*2.7. Photocatalytic Experiments*

The photocatalysts measurement were carried out in our laboratory, SRM Institute of Science and Technology, Chennai (28◦4 N; 82◦25 E), in April 2020 (8 April to 10 May). Daylight from 9 am to 12 pm was utilized to perform the photocatalytic experiment with an average light intensity of 68.2~89.4 mW/cm2. The photocatalytic properties of the as-synthesized samples were performed using MB dye as a model pollutant. In a typical photocatalytic reaction, 10 ppm of MB dye was added to 50 mL of DI water and stirred for 5 min. The solution was maintained in the dark for 20 min under stirring to achieve an adsorption–desorption equilibrium. At regular time intervals of the photocatalytic dye degradation reaction solution (20 min), 3 mL aliquots of the solution were sampled and UV–Vis spectra were measured.

## *2.8. DSSC Device Fabrication*

Prepared samples, such as TiO2, TiO2/rGO and TiO2/rGO/Ag (0.25 g each), were dispersed in 2 mL of stock solution (prepared by mixing an equal amount of DI water and ethanol) and ground for 15 min. To obtain a paste-like formation, 2 mL of the stock solution was mixed with 200 μL of acetic acid and 100 μL of surfactant (Triton 100-X) during the grinding process. The resultant colloidal solution was then uniformly coated on a fluorine-doped tin oxide (FTO) substrate with optimized conditions using a doctor blade technique. After drying at 100 ◦C for 10 min, the coated substrate was annealed at 450 ◦C for 30 min. The coated FTO was then immersed in dye solution for 12 h, which consisted of 0.03 M dis-tetrabutylammonium cis-bis(isothiocyanato)bis(2,2-bipyridyl-4,4 dicarboxylato) ruthenium (II) (N719). The prepared photoanode was subsequently clamped with a Pt-coated counter electrode to form a sandwich type of device. An electrolyte solution consisting of 0.6 M dimethylpropylimidazolium iodide, 0.1 M lithium iodide, 0.01 M iodine and 0.5 M 4-tert-butylpyridine in acetonitrile was then filled in between the layers of the sandwich-type device.

#### **3. Results and Discussion**

#### *3.1. Structural and Compositional Analysis*

Figure 1a shows the XRD patterns of prepared samples of pristine GO, mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag. A strong and sharp diffraction peak is observed at 9.8◦, which corresponds to the (0 0 1) crystal plane of GO. Using Bragg's equation, the interlayer distance of the as-prepared GO sheet was estimated to be 0.9 nm, whereas graphite powder shows an interlayer distance around 0.33 nm [30]. The XRD pattern obtained for GO indicates that the bulk graphite is successfully reduced as rGO nanosheets [31]. Furthermore, the characteristic peaks of TiO2 are observed at 25.28◦, 37.97◦, 47.95◦, 53.84◦, 55.02◦, 62.40◦, 68.70◦ and 75.20◦, which are assigned to the (1 0 1), (0 0 4), (2 0 0), (1 0 5), (2 1 1), (2 0 4), (1 1 6) and (2 1 5) planes, respectively. In the case of TiO2/rGO/Ag, the peaks observed at 38.12◦, 44.26◦, 64.33◦ and 77.35◦ are associated with the cubic phase of the Ag (1 1 1), (2 0 0), (2 2 0) and (3 1 1) planes (JCPDS: 04-0783). Therefore, the obtained XRD pattern of the TiO2/rGO/Ag composite reveals the coexistence of rGO, TiO2 and Ag materials and confirms the effective formation of a hybrid nanostructure.

**Figure 1.** (**a**) XRD patterns, (**b**) Raman spectra of GO, mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag. XPS core level spectra of the (**c**) C1s spectra of GO; (**d**) C 1s spectra of TiO2/rGO; (**e**) Ti 2p spectra of mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag; (**f**) Ag 3d spectra of TiO2/rGO/Ag.

To explore the structural characteristics, Raman spectroscopy was performed for GO, TiO2, TiO2/rGO and TiO2/rGO/Ag samples, as shown in Figure 1b. The obtained Raman spectrum of pristine GO revealed the D band at 1350 cm−<sup>1</sup> due to the disordered nature of the graphene structure with sp<sup>3</sup> defects (as given in Figure S1). In addition, the G band (at 1587 cm−1) was observed in the Raman spectrum of pristine GO, which was attributed to the in-plane vibrations of C–C stretching in graphitic materials, as well as the doubly degenerate phonon mode in the Brillouin zone [32]. As shown in the inset image of Figure 1b, the G band shifted from 1587 to 1599 cm−1, which is direct evidence of the effective reduction of GO to rGO with a constant D band for TiO2/rGO [33]. In comparison

with pristine GO, the ID/IG intensity ratio of the TiO2/rGO and TiO2/rGO/Ag samples was slightly reduced to 0.83 from 0.87 (as shown in Supporting Information Figure S1). However, the reduction in the ID/IG ratio indicates a considerable reduction in the sp<sup>2</sup> domain size of carbon atoms, as well as the reduction of sp<sup>3</sup> to sp2 [34]. In the case of TiO2/rGO/Ag, the Raman vibrational modes were observed at 147, 199, 396, 515 and 639 cm−1, which corresponds to the Eg, Eg, B1g, A1g and Eg modes of anatase TiO2 [35]. On the other hand, the broadening of the Eg mode (147 cm−1) with a considerable peak shift revealed the presence of TiO2 mesospheres, as well as Ag NPs, on multilayer rGO. Furthermore, there were no significant changes in the observed Raman scattering modes of the plasmonic hybrid nanostructures compared with the hybrid sample (TiO2/rGO). Nevertheless, the single Raman characteristic peak of the anatase TiO2 phase was considerably enhanced due to the LSPR properties of Ag NPs on the TiO2/rGO/Ag hybrid surface [36]. The obtained Raman characteristic modes further confirm the successful formation of plasmonic hybrid nanostructures.

XPS measurements were performed to evaluate the chemical state and interaction of the prepared hybrid nanostructures and pristine GO samples. The surveyed spectra of TiO2, TiO2/rGO and TiO2/rGO/Ag indicate the presence of Ti, O, Ag and C elements without any impurities in the prepared hybrid nanostructures (as given in Supporting Information Figure S2). The deconvoluted C 1s spectrum of the as-prepared GO sample is shown in Figure 1c as follows: (i) At 282.55 eV, C bonds with sp1 carbon atoms [37], (ii) C=C bonds denote the aromatic sp2 structure groups present in the GO at 284.66 eV, (iii) the peak at 286.30 eV represents the carboxyl groups (C–O bonds) including epoxy and hydroxyl group [38]. In the case of the TiO2/rGO sample, the deconvoluted XPS peak of the C 1s spectrum shows the four diverse carbon bonds with various binding energies of 283.18 eV, 284.74 eV, 286.07 eV and 288.72 eV. Thus, the peak that appears at 283.18 eV indicates the chemical bonding between the C atom and the Ti atom (Ti–C) [34]. Furthermore, the peaks with reduced intensity emerges at 284.74 eV, 286.07 eV and 288.72 eV with extensively reduced intensity purely attributed to C=C, C–O, O=C–OH bonds as oxygenated functional groups, which highlights the formation of rGO from GO [39]. Figure S3 shows the deconvoluted O 1s spectrum with two bands at 529 eV and 530 eV that correspond to the Ti–O–Ti and Ti–O–C groups. There is a slight peak shift in the TiO2/rGO sample from 530.95 eV to 531.12 eV, which indicates the strong binding of rGO with TiO2 mesosphere during the solvothermal process [40,41]. The chemical states of Ti species were thoroughly investigated using the XPS spectra, as shown in Figure 1e. The core level peak of Ti is split into a doublet peak observed at 458.40 eV and 464.10 eV, which corresponds to Ti 2p3/2 and Ti 2p1/2 with a splitting energy of 5.7 eV, mainly attributed to the Ti4+ state for the TiO2/rGO/Ag samples [42,43]. After infusing the Ag NPs into the hybrid nanostructure, the doublet peaks (Ti 2p3/2 and Ti 2p1/2) of the Ti 2p state have observed notable shifts to higher order, such as 458.40 eV to 458.51 eV (Ti 2p3/2) and 464.10 eV to 464.27 eV (Ti 2p1/2). As a result, a positive shift in the doublet peak of the Ti species directly reflects the alteration of the Fermi level, whereas the Ag species holds a lower Fermi level offset than that of TiO2. Therefore, the resultant Fermi level alignment in the plasmonic hybrid nanostructures are more favorable for fast electron transfer processes at the rGO/TiO2/Ag interface [44]. Figure 1f shows the doublet peaks of the Ag element at binding energies of 367.23 eV and 373.31 eV, which correspond to Ag 3d5/2 and Ag 3d3/2 with a splitting energy of 6.12 eV, which confirms the presence of the Ag NPs in the prepared TiO2/rGO/Ag hybrid nanostructures [45].

#### *3.2. Morphological Analysis*

The surface morphology features of GO nanosheets, TiO2 mesospheres, TiO2/rGO and TiO2/rGO/Ag hybrid nanostructures were investigated using HR-SEM, TEM and HR-TEM measurements, as shown in Figure 2. The transparent ultrathin GO nanosheets are visible in the HR-SEM and HR-TEM micrographs, as shown in Figure 2(a1–a3). As-prepared TiO2

mesosphere samples are agglomerate-free, uniformly distributed and spherical shaped, with an average size of ~570 nm (see Figure 2(b1)).

**Figure 2.** HR-SEM, TEM and HR-TEM images of prepared samples of GO (**a1**–**a3**), TiO2 (**b1**–**b3**), TiO2/rGO (**c1**–**c3**) and TiO2/rGO/Ag (**d1**–**d3**).

The surface morphology and atomic interplanar distance of prepared samples were investigated using TEM and HR-TEM, as shown in Figure 2(b2,b3). Figure 2(c1–c3) shows the uniform TiO2 mesospheres with an interplanar distance of 0.35 nm for the pure anatase TiO2 phase over the rGO nanosheet surfaces. The presence of Ag NPs embedded in the TiO2 mesosphere-coated rGO nanosheet surfaces is revealed by the TEM and HR-TEM images of the hybrid nanostructures. The interplanar distance of the prepared hybrid nanostructures was estimated from the HR-TEM image using a line profile analysis, as shown in Figure 2(d3). The calculated interplanar distance values were 0.35 nm and 0.23 nm for the anatase TiO2 crystal plane of (1 0 1) and the (1 1 1) crystal plane of the Ag NPs, respectively, which is well matched with the standard JCPDS cards (Nos. 21-1272 and 04-0783) [46]. Moreover, the elemental mapping and energy dispersive X-ray spectroscopy (EDX) spectrum of the prepared hybrid nanostructures indicate the coexistence of C, Ti, O and Ag elements without the presence of any impurities in the resultant hybrid sample (as shown in Supporting Information Figure S4). Therefore, the resultant surface morphology

of the prepared hybrid ternary nanostructures indicates that the plasmonic Ag NPs are uniformly embedded into the TiO2 mesospheres anchored in the rGO nanosheet surfaces. Moreover, the thickness of the fabricated photoanode was estimated to be 14.2 μm (TiO2) and 13.8 μm (TiO2/rGO/Ag) and presented Figure S5a,b.

#### *3.3. Optical and Surface Area Analysis*

UV-DRS was used to investigate the optical absorption of the prepared TiO2, TiO2/rGO and TiO2/rGO/Ag samples and is shown in Figure 3a. Due to the electronic transition of O 2p to Ti 3d, all of the prepared samples exhibit a typical optical absorption edge located at 392 nm of TiO2, which corresponds to a bandgap energy of 3.3 eV for TiO2 [47]. After the introduction of Ag NPs over the TiO2/rGO surface, the optical absorption of the hybrid nanostructure is significantly higher in the visible region compared to the other samples, due to the LSPR of Ag NPs on the TiO2/rGO surface [48]. Overall, the improved visible-light-region absorption observed for the prepared hybrid nanostructures indicates a greater number of photons harvested from the visible-light region, which is favorable for efficient photocatalysts and photoanodes for solar energy conversion devices.

**Figure 3.** (**a**) UV-DRS absorption spectra, (**b**) PL spectra of mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag. Nitrogen adsorption–desorption isotherms of (**c**) surface area and corresponding (**d**) pore size distribution of prepared samples TiO2, TiO2/rGO and TiO2/rGO/Ag.

PL emission spectroscopy measurements are widely used to understand the desired photon-induced charge separation properties of the prepared samples. PL spectra of the prepared samples with an excitation wavelength of 300 nm and in the range from 350 to 500 nm are shown in Figure 3b. A strong band edge emission peak is observed at around 389 nm for all the prepared nanostructure samples [49]. Another emission peak centered in the visible region of 480 nm indicates the presence of oxygen vacancies on the TiO2 surface. For the plasmonic hybrid nanostructures, the resultant emission peaks are significantly quenched, which is a direct indication of the reduction of the photon-induced electron-hole pair recombination rate that facilitates electron transfer at the TiO2/rGO/Ag interface [50,51]. The function group of GO and prepared materials were analyzed by FTIR spectrum [52,53] (Figure S6).

The surface area and pore size distribution of the prepared samples were evaluated using the BET and BJH methods shown in Figure 3c,d. In comparison, the specific surface area of Ag NP-based hybrid nanostructures have a surface area of 297.71 m2/g due to TiO2 mesosphere formation with the Ag NP-coated rGO nanosheets. The estimated surface areas of the TiO2 mesosphere and TiO2/rGO samples were 234.16 m2/g and 281.31 m2/g, respectively. As shown in Figure 3d, the estimated pore sizes were 2.15 nm, 1.52 nm and 1.91 nm for the TiO2, TiO2/rGO and TiO2/rGO/Ag hybrid nanostructures, respectively. The enhanced specific surface areas and pore size distributions of the hybrid nanostructures are more beneficial for the photoanodes in DSSC assemblies and photocatalysts for photocatalytic reactions.

#### *3.4. Performance of DSSC Device*

The influence of the LSPR properties of Ag NPs on the photovoltaic characteristics of a DSSC based on a TiO2/rGO/Ag photoanode under simulated sun irradiation with an AM 1.5 G filter was investigated. The current density vs. voltage characteristics of DSSC devices assembled with various photoanodes (TiO2, TiO2/rGO and TiO2/rGO/Ag) are given with error bars in Figure 4a. Compared with the TiO2 and TiO2/rGO photoanodes, the TiO2/rGO/Ag photoanode show superior photovoltaic performance with a maximal *JSC* of 16.05 mA/cm<sup>2</sup> and an improved open circuit voltage (*Voc*) of 0.74 V, with a fill factor (*FF*) of around 62.50%. In addition, the hybrid-nanostructure-based plasmonic DSSC device achieved a higher power conversion efficiency of 7.27%, which is 1.7 times higher than that of a pristine TiO2 photoanode (4.01%). An overall comparison of a fabricated DSSC device's photovoltaic characteristics are given in Table 1 and SI. Figure 4b shows the IPCE of all of the fabricated DSSC devices in the range of 400 nm to 800 nm. The Ag-based DSSC device exhibited the maximum IPCE owing to the presence of N719 dye adsorption. The IPCE of the TiO2/rGO/Ag photoanode-based plasmonic DSSC device achieved a maximum value of ca. 77.82% at an incident wavelength of 550 nm due to improved photon harvesting efficiency and a higher number of electron extraction from the N719 dye molecule than the other devices [27,28,52]. The photovoltaic performance of the fabricated TiO2/rGO/Ag photoanode-based plasmonic DSSC cell has been compared with a recent report, as illustrated in Figure 4c [54–59]. A tentative operating mechanism of the proposed plasmonic DSSC device is illustrated to understand its enhanced photovoltaic performance. As shown in Figure 4d,e, upon 1 sun irradiation the electrons are excited from higher occupied molecular orbitals (HOMO) to lower occupied molecular orbitals (LUMO) levels of the dye molecule (N719). Under this irradiation, if the incident light coincides with the LSPR effect of Ag NPs (work function (WF) = −4.2 eV), then the electrons close to the Fermi level are excited into a higher energy state by receiving the energy from plasmon resonance, known as a hot electron, via non-radioactive process [60]. These electrons have sufficient energy to break the Schottky barrier formed between the TiO2 mesosphere (WF = −4.4 eV) and Ag NPs and can flow through the conduction band of the TiO2 mesosphere. Moreover, it is possible that this hot electron can be accepted and shuttled via rGO nanosheets (WF = −4.4 eV) [61–63]. The superior conductivity of rGO nanosheets could assist in accepting or transmitting electrons generated by TiO2 or Ag NPs to the appropriate photoanode. The hybrid nanostructure attained higher efficiency than the TiO2 mesospheres for the following reasons. (i) The large surface area provides

high dye loading and efficiently scatters the incoming light within the device. (ii) The rGO nanosheets promote extraordinary charge transport where electrons come from TiO2 mesospheres or Ag NPs and limit the electron-hole recombination rate. (iii) The LSPR effect of the Ag NPs extends the light absorption range from the UV to the visible region.

**Figure 4.** (**a**) Current density–voltage characteristics (with error bar), (**b**) IPCE spectra for different photoanode-based DSSC devices (mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag) measured under one sun illumination, (**c**) compared efficiency of recent reported photoanode materials with TiO2/rGO/Ag device performance, (**d**) schematic representation of the charge transfer process influenced by a plasmonic hybrid nanostructure (TiO2/rGO/Ag)-based DSSC device and (**e**) DSSC device consists of hybrid photoanode and energy level diagram.

**Table 1.** Comparison of the photovoltaic properties and surface area of the DSSC device based on the mesosphere TiO2, TiO2/rGO and TiO2/rGO/Ag photoanodes measured under AM 1.5 G one sun illumination.


Figure S7 shows the transient photocurrent response of the TiO2/rGO/Ag composite under 1 sun light irradiation with the highest photocurrent density, which is approximately one-fold greater than that of TiO2/rGO. This confirms that the incorporation of Ag into the TiO2/rGO/Ag nanocomposite not only enhances the light harvesting but also significantly expedites the photon-induced charge carrier separation and transport properties of Ag NP-coupled hybrid composites. The transient photocurrent response of the TiO2/rGO/Ag composite was also performed under a UV filter, which confirmed the plasmonic Ag response photocurrent in the visible region. These results confirm that Ag could act as an acceptor of the photogenerated electrons by TiO2/rGO and encourage fast charge transportation due to the high metallic conductivity, which effectively suppresses the charge recombination in the composites.

#### *3.5. Photocatalytic Performance*

To demonstrate the photocatalytic MB dye degradation process, TiO2-, TiO2/rGO- and TiO2/rGO/Ag-based photocatalysts were employed under natural sunlight irradiation for three repetitions as shown in Figure S8. Figure 5a–c shows dye degradation profiles with MB as the pollutant with different irradiation times under natural sunlight for all of the prepared photocatalysts. The maximum absorption peak for the photocatalytic dye degradation process in the presence of a photocatalyst is displayed at around 664 nm due to the absorbance characteristics of MB dye. Under dark conditions, there were no significant changes in the UV–vis absorption spectra of all the prepared photocatalysts. The UV–vis absorption spectra of all the prepared photocatalysts were successfully recorded with a time interval of 20 min under continuous natural sun irradiation. The hybrid nanostructures of the TiO2/rGO/Ag photocatalyst exhibited a higher dye degradation efficiency of 93% under continuous natural sunlight irradiation for 160 min. Compared with pristine TiO2, the enhanced dye degradation efficiency of the plasmonic hybrid photocatalyst was almost 1.3 times higher due to the LSPR properties of the Ag NPs. On the other hand, the presence of rGO/Ag on the TiO2 surface creates a more favorable environment for fast photon-induced charge separation and transfer with an extended lifetime of charge carriers, whereas Ag NPs act as electron sinks for improved photocatalytic dye degradation [64].

**Figure 5.** (**a**–**c**) UV absorption spectra of MB degradation with different interval time under natural sunlight, (**d**) plots of C/C0 as a function of time (min) towards the photo degradation of MB, (**e**) TOC analysis was observed for TiO2/rGO/Ag at different intervals and the (**f**) photocatalytic degradation of MB in the presence of different scavengers of the TiO2/rGO/Ag sample under natural sunlight.

To further understand the effect of dye degradation in the form of mineralization efficiency, TOC analysis of the TiO2/rGO/Ag photocatalyst was performed with respect to various time intervals, as shown in Figure 5e. After 160 min of natural solar irradiation, the mineralization efficiency of the TiO2/rGO/Ag photocatalyst in terms of carbon content removal in MB dye was approximately 80.35% (Figure 5e). The improved TOC of the plasmonic hybrid TiO2/rGO/Ag photocatalyst suggests that it efficiently mineralized MB dye into residual organic molecules, such as CO2 and H2O, in the MB dye solution. Furthermore, the photocatalytic MB dye degradation activity of TiO2/rGO/Ag was evaluated

by the addition of various radical scavengers (Ag, EDTA, IPA and BQ) to determine the more active species e−, h+, OH<sup>−</sup> and O2<sup>−</sup> in the degradation system [65,66]. Figure 5f shows various scavenger photocatalytic MB dye degradation (MB) experiments with the TiO2/rGO/Ag system as a photocatalyst. For this study, 2 mg of various scavengers were mixed with MB dye solution in the presence of a prepared photocatalyst (15 mg) and exposed to natural sunlight for 160 min. The photocatalytic MB dye degradation efficiency of TiO2/rGO/Ag was significantly inhibited by 58.89% and 23.00% in the presence of IPA and BQ as scavengers, respectively. This reduced the photocatalytic dye degradation efficiency, and it reveals that OH<sup>−</sup> and O2<sup>−</sup> radicals are the main active species during the degradation of MB by TiO2/rGO/Ag under natural sunlight [67].

A possible mechanism for the photon-induced charge carrier separation and transfer in the prepared photocatalyst surface is depicted in Figure 6. The valence band (*EVB*) and conduction band (*ECB*) potentials of TiO2, TiO2/rGO and TiO2/rGO/Ag were estimated using the Mulliken electronegativity theory [53] and the following equations:

$$E\_{VB} = \text{\textdegree } -E^{\varepsilon} + 0.5E\_{\text{\textdegree}} \tag{1}$$

$$E\_{\mathbb{C}B} = E\_{VB} - E\_{\mathbb{S}} \tag{2}$$

where *x* is the electronegativity (5.81 eV for TiO2), *ECB* is the conduction edge potential, *EVB* is the valence band edge potential, *Ee* is the free energy of the electrons in the reversible hydrogen scale (4.5 eV), and *Eg* is the bandgap of the material. The *EVB* and *ECB* potentials of the TiO2/rGO/Ag system were estimated to be −0.34 eV and 2.96 eV, respectively, while *EVB* and *ECB* for pure TiO2 were estimated to be −0.32 and 2.94 eV, respectively.

**Figure 6.** Schematic illustration of the possible mechanism of rGO, Ag NPs roles with mesosphere TiO2 in MB photocatalytic degradation under natural sunlight irradiation.

Figure 6 shows the proposed photocatalytic mechanism for the natural-sunlightresponsive TiO2-RGO-Ag nanostructure. After natural sunlight irradiation, the Ag NPs' dipolar characteristics of the SPR effect enable electrons generated from the Ag NPs to migrate to the CB of TiO2. In contrast, electrons in the CB of TiO2 could be transferred to the surface of the RGO nanosheets due to the favorable electron affinity of TiO2 and the lower WF of RGO. Further, it is postulated that the photogenerated electron from Ag particles can also be transferred into the RGO sheets. TiO2- and plasmon-excited Ag particles serve as electron transfer channels, which ensure charge separation efficiency. The oxidized Ag<sup>+</sup> species accept electrons from water molecules (H2O) or OH<sup>−</sup> adsorbed on the TiO2 surface or from the dye molecules present in the solution and are regenerated. The reaction with H2O or hydroxide ions produces hydroxyl radicals (OH). These radicals (O2 and OH) are influential oxidizing agents for the degradation of MB dye molecules. Therefore, the

prepared ternary nanostructures offer superior photocatalytic dye degradation (MB) due to efficient photoinduced charge carrier separation [52,53].

#### **4. Conclusions**

In this study, TiO2/rGO/Ag hybrid nanostructures were successfully synthesized using a combination of solution processes and in situ growth and were then employed as photoanodes for DSSCs and as catalysts for photodegradation applications. The plasmonenhanced DSSC devices demonstrate enhanced photovoltaic performance of 7.27% along with a higher short-circuit current of 16.05 mA/cm2 and an IPCE efficiency of 77.82% at 550 nm. The results suggest that the high photovoltaic performance of the plasmon-based TiO2/rGO/Ag device can be attributed to (i) the large specific area of TiO2/rGO/Ag, which leads to high dye loading; (ii) TiO2 mesospheres enhancing the light scattering effect of incoming light; and (iii) the incorporation of Ag NPs facilitating more induced photons and fast electron transport in the device. Upon natural sunlight irradiation, the prepared hybrid nanostructure shows an improved photocatalytic degradation of MB by 93% within 160 min, and the effects of different scavengers on the obtained photocatalytic activity were systematically investigated. The effects of optimum active surface area, the LSPR properties of Ag NPs and the enhanced electrical conductivity of the prepared ternary nanostructures combine to provide an enhanced visible-light-driven plasmonic DSSC device and photocatalyst for dye degradation (MB). The proposed plasmonic and hybrid-based nanostructures demonstrate an emerging strategy to establish large-scale applications of solar energy conversion technologies.

**Supplementary Materials:** The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/nano13010065/s1. Figure S1: Raman spectra of prepared GO nanosheets. Figure S2: XPS survey spectrum of TiO2, TiO2/rGO and TiO2/rGO/Ag. Figure S3: XPS O 1s spectra of (a) TiO2 (b) TiO2/rGO and (c) TiO2/rGO/Ag. Figure S4: (i) STEM elemental mapping and (ii) EDX spectrum of TiO2/rGO/Ag hybrid ternary nanostructures. Figure S5: The HR-SEM cross section of (a) TiO2 (b) TiO2/rGO/Ag photoanode. Figure S6: FTIR spectra of GO, TiO2, TiO2/rGO and TiO2/rGO/Ag. Figure S7: Transient photocurrent spectra of TiO2/rGO and TiO2/rGO/Ag (under with and without UV filter). Figure S8: (a–c) Three different trials of UV absorption spectra of MB degradation with different intervals under natural sunlight. (d) Plots of Ln (C/C0) as a function of time (min) towards the photo degradation of MB. References [52,53] are cited in supplementary materials.

**Author Contributions:** S.A.—designed the experiment procedure and wrote the paper, V.S.M.—data analysis, S.K.H.—draft, data analysis, conceptualization, K.S.—formal analysis, S.G.—original draft preparation, H.I.—resources, M.N.—methodology, J.A.—draft, data analysis, funding acquisition, supervision. All authors have read and agreed to the published version of the manuscript.

**Funding:** The financial support by "DAE-BRNS Young Scientist Research Award" No.34/20/02/2017- BRNS/34277 awarded to Dr. J. Archana to carry out the research work is gratefully acknowledged. The authors are thankful to SRMIST, Chennai, India for their SEED/START UP research grant and constant support for the research.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

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

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
