1 Introduction

TiO2 films composed of nanoparticles in the range from 10 nm to 30 nm can have a roughness factor (the actual surface area to the apparent one) reaching 1000 [1]. Therefore, the amount of dye adsorbed in a DSSC is drastically increased due to the much larger accessible internal surface when the mesoporous TiO2 is used. In fact, the space-filling capability of the dye can reach almost 100% over a wide range in the visible region [1]. By virtue of comparison, the amount of adsorbed dyes on the surface of single-crystal and poly-crystal materials is quite small, causing only about 1% absorption even at the peak wavelength [1]. Such properties nominate mesoporous TiO2 to be integrated in various optoelectronic applications [1,2,3,4,5,6]. One of these applications as an active layer in dye-sensitized solar cells (DSSC) [6,7,8].

The absorption mechanism in the DSSC is mainly rely on the visible absorption due to the absorbed dye, as TiO2 is an UV absorber (3.2 eV) [4, 9]. Consequently, the reported DSSC efficiency is rarely reaching 10% [3, 10, 11]. Previous attempts in the literature acted to enhance the DSSC efficiency either by adding new trapping structures or by customizing the optical properties of the mesoporous TiO2 layer. In such context the AgVo3-doped TiO2 thin film can be deposited using a simple sol–gel method [3]. Such doping can narrow down the energy band gap, turning mesoporous TiO2 to be a visible light absorber. Concerning the used dye, previous attempts were demonstrated with commercial N719 dyes. [1, 12, 13]. Alternatively, other attempts replaced the inorganic N719 dye with a low-cost organic dye based on beetroot. Organic beetroot dye recorded acceptable sustainability in previous attempts in the literature as well as matched optical spectrum with mesoporous TiO2 [14, 15].

The work investigated in this paper demonstrates the impact of AgVO3 doping on a mesoporous TiO2 layer used in organic-based dye-sensitized solar cells. Experimentally segmented and complete solar cells have been fabricated and characterized to evaluate the effect of doping on the overall conversion efficiency. The fabricated cell shoed acceptable output characteristics with 8.6 mW/cm2 harvested power density, as well as Voc of 0.514 V and Jsc of 20.9 mA/cm2. Herein, our fabricated cell can be used to harvest light for satisfying the energy demands of a micro/nano-sensing unit. The low-power sensor was normally operated via the connected DC battery (cf. Fig. 1).

Fig. 1
figure 1

a Architecture of our DSSC b Diagram showing the location of the DSSC in our underwater light harvesting unit

Full size image

2 Experimental work and characterization

This section describes the fabrication process for a complete DSSC with organic dye and various AgVO3-doping concentrations, applying a bottom-up approach. In addition, all morphological, optical, and electrical characterization procedures are explained.

2.1 Materials

Titanium (IV) Iso-propoxide (C12H28O4Ti, 97%), carboxy methyl cellulose sodium salt (CMC), and Isopropyl alcohol (C3H8O) were purchased from Sigma Aldrich for preparing TiO2 powder. Silver nitrate (AgNO3) and sodium vanadate (NaVO3) were purchased from Merck and used in preparation of the doped material. Acetonitrile (CH3CN), sodium hydroxide (NaOH), ethylene glycol (C2H6O2), fluorescent dye, fluorine-doped tin oxide (FTO), iodine (I) G. R. grade, 99.8%, and potassium iodide (KI), were also purchased from Sigma Aldrich. N719 dye as well as natural dye, extracted from “Beet” and has a dark reddish-purple color, were used.

2.2 TiO2 nanoparticles preparation and deposition

TiO2 nanoparticles were prepared using the sol–gel method. 40 mL of titanium (IV) isopropoxide and 400 mL of isopropyl alcohol (IPA) were mixed on a magnetic stirrer for 30 min at a speed of 100 rpm. Next, distilled water was added (drop by drop using a pipette) until a gel started to form. The solution was left for 2 days before it was later filtered, dried, grinded, and then calcinated for 3 h at 500 °C.

For paste preparation, 2.5 g of the prepared nanoparticles were added to 3 mL of CMC. The mixture was then grinded until it achieved the required homogeneity. Finally, 0.75 mL of IPA was added to the mixture and grinded again. The film was screen printed on a glass substrate, as described in our previous work in [1, 12, 13].

2.3 Ag doing

Silver vanadate (AgVO3) was produced from the reacting sodium vanadate (NaVO3) and silver nitrate (AgNO3): “NaVO3 + AgNO3 → NaNO3 + AgVO3.” NaVO3 (1.5 mg) and AgNO3 (2.09 mg) were placed with 50 mL of distilled water on the magnetic stirrer until the crystals were completely dissolved in water and the solution was colorless. Next, the sodium vanadate solution was added drop by drop to the silver nitrate solution while stirring for 2 h, until it turned into a yellowish mustard color. Finally, after placing the mixture in a probe sonicator for 2 h, it was filtered, dried, and then grinded.

TiO2 was doped with different concentrations of AgVO3, varying from 0% to 25%. Powders were mixed in a beaker with 50 mL distilled water. Next, the mixture placed in a probe sonicator for 1.5 h and was again filtered, dried, and then grinded. Doped-TiO2 pastes were then prepared using the same recipe mention in Sect. 2.2 and were deposited on glass using the screen-printing technique.

2.4 Segmented and complete DSSCs fabrication

In comparison with our previously published work in [1, 12, 13, 16], we have replaced the commonly used N719 dye with an organic natural dye (detailed recipe is listed Table 1). Moreover, mp-TiO2 was replaced by an AgVO3-dopped mp-TiO2 layer, as discussed in the previous section. Herein, beetroot was used as a natural organic dye, which has a dark reddish purple color and has good sunlight absorption capabilities [6]. Beetroot-based dye was chosen due to its matched light spectrum, acceptable lifetime as well as significantly reduced cost with respect to commercial N719 dye. For example, one gram of N719 approximately costs 330 euros, while the extracted organic dye costs less than 5% of this.

Table 1 Recipe and preparation of each layer for DSSCs
Full size table

The extraction process was initiated by first rinsing the beetroot with water. They were later peeled and chopped to smaller parts. In a pan, distilled water was boiled with the chopped beetroot for around 15 min. After cooling down, the beetroot was squeezed using a hand blender until it turned into a paste. The observed consistency was very thick. A paper filter was then placed in a conical flask with a funnel. By adding pressure, the liquid beet natural dye was extracted, filtered, and ready for use.

Seeking for a bare cell, yellowish green N719 dye was prepared. The prepared concentration of the synthetic dye was 2 × 10–5 Molar. The medium is alkaline by using NaOH to increase the solubility of the dye in distilled water (see Fig. 2, green color in Fig. 2b is due to the I3−/I− electrolyte). All DSSCs were sealed using a hot press at a temperature of 120 °C.

Fig. 2
figure 2

Thin layer of mp-TiO2 immersed in a beet natural dye b N719 dye

Full size image

2.5 Characterization

The crystalline size and structure of the prepared nanoparticles were estimated by P Analytical X’PERT MPD diffractometer using (Cu [Kα1/Kα2]) radiation. The diffraction angle ranges from 10° to 90° with a step of 0.02° with integration time of three s/step were utilized. The morphologies of the samples were analyzed using high-resolution transmission electron microscopy (SEM JEOL 6340). To electrically characterize the fabricated DSSCs, an LED-based solar simulator with an integrated NIR-UV–Vis spectrometer and Keithley 2401 current–voltage source meter were used [17]. The optical transmission spectra of the fabricated as well as semifinished DSSCs are measured through a V-770 UV–Visible/NIR spectrophotometer that was used, which has a wavelength range from 190 nm to 2700 nm.

3 Results and Discussions

3.1 TiO2 nanoparticles characterizations

Following the AgVO3-doped TiO2 nanoparticles recipe introduced in Sect. 2, XRD analysis was carried to investigate the crystal structure of the prepared nanoparticles. XRD was conducted for TiO2, AgVO3, TiO2 doped with 5% to 25% to ensure that the required structures were formed. Figure 3 shows the XRD patterns of TiO2, TiAg5, and TiAg15. The XRD pattern of the nanocomposite shows the existence of broad peak revealing the reduction in Titania crystallites upon incorporation of bismuth vanadate. The diffraction pattern of pure Titania shows several peaks at 2θ = 25.3°, 36.9°, 37.7°, 38.5°, 48°, 51.9°, 53.8°, 55.1°, 62.6°, 68.7°, and 75° [JCPDS No. 71-1167 were a = 3.786 Å and c = 9.507 Å] referred to Titania anatase phase. These peaks were indexed in order of increasing diffraction angles indicated a body-centered tetragonal crystalline structure of Titania crystal. The XRD pattern of the nanocomposite shows the existence of broad peak revealing the reduction in Titania crystallites upon incorporation of bismuth vanadate. This result reflects the strong chemical interaction between AgVO3 and anatase TiO2 to form titanium vanadium hydroxide.

Fig. 3
figure 3

XRD patterns of TiO2, TiAg5, and TiAg15. The diffraction pattern of pure Titania shows several peaks at 2θ = 25.3°, 36.9°, 37.7°, 38.5°, 48, 51.9°, 53.8°, 55.1°, 62.6°, 68.7°, and 75° [JCPDS No. 71–1167 were a = 3.786 Å and c = 9.507 Å] referred to Titania anatase phase. These peaks were indexed in order of increasing diffraction angles indicated a body-centered tetragonal crystalline structure of Titania crystal

Full size image

3.2 TiO2 thin-film characterizations

Using the recipe discussed in Sect. 2, TiO2 paste was deposited on microscopic glass substrates. Three sets of samples were prepared using TiO2 paste as well as AgVO3 doped. SEM characterization analysis is carried out for the deposited thin layer to determine the surface characteristics, size and shape of particles, composition, and arrangement of atoms in the sample. Measurements were conducted for TiO2 and TiO2 doped with 5% to 25%. As shown in Fig. 4, the anatase titanium dioxide mesoporous layer which is prepared by sol–gel method [1, 12] has a homogenous structure and is highly crystalline with spherical agglomerated particles.

Fig. 4
figure 4

SEM for a and b TiO2, c and d TiO2 doped with 5% AgVO3, and e and f TiO2 doped with 15% AgVO3

Full size image

EDX characterization analysis is attached in Fig. 5. EDX characterization analysis is carried out to determine and identify the elements present in the sample with information about their percentages (see Fig. 5). Analyses illustrate the elements presented in TiO2 are Ti and O, while the elements presented in TiO2 doped 5% and 15% are Ti, O, Ag, and V, with the indicated percentages in the graphs (c.f. Fig. 5b and c).

Fig. 5
figure 5

EDX for a TiO2m b TiO2 doped with 5% AgVO3, and c TiO2 doped with 15% AgVO3

Full size image

3.3 Optical characterizations

We compared the absorption spectrum of beetroot dye standard N719 dye (see Fig. 6). For the N719 commercial dye, double absorption peaks were recorded (at 289 nm and 456 nm), which agrees with previous reported data in the literature in [18]. The variation in the wavelength allocation of the two detected peaks in the N719 spectrum as well as the spectrum sharpness can be attributed to the dye treatment method along with the dilution percentages. It can be observed that the natural beetroot dye, with peak absorption around 550 nm (centralized visible absorption spectrum), is more appropriate for the proposed DSSC compared with N719 dye. This is due to the enhanced absorption of the doped TiO2 layer, making it capable of absorbing near ultra-violet photons in the visible portion of the spectrum (350 nm to 500 nm). Accordingly, the integration of the AgVO3-doped TiO2 layer with the beetroot natural dye will broaden the absorption spectrum of the effective active layer in DSSC.

Fig. 6
figure 6

Absorption spectra for a N719 and b Beetroot dye

Full size image

3.4 Electrical characterizations

The J–V characteristic curves for the fabricated DSSCs are shown in Fig. 7. In the current study, we investigated the impact of both AgVO3 doping as well as the beetroot dye. First, it can be observed that the higher the doping rate, the higher the short-circuit current density. Maximum is obtained when the doping rate reached 15% with current density of 20.9 mA/cm2. This condition represents the maximum efficiency of around 8.6% with total extracted power density around 8.6 mW/cm2. Consequently, by referring to the bare cell with N719 dye, still the beetroot dye recorded higher current as well as efficiency with respect to N719 DSSC. Beetroot dye showed an enhanced performance by nearly 10% while considering N719 dye as a reference keeping the doping level constant in both as 15% (cf. Fig. 7). This agrees with the interpretation proposed earlier related to the matching in the spectra between organic beetroot dye and the mesoporous TiO2 layer with narrowed energy band gap. Herein, a merged absorption profile with visible absorption overlapping is resulted, which can maximize the absorption in the active layer. Accordingly, we can describe the proposed DSSC as a double absorption effect DSSC, where both the AgVO3-doped TiO2 and the dye are contributing to the absorption mechanism. All output characteristic parameters are listed in Table 2.

Fig. 7
figure 7

J-V characteristic curves for fabricated DSSCs with N719 dye and beetroot dye associated with variable doping level of AgVO3 in the mesoporous active layer

Full size image
Table 2 J-V characteristic outputs for fabricated DSSCs with N719 dye and Beetroot dye associated with variable doping level of AgVO3 in the mesoporous active layer. Comparison is made with previous reported data in [13, 15, 19]
Full size table

By referring to the literature, we can figure out that the measured efficiency with 15% doping and beetroot dye enhanced the efficiency of the proposed DSSC from 2.1% (see reported efficiency in [13], under DSSC-A samples) to nearly 8.6%. Other literature attempts recorded higher references [15, 19]; however, the reported efficiencies were still below 8.6%, see Table 2. Our proposed cell can be categorized as a very low cost, simple, and low-temperature fabricated cell. Such conservations promoted organic dye-based DSSCs in various micro/nanopower applications. Finally, the sustainability performance analysis of these devices needs to be investigated in the future.

4 Conclusion

This work presents a boosted 6.1% DSSC through narrowing the energy band gap of the active mesoporous TiO2 layer as well as integrating a natural dye. Mesoporous TiO2 is treated with 15% of AgVO3 to reach an energy band gap of 2.6 eV. The optimized cell with energy harvesting density of 8.6 mW/cm2 was recorded. The proposed cell showed enhanced performance with respect to a bare cell fabricated with N719 dye as well as similar recipes reported data in the literature. The harvested output power is capable of operating various IoT low-power applications towards a self-power sensor.