller amounts of energy are needed to generate LSPR signal consequently, the peak absorption wavelength shifts towards higher wavelength values.
Now, from the calibration curve, sensitivity of the refractive index sensor can be evaluated. Sensitivity is a measure of the shift attained in peak absorption wavelength with respect to the change in refractive index. Mathematically, sensitivity is calculated from the slope of the calibration curve. For each refractive index value, the value of sensitivity is measured and is plotted as a function of the refractive index in inset of Fig. 7 (b). As is evident from figure, the sensitivity curve pursue a declination trend with an increase in the refractive index. The maximum value sensitivity is determined to be ~(156-260) nm/RIU in (1.33-1.38) refractive index range. Since the mathematical equation narrating the calibration curve is a polynomial of second degree, its differentiation yields a linear equation, and thus, the sensitivity is noticed to follow a linear trend with refractive index.
Limit of detection (LOD) is yet another parameter to characterize a sensor. A further investigation has been made to measure LOD of the present refractive index sensor. For LOD, refractive index resolution should be measured. LOD is used to estimate the minimum possible value of refractive index that can be measured with the sensing layout.
Interpreted mathematically, refractive index resolution provides the minimum possible change in the refractive index wavelength attainable with the sensing framework. In this aspect, refractive index resolution of present sensor is evaluated. Resolution is evaluated from the below mentioned formula [15]
(3)
Here, ΔλS.D. denotes the standard deviation calculated in the measurement of peak absorption wavelengths for different refractive index solutions and S(RI: 1.33) represents the sensitivity attained at the minimum refractive index point (1.33). From Fig. 3(b), ΔλS.D. is estimated to be 0.4148 nm. Also, the sensitivity at refractive index value 1.33 is 155.56 nm/RIU. Using these values in equation (5), a refractive index resolution of 2.66×10−3 RIU is achieved.
From the obtained value of resolution, LOD of the present refractive index sensor is calculated using the formula [16]
(4)
where, σ represents the resolution which is taken as the wavelength resolution of the UV–Vis spectrophotometer used to perform absorption spectra experiments which is 0.13 nm. From eq. (6), LOD of 5.14×10−3 RIU is obtained at 1.33. Since the strength of interaction between Ta2O5:Si:Graphite tricomposite nanostructure and external refractive index and consequent generation of LSPR signal depends on the volumefilling factor (f) of Si in Ta2O5. Therefore, 'f' is very important to optimize in terms of the sensing performance of the Ta2O5:Si:Graphite. In this section, 'f' have been optimized by measuring the shift attained in peak absorption wavelength corresponding to Ta2O5:Si:Graphite for different values of 'f ' ranging from 0 to 1 in step of 0.1. These experiments for different values of 'f' were performed in same refractive index difference i.e. 1.33 to 1.38. The variation of shift in peak absorbance as a function of 'f' is plotted in Fig. 8. It may be noticeable from figure that as the value of 'f ' increases from 0 to 0.4, shift in peak absorbance wavelength increase. But on further increasing the value of 'f' from 0.5 to 1, shift in peak absorbance wavelength starts decreasing and become minimum at f=1. This trend is found to be in reasonable agreement with PL intensity shown and discussed previously in Fig. 6 (a). The maximum peak in PL intensity is found to be at f=0.4. Therefore it may be justified that the amount of defects level is the reason for absorption of light and hence for the shift in peak absorbance wavelength. It is worthwhile to mention here peak absorbance is observed due to the interaction of photons with that electron cloud results in the generation LSPR signal. When there is difference in refractive indices of the surrounding medium then the shift in peak absorbance wavelength is observed. This phenomenon is depicted well in Fig. 9.