Efficient Dye-Sensitized Solar Cells Based on Nanoflower-like ... - MDPI

molecules Communication

Efficient Dye-Sensitized Solar Cells Based on Nanoflower-like ZnO Photoelectrode Xiaobo Chen 1, *, Yu Tang 2 and Weiwei Liu 1 1 2

*

School of New Energy and Electronic Engineering, Yancheng Teachers University, Yancheng 224051, China; wwliu_ciomp@126.com School of Intelligent Manufacturing, Sichuan University of Arts and Science, Dazhou 635000, China; love-ty11@163.com Correspondence: chenxbok@126.com; Tel.: +86-515-8823-3177

Received: 5 July 2017; Accepted: 1 August 2017; Published: 3 August 2017

Abstract: A photoanode material ZnO nanoflower (ZNFs) for efficient dye-sensitized solar cell (DSSC) was prepared. This unique structure can significantly increase the specific surface area and amount of light absorption, leading to a higher short-circuit current density. Furthermore, ZNFs resulted in closer spacing between the nanorods and more direct conduction paths for electrons, leading to higher open-circuit voltage. The overall promising power conversion efficiency of 5.96% was obtained with photoanodes of 8.5 µm thickness. This work shows that ZNFs is an attractive material and has good potential for application in high efficiency ZnO-based DSSCs. Keywords: energy storage and conversion; solar energy materials; microstructure

1. Introduction Dye-sensitized solar cells (DSSC) are promising alternatives to conventional solar cells because of their cheap, environmentally friendly, and easy fabrication [1,2]. A porous-structured wide band gap metal oxide film (such as TiO2 [3], ZnO [4], and SnO2 [5]) as a photoanode is a key component of a typical DSSC, which determines the light-harvesting capability, charge diffusion, and collection efficiency [6]. ZnO has been considered a fascinating alternative photoanode material in DSSCs, since it has a similar electronic band-gap level and higher electron mobility with respect to conventional photoanode material-TiO2 [7]. The structure, morphology, crystallinity, and size of ZnO significantly affect the final power conversion efficiency [7,8]. Therefore, various nanostructured ZnO films (such as nanoparticles, nanotubes, nanowires, nanosheets, nanoflakes, hollow spheres, microspheres, and flowers) have been synthesized to enhance PCE in ZnO-based DSCs, which can offer larger surface areas, effective light-scattering centers, or direct electron pathways [7–9]. Here, we report on the deposition of ZnO nanoflower (ZNF) photoanodes, which we have developed in an attempt to increase the surface area and ability to reflect and scatter light. An overall power conversion efficiency of 5.96% is obtained by using this highly connected ZNF photoanode with the dye N719, yielding 35.8% enhancement in comparison to DSSCs with ZnO nanoparticle (ZNP) photoanodes. 2. Results and Discussion Field-emission scanning electron microscopy has been used to investigate the microstructure of the ZnO photoanode, as shown in Figure 1a. The well-defined flower-like morphology could be easily seen. Such a compact flower-like structure of ZnO could benefit the adsorption of dye as well as could provide a direct pathway for the electron transport, which is crucial for light conversion. SEM-coupled energy-dispersive X-ray spectroscopy (EDS) is performed for determining the composition of the

Molecules 2017, 22, 1284; doi:10.3390/molecules22081284

www.mdpi.com/journal/molecules

Molecules 2017, 22, 1284

2 of 6

Molecules 2017, 22, 2 of 6 composition of 1284 the grown thin film. Figure 1b shows the EDS of the grown ZNF thin film, in which 2017, 1284 with weak Sn peaks are observed. The presence of Zn and O atoms indicates 2 of 6 strongMolecules Zn and O22, peaks the presence of ZnO, and Sn peaks originate from ZNF the FTO substrate. Figure 1c shows the XRD grown thin film. Figure 1b the shows of the grown thin film, in which Zninand O peaks composition of the grown thin the film.EDS Figure 1b shows the EDS of the grown ZNF strong thin film, which pattern (XRD, Rigaku) of ZNFs, exhibiting the peaks corresponds to hexagonal wurtzite structure of with strong weak Zn Sn and peaks are observed. presence of Zn and atoms indicates the presence of ZnO, O peaks with weakThe Sn peaks are observed. TheOpresence of Zn and O atoms indicates ZnO (JCPDS 36-1451). diffraction peaks are associated with FTO(XRD, substrate. In the presence of ZnO,Other andfrom theobserved Sn from the FTO substrate. 1c shows the XRD and the Sn peaks originate thepeaks FTOoriginate substrate. Figure 1c shows the Figure XRD pattern Rigaku) Figure 1d, it is clear to see that the ZNPs are almost spherical and uniformly distributed with a size pattern (XRD, Rigaku) of ZNFs, exhibiting the peaks corresponds to hexagonal wurtzite structure of of ZNFs, exhibiting the peaks corresponds to hexagonal wurtzite structure of ZnO (JCPDS 36-1451). around 50(JCPDS nm. diffraction The components and crystallinity of the ZNPs respectively 36-1451). Other are observed diffraction peaks arecommercial associated with FTO In that OtherZnO observed peaks associated with FTO substrate. In Figure 1d, itwere issubstrate. clear to see Figure by 1d, EDS it is clear to see that the ZNPs arepattern almost spherical and1e, uniformly distributed with a sizewere confirmed in Figure 1e and XRD in Figure indicating that the ZNPs the ZNPs are almost spherical and uniformly distributed with a size around 50 nm. The components around 50 hexagonal nm. The components and crystallinity of the commercial ZNPs were respectively relatively pure wurtzite structure ZnO. and crystallinity of the commercial ZNPs were respectively confirmed by EDS in Figure 1e and XRD

confirmed by EDS in Figure 1e and XRD pattern in Figure 1e, indicating that the ZNPs were

pattern in Figure indicating that the ZNPs ZnO. were relatively pure hexagonal wurtzite structure ZnO. relatively pure1e, hexagonal wurtzite structure

ff (101)

(112)

(110)

(112) (103)

(110)

(103)

(102)

(100)

(100)

(102)

(103)

(200)

(200)

(112)

(112)

(103)

(110)

(102)

(102)

(110)

(100) (002)

Intensity(a.u.) (a.u.) Intensity

(101) (100) (002) (101)

FTO

(002) (002) (101)

FTO

Intensity (a.u.)

Intensity (a.u.)

cc

20

20

30

30

40

40

50

50

60

60

2θ (degree)

70

70

20 20

30

40

30

40

50

50

60

60 70 70

2θ (degree)

2θ (degree)

2θ (degree)

Figure 1. SEM photographs of ZNFpowder powder and and (d) EDS spectra of (b) ZNFs on FTO Figure SEM photographs of(a) (a)(a) ZNF (d)ZNP ZNPpowder; powder; EDS spectra of(b) (b)ZNFs ZNFs on FTO FTO Figure 1.1.SEM photographs of ZNF powder and (d) ZNP powder; EDS spectra of on glass and (e) ZNP powder; XRD patterns of (c) ZNFs on FTO glass and (f) ZNP powder. glass and (e) ZNP powder; XRD patterns of (c) ZNFs on FTO glass and (f) ZNP powder. glass and (e) ZNP powder; XRD patterns of (c) ZNFs on FTO glass and (f) ZNP powder.

The porosity of the ZNF is verified by N2-sorption isotherm measurement (ASAP 2020,

0.2 0.1

100 0.3 150 0.2

0.0

100 50

-1 -1

0

b

3

0

20 40 60 80 100 120 Pore W idth (nm)

50 0.1 0.0 0 0.0

3

-1

0.3

Volume Adsorbed Volume Adsorbed (cm /g) (cm /g)

-1

a

150

0.4

0.4

200

Dv (cm3 g nm )

3

Volume Adsorbed (cm /g)

Dv (cm3 g nm )

200

3

Volume Adsorbed (cm /g)

The porosity of the ZNF is verified verified by N22-sorption -sorption isotherm isotherm The porosity as of shown is N measurement (ASAP 2020, Micromeritics), in Figure 2a. The by Brunauer-Emmet-Teller (BET)measurement specific surface(ASAP area of2020, 2 −1 Micromeritics), asasshown ininFigure The specific Micromeritics), shown Figureis2a. 2a. TheBrunauer-Emmet-Teller Brunauer-Emmet-Teller (BET) specific surface surface area of as-synthesized ZNF photoanode calculated to be 74 m g . The pore(BET) size distribution of ZNFsarea is of − −11 shown in the inset of Figure 2a.is exhibit to a broad distribution, mostly in the range of as-synthesized ZNF photoanode calculated m22 gsize . .The distribution of is as-synthesized ZNF photoanode isZNFs be 74 pore m Thepore poresize size distribution of ZNFs ZNFs is 20–100 nm, ensuring that dye be absorbed throughout ZNFs. The BET surface of a of shown in inset of 2a. ZNFs exhibit aa broad pore distribution, mostly in the shown in the the inset of Figure Figure 2a.could ZNFsget exhibit broad pore size sizethe distribution, mostly inarea the range range of −1 (Figure 2b). The thickness of these two photoanode films with similar ZNP photoanode is 27dye m2 gcould 20–100 nm, ensuring get 20–100 nm, ensuring that that dye could get be be absorbed absorbed throughout throughout the the ZNFs. ZNFs. The The BET BET surface surface area area of of aa thickness is was to be approximately 8.5 μm, of using an optical profilometerfilms (NanoMap-D, −1(Figure ZNP photoanode 2b). two with ZNP film photoanode is 27measured m22 g−1 (Figure 2b).The Thethickness thickness ofthese these two photoanode photoanode films with similar similar AEP Technology) in the optical mode. film film thickness thickness was was measured measured to to be be approximately approximately8.5 8.5μm, µm,using usingan anoptical opticalprofilometer profilometer(NanoMap-D, (NanoMap-D, AEP AEP Technology) Technology) in the optical mode. 80 a

20 40 60 80 100 120

0.2Pore W0.4 idth (nm)0.6

Adsorption Desorption

0.8

Relative Pressure (P /P 0)


1.0

60

80

b Adsorption Desorption

40

60

20


40 0 0.0

20

0.2

0.4

0.6

0.8

1.0

Relative Pressure (P/P0)

0 0 The N2 adsorption/desorption isotherms Figure 2. Nitrogen sorption isotherms of the photoanodes. (a) 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 and the pore size distribution (inset) of the ZNF photoanode; (b) Nitrogen sorption Relative Pressure (P /P 0) Relative Pressure (P/P0)isotherms of the ZNP photoanode.

Figure photoanodes. (a) (a) The The N N2 adsorption/desorption adsorption/desorption isotherms Figure2. 2. Nitrogen Nitrogen sorption sorption isotherms isotherms of of the the photoanodes. isotherms 2 and the pore size distribution (inset) of the ZNF photoanode; (b) Nitrogen sorption isotherms and the pore size distribution (inset) of the ZNF photoanode; (b) Nitrogen sorption isotherms of of the the ZNP photoanode. ZNP photoanode.

Molecules 2017, 22, 1284

3 of 6

Molecules 2017, 22, 1284

3 of 6

The photovoltaic performance of ZNFs was compared against ZNPs. Figure 3 reports The density-photovoltage photovoltaic performance ZNFs was compared against ZNPs. Figure reports photocurrent (J-V)ofmeasurements under AM 1.5 illumination. The3photovoltaic photocurrent density-photovoltage (J-V) measurements under AM 1.5 illumination. The photovoltaic parameters are summarized in Table 1. It can be seen that the two devices have a similar V oc ; because arethe summarized in Table 1. It(including can be seensame that the have a similar Voc; because theseparameters DSSCs have same compositions Pt two CE devices and electrolyte), it makes sense that these DSSCs have the same compositions (including same Pt CE and electrolyte), it makes sense that Molecules 2017, 22, 1284 3 of 6 their V oc values are close. A comparison of the J-V characteristics of the two photoanode materials their Voc values are close. A comparison of the J-V characteristics of the two photoanode materials indicated that Jsc was the parameter that contributed to the enhancement of the efficiency for the Thethat photovoltaic of ZNFs was compared against ZNPs. Figure 3 reports indicated Jsc was theperformance parameter that contributed to the enhancement of the efficiency for the flower-like ZnO.ZnO. The Jsc of ZNF-based-DSSC is about 33.7% larger than the JscThe of ZNP-based-DSSC. photocurrent density-photovoltage (J-V) measurements under larger AM 1.5than illumination. photovoltaic flower-like The Jsc of ZNF-based-DSSC is about 33.7% the Jsc of ZNP-based-DSSC. 2 −1 Such Such a relatively high photocurrent is ispossibly higher specific (34 parameters are summarized in Table 1. possibly It can becaused seen thatby theaatwo devices havesurface asurface similararea Varea oc; (34 because a relatively high photocurrent caused by higher specific m2 m g−1) g ) these DSSCs thedye same compositions (including same Pt CE and electrolyte), it makes sense that and therefore a higher dye loading ononthe and therefore ahave higher loading theZNF ZNFphotoelectrode. photoelectrode. -2

-2 density (mA cm ) Current density (mA cm Current )

their Voc values are close. A comparison of the J-V characteristics of the two photoanode materials indicated that Jsc was the parameter 14 that contributed to theZNFs enhancement of the efficiency for the 12 flower-like ZnO. The Jsc of ZNF-based-DSSC is about 33.7% ZNPs larger than the Jsc of ZNP-based-DSSC. Such a relatively high photocurrent is10 possibly caused by a higher specific surface area (34 m2 g−1) 8 and therefore a higher dye loading on the ZNF photoelectrode. 6 4 14 2

ZNFs ZNPs

12

0 0.0 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Voltage (V)

8 6

Figure 3. The J-V curves of the DSSCs based on ZNF and ZNP photoelectrodes. 4 DSSCs based on ZNF and ZNP photoelectrodes. Figure 3. The J-V curves of the 2

0 Table 1. Summary of photovoltaic and electrochemical properties of the cells. 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Table 1. Summary of photovoltaic and electrochemical properties of the cells. Voltage (V) Photoanode

Voc (mV)

Jsc (mA cm−2)

FF (%)

PCE (%)

Figure 3. The J-V curves680 of the based on and±ZNP photoelectrodes. ZNP ± 1DSSCs10.1 ± 0.2 1 4.39 ± 0.12 −2ZNF64.0

Photoanode

V oc (mV)

ZNF

682 ± 2

Jsc (mA cm

13.5 ± 0.1

)

FF (%)

64.6 ± 1

PCE (%)

5.95 ± 0.13

of photovoltaic and ± electrochemical properties ZNPTable 1. Summary 680 ± 1 10.1 0.2 64.0 ± 1 of the cells. 4.39 ± 0.12 ZNF 682 ± 2 13.5 ± 0.1 64.6 ± 1 5.95 ± 0.13 −2) determined The effective dye loading of theVocphotoelectrode was from (mV) Jsc (mA cm FF (%) PCE (%) the absorption value for Photoanode ZNP 680 ± 1 law 10.1 ± 0.2and the 64.0 UV/Vis ± 1 4.39absorption ± 0.12 each NaOH-dye solution according to Beer’s [10], spectra are shown ZNFon ZNFs 682(1.5 ± 2 × 1013.5 ± 0.1 ± 1 5.95 ± 0.13 −7 mol −2) 64.6 −2), in Figure 4a. The dye loading cm is higher than ZNPs (1.2 × 10−7 mol cm The effective dye loading of the photoelectrode was determined from the absorption value for which will lead to more photogenerated charge carriers and higher current density for the former. The each NaOH-dye solution according to Beer’s law [10], and the UV/Vis absorption spectra are shown in The effective dye loading of the photoelectrode was determined from the absorption value for second factor that contributed to the Jsc was−7the enhanced light scattering by the microstructures. A −2 − 2 − 7 Figure 4a. The dye loading on according ZNFs (1.5to×Beer’s 10 law mol cmand) the is higher ZNPsspectra (1.2 ×are 10 shown mol cm ), each NaOH-dye solution [10], UV/Visthan absorption comparison of the diffuse reflectance spectra from−7 ZNF and ZNP photoelectrodes is given−7 in Figure 4b. −2), in Figure The dyephotogenerated loading on ZNFs (1.5 × 10 carriers mol cm−2)and is higher thancurrent ZNPs (1.2 × 10 mol cm whichCompared will leadto4a. to more charge higher density for the former. the ZNP photoelectrode, the ZNF photoelectrode has higher diffuse reflection capability, which will lead to more photogenerated charge carriers and higher current density for the former. The The second factor that contributed the Jsc was[11]. the enhanced light scattering by the microstructures. thus definitely allowing a higherto photocurrent

second factor that contributed to the Jsc was the enhanced light scattering by the microstructures. A

A comparison of the diffuse reflectance spectra from ZNF and ZNP photoelectrodes is given in comparison of the diffuse reflectance spectra from ZNF80 and ZNP photoelectrodes is given in Figure 4b. Figure 4b. Compared thephotoelectrode, ZNP photoelectrode, the ZNF photoelectrode hasreflection higher diffuse reflection ato Compared to the ZNP the ZNF photoelectrode has higher diffuse capability, ZNFs 70 b ZNPs ZNPs capability, definitely allowing higher [11]. ZNFs photocurrent thusthus definitely allowing a higheraphotocurrent [11].

400

ZNPs ZNFs

450

500

550

600

Wavelength (nm)

650

700

Reflectance (%)Reflectance (%)

a Absorption (a.u.)

Absorption (a.u.)

60 50

80

40

70

b

ZNFs ZNPs

30

60

20

50

10 40 300

300

20

400

500

600

700

800

Wavelength (nm)

10 0 Figure 4. UV/Vis absorption and diffuse reflectance spectra of ZNFs and ZNPs photoelectrodes. (a) 400 450 500 550 600 650 700 300 400 500 600 700 800 The UV/Vis absorption Wavelength spectra (nm) of solutions containing dyes detached from the films in NaOH; Wavelength (nm) (b) Diffuse reflectance spectra of ZNF and ZNP photoelectrodes before dye adsorption. EIS spectra of Figure 4. UV/Vis absorption and diffuse reflectance spectra of ZNFs and ZNPs photoelectrodes. (a) Figure UV/Vis the4.two DSSCs. absorption and diffuse reflectance spectra of ZNFs and ZNPs photoelectrodes.

The UV/Vis absorption spectra of solutions containing dyes detached from the films in NaOH;

(a) The UV/Vis absorption spectra of solutions containing dyes detached from the films in NaOH; (b) Diffuse reflectance spectra of ZNF and ZNP photoelectrodes before dye adsorption. EIS spectra of To understand transport of photoexcited electrochemical (b) Diffuse reflectancethe spectra of ZNFand andrecombination ZNP photoelectrodes before dyeelectrons, adsorption. EIS spectra of the two DSSCs. impedance spectroscopy (EIS) was performed in both the ZNFand ZNP-based DSSCs by an the two DSSCs. Electrochemical Workstation. Nyquist and Bode phase plot fromelectrons, EIS studies are shown in To understand the transport andplot recombination of photoexcited electrochemical Figure 5a,b, and the corresponding parameters summarized Tableby 2. From impedance spectroscopy (EIS) was electrochemical performed in both the ZNF-are and ZNP-basedinDSSCs an

To understand the transport and recombination of photoexcited electrons, electrochemical Electrochemical Workstation. Nyquist plot and Bode phase plot from EIS studies are shown in impedance spectroscopy (EIS) was performed in both the ZNF- and ZNP-based DSSCs by an Figure 5a,b, and the corresponding electrochemical parameters are summarized in Table 2. From Electrochemical Workstation. Nyquist plot and Bode phase plot from EIS studies are shown in

Molecules 2017, 22, 1284

4 of 6

Figure 5a,b, and the corresponding electrochemical parameters are summarized in Table 2. From the Nyquist EIS2017, plots Molecules 22, of 1284DSSCs (Figure 5a), the starting point correlates with the serial resistance 4 of 6 (Rs ) of the DSSCs. While three semicircles are exhibited in the frequency ranges of 105~103, 103~1, and 2 Hz, Nyquist EIS plots ofto DSSCs (Figure 5a), the starting point the serial resistanceinterface (Rs) 1~10−the corresponding charger transfer resistance at thecorrelates counter with electrode/electrolyte the DSSCs. While three semicircles exhibited in the of 105~103, 103~1,ofand (Rct1 ),ofphotoanode/electrolyte interfaceare (Rct2 ) [12,13], andfrequency Warburgranges diffusion resistance I− /I3 − −2 Hz, corresponding to charger transfer resistance at the counter electrode/electrolyte interface 1~10 (W).The Rs values have no obvious change, indicating the difference of the photoanode may exert (Rct1), photoanode/electrolyte interface (Rct2) [12,13], and Warburg diffusion resistance of I−/I3− almost no effect on the contact resistance. It is found that there is no obvious deviation for Rct1 , (W).The Rs values have no obvious change, indicating the difference of the photoanode may exert because of the utilization of same Pt CE and I− /I3 − redox electrolyte. According to the method of almost no effect on the contact resistance. It is found that there is no obvious deviation for Rct1, Adachi et al. of [14], charge-transfer ZnO/dye/electrolyte (Rct2 ) of can be because thethe utilization of same Ptresistance CE and I−at /I3−the redox electrolyte. Accordinginterface to the method obtained from the Nyquist plot. It was found that the R for ZNFs was smaller than that for ZNPs, Adachi et al. [14], the charge-transfer resistance at thect2 ZnO/dye/electrolyte interface (Rct2) can be whichobtained indicates that are Iteasier to move at the surface and contribute to the charge from theelectrons Nyquist plot. was found that the Rct2 ZNF for ZNFs was smaller than that for ZNPs, whichatindicates that electrons move at the ZNFasurface and contribute charge transport the photoanode [15]. are Theeasier W fortothe ZNFs shows value lower than thattoofthe ZNPs, which transport at can the photoanode [15]. The W ZNFs shows value lower that ofInZNPs, which reveals that I3 − be rapidly reduced to for I− the to speed up thea diffusion ofthan I3 − [16]. the Bode phase reveals that I 3− can be rapidly reduced to I− to speed up the diffusion of I3− [16]. In the Bode phase plot (Figure 5b), the electron lifetime (τ) can be estimated using the equation τ = 1/(2πf p ), where f p (Figure 5b), thecorresponding electron lifetimeto(τ)the cancharger be estimated using the equation = 1/(2πfp),where fp is is theplot peak frequency transfer process at the τphotoanode/electrolyte the peak frequency corresponding to the charger transfer process at the photoanode/electrolyte interface. The electrons live longer in the DSSCs based on ZNFs (5.34 ms) compared to that in the interface. The electrons live longer in the DSSCs based on ZNFs (5.34 ms) compared to that in the DSSCs based on ZNPs (4.84 ms). The enhanced electron lifetime can be attributed to the reduced DSSCs based on ZNPs (4.84 ms). The enhanced electron lifetime can be attributed to the reduced recombination process, resulting transfer,increasing increasing electron density, recombination process, resultingininaccelerating accelerating electron electron transfer, electron density, and and improving device performance [17]. attributelarger larger surface area, greater ability improving device performance [17].InInshort, short, we we can can attribute surface area, greater ability to to reflectreflect and scatter light, andand better propertiespossessed possessed ZNFs to their higher and scatter light, bettercharge chargetransport transport properties by by ZNFs to their higher conversion efficiency over ZNPs. conversion efficiency over ZNPs. 10

50

a

CPE1

ZNPs ZNFs

CPE2

8

b

40

ZNPs ZNFs

W

Rct2

-Phase (Degree)

Rct1

2

-Z'' (ohm cm )

Rs

6 4

30 20 10

2 0

0

0

2

4

6

8

10

12

14

16

2

-2

10

-1

10

0

1

10

10

2

10

3

10

4

10

5

10

Frequency (Hz)

Z' (ohm cm )

Figure 5. Nyquist Bode phaseplot plotof ofZNF ZNF and and ZNP (a) (a) Nyquist plot plot (Inset(Inset Figure 5. Nyquist plotplot andand Bode phase ZNPphotoelectrodes. photoelectrodes. Nyquist gives the equivalent circuit used to fit the impedance data) and (b) Bode phase plot. gives the equivalent circuit used to fit the impedance data) and (b) Bode phase plot. Table 2. The electrochemical parameters extracted from the DSSCs with various CEs.

Table 2. The electrochemical parameters extracted from the DSSCs with various CEs. Photoanode

Photoanode ZNP Rs (Ω cm2 ) ZNF

ZNP 10.02 ZNF 9.89 3. Materials and Methods

Rs (Ω cm2) Rct1 (Ω cm2) Rct2 (Ω cm2) W (Ω) 2) 10.02 W 1.71 (Ω) Rct1 (Ω cm2 ) 3.01 Rct2 (Ω cm12.75 9.66 1.35 9.89 2.94

3.01 2.94

12.75 9.66

1.71 1.35

fp (ms) Τ (ms) 32.90 f (ms) 4.84 p 29.82 5.34

32.90 29.82

T (ms) 4.84 5.34

ZNFs were synthesized by modifying the procedure as reported by Gupta et al. [18]. Briefly, ZnCl2 and NaOH were mixed with a molar ratio of 1:7 and dissolved in pure water. Subsequently, the mixture was placed by in amodifying glass reaction vessel andasautoclaved byGupta a single mode ZNFs were synthesized the procedure reported by et al. [18].microwave Briefly, ZnCl2 reactorwere (2.45 mixed GHz, Discover SP, CEM) °Cand for dissolved 0.5 h. Finally, the white product was obtained by and NaOH with a molar ratioatof801:7 in pure water. Subsequently, the mixture centrifugation at 7000 rpm for 5 min. The white powder was washed with distilled water and was placed in a glass reaction vessel and autoclaved by a single mode microwave reactor (2.45 GHz, absolute alternately several times to remove the impurities, dried in air atby 50 °C for 3 h. Discover SP, ethanol CEM) at 80 ◦ C for 0.5 h. Finally, the white productand was obtained centrifugation Ethyl cellulose, terpineol, and ZNF powder were added into an ethanol solution and stirred to at 7000 rpm for 5 min. The white powder was washed with distilled water and absolute ethanol get a paste. Subsequently, the paste was deposited on fluorine-doped tin oxide (FTO) glass substrate alternately several times to remove the impurities, and dried in air at 50 ◦ C for 3 h. by doctor-blade technique. The active area of photoanodes was controlled at 0.16 cm2 (0.4 × 0.4 cm). Ethyl terpineol, ZNF powder were added into°C anfor ethanol solution Beforecellulose, dye adsorption, theand ZnO films were annealed at 500 1 h to removeand thestirred organicto get a paste. Subsequently, the paste was deposited on fluorine-doped tin oxide (FTO) glass substrate contamination in the paste and improve the interconnecting network. The thermal treatment also by

3. Materials and Methods

doctor-blade technique. The active area of photoanodes was controlled at 0.16 cm2 (0.4 × 0.4 cm). Before

Molecules 2017, 22, 1284

5 of 6

dye adsorption, the ZnO films were annealed at 500 ◦ C for 1 h to remove the organic contamination in the paste and improve the interconnecting network. The thermal treatment also increased the crystallinity of the ZnO powder. A conventional ZnO photoanode made from commercial ZNPs (