3.1 Sample preparation
3.1.1 VLS method
The ZnO nanowires were grown by thermal vapor deposition from zinc vapor and a little oxygen gas on FTO (F:SnO2) conducting glass substrate, which has wire resistance of 10 Ω/cm2. The vertical distance between the metal source and the sample was about 3~5 mm. Then the alumina boat which carried the substrate was inserted into a quartz tube. This quartz tube was placed inside a furnace, with the center of the alumina boat positioned at the center of the furnace and the substrates placed downstream of growth metal powder (Fig. 3-1). The quartz tube was evacuated to a pressure below 2 x10-2 torr using mechanical pump. After the high-purity argon gas was infused into the system with a flow rate of 500 sccm and pressure controlled at 10 torr. The zinc vapor was transported with argon carrier gas with controllable flow rate ranging from 500 sccm to 700 sccm, and pressure ranging 10torr to 30 torr, to change the rods’ length.
Fig. 3-1 Thermal vapor transport system
3.1.2 SLS method
First, the arrays of ZnO nanowires were synthesized on seeded fluorine doped tin oxide (FTO) substrates (3 mm thickness, 10 Ω per square, Nippon Sheet Glass) similar to P. D. Yang et al. by immersing the seeded substrates in aqueous solutions containing 0.08 M zinc nitrate hydrate (98%, Riedel-deHaën), 0.08 M hexamethylenetetramine (99%, Showa), and 12 mM polyethylenimine (branched, low molecular weight, Aldrich) at 95 °C for 10 hours. Second, in order to fabricate the branched nanowires, the ZnO nanowires substrate obtained from the first step were re-coated with ZnO seed layer by dip-coating in a 0.005 M zinc acetate dihydrate (99%, Riedel-deHaën) in ethanol. Then the branched nanowires were grown by immersing the seeded ZnO nanowires in aqueous solutions containing 0.02 M zinc nitrate hydrate, 0.02 M hexamethylenetetramine and 3 mM polyethylenimine at 95 °C
for 5 hours. The branched ZnO nanowires were finally rinsed with deionized water and baked in air at 450 °C for 30 minutes to remove any residual organics. The growth procedure was illustrated in Fig. 3-2
Fig. 3-2 The schematic growth procedure from the original ZnO nanowires to the branched ZnO
nanowires.
3.2 Fabricate the DSSCs
Dye-sensitized solar cells were prepared by immersing the branched ZnO nanowires on substrates into a solution of 0.5 mM cis-bis(isothiocyanato)bis(2,2’-bipyridyl-4,4’-dicarboxylato)-ruthenium(II)
bis-tetrabutylammonium (N719, Solaronix) in acetonitrile/tert-butanol (1:1) for 15 mins, and the films were then rinsed with acetonitrile. The sensitized electrodes were sandwiched together with thermally platinized FTO counter electrodes separated by 25-μm-thick hot-melt spacers (Surlyn, Dupont). The internal space of the cell was filled with an electrolyte solution (0.1M LiI, 0.5M 1,2-dimethyl-3-propylimidazolium iodide, 0.03M I2, and 0.5M tert-butylpyridine in acetonitrile).
3.3 Scanning Electron Microscope (SEM) system
The morphology of ZnO-based nanostructures was observed by the Field Emission Gun Scanning Electron Microscopy (FEG-SEM) [JEOL JSM-6500F]. The accelerated voltage is 0.5-30kV and the magnification is 20-300k times, as shown in Fig. 3-3.
Fig. 3-3 SEM system
3.4 The equipments of conversion efficiency measurement and impedance
spectroscopy
Efficiencies for solar energy conversion and ac impedance spectroscopy were immediately evaluated under AM 1.5G simulated sunlight (Yamashita Denso, YSS-100A) and electrochemical analyzer (Autolab, PGSTAT3). The light power was calibrated with a set of neutral density filters by using a silicon photodiode (BS-520, Bunko Keiki).
Chapter 4 Results and Discussion
4.1 ZnO nanowire based dye-sensitized solar cell manufactured by VLS
4.1.1 Growth of ZnO nanowires with VLS
The images shown in Fig. 4-1 are the SEM cross-sectional views of the nanostructures. Figure 4-1(a) shows the ZnO film and Fig. 4-1(b) shows the ZnO rods grown at 550 ℃ under 10 torr tube pressure with 500 sccm of flow rate for 60 mins. Their lengths are about ~ 2 μm. Figure 4-1(c) shows the rods grown at 550
℃ under 100 torr tube pressure with 700 sccm of flow rate for 60 mins. Their lengths are about ~ 4 μm.
(a)
(b)
(c)
Fig. 4-1 The SEM images of the ZnO rods（a）ZnO film, （b）1.5-2 μm rods, and（c）3-4 μm rods
4.1.2 The energy conversion of DSSCs base on ZnO nanowires grown by VLS
The current-voltage characteristics for a solar cell constructed using nanowires with Pin = 100 mW/cm2 broadband illumination were shown in Fig. 4-2. Active electrode area was typically 0.2827 cm2. Figures 4-2(a) and (b) show the I-V curve with the incident light came from the top and the bottom of the solar cell, respectively.
Because the incident light has to pass through the translucent ZnO buffer layer, less of the absorbed energy of incidence by the dye on the nanowires and so the Jsc for which the light came from the bottom of the solar cell is less than that the light came from the top of the solar cell. The relations of the short circuit current, the open circuit voltage, the fill factor, and the overall efficiency to length of nanowires are shown in Table 1. Obviously, the solar cell with longer nanowires has the better conversion efficiency, and the worst conversion efficiency obtained for using the ZnO film. It can be interpreted as the larger surface area can adsorb the more dye for the longer ZnO nanowires. The external quantum efficiency (ECE), which is the percentage of electrons collected per incident photon (with no correction for reflection losses), will increase as increasing length [16].
Additionally, after having been grown the ZnO nanowires by thermal vapor deposition at 550oC for an hour, the resistance of the FTO conducting glass substrates became 125 Ω/cm2. We found there are current jump around applied voltages of
0.05, 0.3, and 0.4 V in Fig. 4-2 for the films of ~2 μm, and ~4 μm thick. The higher sheet resistance reduces short-circuit current which causes the lower conversion efficiency as compared with the reported results [17, 18] of the ZnO nanowire-based dye-sensitized solar cell.
Table 1 The short circuit current, the open circuit voltage, the fill factor, and the overall efficiency of the solar cell with different lengths of nanowires.
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Fig. 4-2 Current-voltage characteristic of a ZnO nanowire-based dye sensitizes solar cell with 100
mW/cm2 broadband visible illumination.（a）the incident light came from the top of the solar cell and
（b）the incident light came from the bottom of the solar cell.
4.2 ZnO nanowire based dye-sensitized solar cell manufactured by SLS
The arrays of ZnO nanowires were synthesized on seeded fluorine doped tin oxide (FTO) substrates (3 mm thickness, 10 Ω per square, Nippon Sheet Glass) by immersing the seeded substrates in aqueous solutions containing 0.04 M zinc nitrate hydrate (98%, Riedel-deHaën), 0.04 M hexamethylenetetramine (99%, Showa) and at 75 °C for 8 hours. The image shown in Fig. 4-3 is the SEM cross-sectional view of the nanowires. The efficiency of the DSSC made by this nanowires sample is about
0.153 at % and the I-V curve is shown in Fig. 4-4. Active electrode area was typically 0.2827 cm2. The short circuit current density, the open circuit voltage, the fill factor are 0.905mA/cm2, 0.67V and 0.252, respectively and are summarized in Table 4-2. As compared with the report of P. D. Yang, et al. [17], it still has a lot to improve in efficiency. The I-V curve indicates not only there is a large barrier between nanowires and FTO but also the nanowires with length of ~1 μm and diameter of 100 nm may not offer sufficient surface area for dye to stick on.
Fig. 4-3 The SEM images of the ZnO rods fabricated by SLS with 0.04 M zinc nitrate hydrate (98%,
Riedel-deHaën), 0.04 M hexamethylenetetramine (99%, Showa) and at 75 °C for 8 hours.
Fig. 4-4 The I-V curve of the ZnO rods fabricated by SLS with 0.04 M zinc nitrate hydrate (98%,
Riedel-deHaën), 0.04 M hexamethylenetetramine (99%, Showa) and at 75 °C for 8 hours.
η(%) Voc (V) Jsc (mA/cm2) FF
0.153 0.666 0.905 0.252
Table 4-2 The short circuit current, the open circuit voltage, the fill factor, and the overall efficiency of
the solar cell, in which the nanowires were manufactured by SLS with 0.4M nitrate hydrate, etc.
0.0 0.2 0.4 0.6 0.8
4.3 Ultra long nanowires to improve the efficiency of the DSSCs manufactured
by SLS
The process of the growth of the ZnO nanowires is also the same as that mentioned in Section 4.2, but some different steps were used, e.g., 0.6 M zinc nitrate hydrate (98%, Riedel-deHaën), 0.06 M hexamethylenetetramine (99%, Showa) and 12 mM polyethylenimine (branched, low molecular weight, Aldrich) and stirred at 95
°C for 2.5 hours three times. The image shown in Fig. 4-5 is the SEM cross-sectional view of the nanowires. It seems that the length of nanowies are much longer than those mentioned in the previous section. The function of the polyethylenimine and repeating grown for three times at 95 °C is to make high aspect ratio nanowires.
The relations of the short circuit current, the open circuit voltage, the fill factor, and overall efficiency to length of nanowires were also listed in Table 4.3. Active electrode area was typically 0.2827 cm2. JSC and the fill factor become 2.82 mA/cm2 and 0.44. So the efficiency is higher than that mentioned in Section 4.2. The longer nanowires offer more surface area for dye adsorption and so have the higher efficiency. In the next chapter we will use another process to improve the efficiency of DSSCs.
(a)
(b)
(c)
Fig. 4-5 The SEM images of the ZnO rods (a) without PEI, (b) with PEI, and (c) with PEI and grow 3
times
Fig. 4-6 The I-V curve of the ZnO rods fabricated by SLS with 0.06 M zinc nitrate hydrate (98%,
Riedel-deHaën), 0.06 M hexamethylenetetramine (99%, Showa) and 12 mM PEI and at 95 °C for 2.5
hours three times.
Table 4-3 The short circuit current, the open circuit voltage, the fill factor, and the overall efficiency of
the solar cell which the nanowires manufacture by SLS with 0.6M nitrate hydrate etc.
4.4 Tree-like ZnO nanowries
Using the process described in section 3.1.2, we obtained slightly off-aligned seeded ZnO nanowires with density of about 7x108 cm-2 grown on the FTO substrate, as shown in Fig. 4-7. It shows the seeded ZnO nanowires were formed of little crystalline nanopaticles with diameter 10-20 nm attached to the backbone bare nanowires having length and diameter in the range of 7-8 µm and 150–250 nm, respectively. After the second growth step, the radial secondary branches emanated from the seeds, as shown in Figs. 4-7(b) and (c).
Fig. 4-7 Before (a) and After (b) re-coating a seed layer of the original ZnO nanowires obtained from
aqueous solution. (c) Low- and (d) High-magnification images of the branched ZnO nanowires after
second growth. The scale bar has a length of 1 μm.
The secondary branches or branched nanowires have length and diameter ranging from 100-300 nm and 20-50 nm, respectively. Although the branched nanowires have non-uniform size due to the roughly dip-coating process, they still can offer more sufficient internal surface area for dye adsorption as compared with the bare nanowires.
The photocurrent–voltage (J-V) characteristics for solar cells constructed using ZnO nanostructures with 100 mW/cm2 broadband illumination from a xenon lamp were shown in Fig. 4-8. Active electrode area was typically 0.2827 cm2. Table 4.4 summarizes the measured and the calculated values obtained from each J-V curve.
The short-circuit current density (Jsc) and the overall light conversion efficiency of the branched ZnO nanowires are 2.37 mA/cm2 and 1.51 %, respectively, and almost twice higher than that of the bare ZnO nanowires. One factor for the increase in short-circuit current density is the enhanced photon absorption associated with the presence of enlargement of internal surface area resulting in sufficient dye-loading.
Although the length and density of ZnO structures exhibit the disadvantage compared with previous study [19], the overall efficiency can achieve almost the same value via utilizing the extra secondary nanobranches. The values of fill factor (FF) for ZnO DSSCs are general lower than those using TiO2 nanoparticles (0.6–0.7). This is attributed to recombination between photoexcited carriers in the photoanodes and
tri-iodide ions in the electrolyte. Slightly different value in the bare ZnO nanowire and branched ZnO nanowire DSSCs reveal the almost same interfacial recombination, which is evidenced by the equivalent value of the shunt resistance Rsh=(dV/dI)V=0
from the I-V curves under illumination. The series resistance Rs=(dV/ dI)I=0 for branched ZnO nanowire DSSC (25.64 Ωcm2) is significantly lower than the bare ZnO one (46.13 Ωcm2) and is previously explained association with the dye loading but not the nanowire length [20].
Fig. 4-8 Current density against voltage (J-V) characteristics
Recently, electrochemical impedance spectroscopy (EIS) measurement has been widely performed to investigate electronic and ionic processes in dye-sensitized solar
cells [21, 22, 23, 24, 25]. The Nyquist plots of the impedance data for bare and branched ZnO nanowire DSSCs were performed by applying a 10 mV ac signal over the frequency range of 10−2–105 Hz under illumination at the applied bias of Voc, as shown in Fig. 4-9(b). Some interior parameters of the devices with thickness of LF
can be derived by well fitting the impedance data of the Nyquist plots to expressions based on the equivalent circuit of nanowire DSSC suggested by Wu et al [23, 24].
Although the concrete equivalent circuit may be more complex than the previous reports. According to the formation of Zn2+/dye complexes from the dissolution of surface Zn atoms in the acid dye system and finally block the electron transport from the dye to semiconductor [26]. Hence, the impedance data obtained at different applied potentials cannot be fitted with a single equivalent circuit and some components such as an inductor element should be considered [27]. In order to avoid the unnecessary interference from the trapped electrons, we presently ignored the low frequency part of impedance data. The fitted results were also listed in Table 4.4, which includes the first-order reaction rate constant for the loss of electrons (keff), the electron lifetime (τ=1/keff), the electron transport resistance (Rw), and the charge transfer resistance related to recombination of electron at the ZnO/electrolyte interface (Rk). We found Rk and Rw are quite similar for both DSSCs in this present work, which indicates the same interfacial recombination and equal crystallinity for either
bare ZnO nanowires or branched ZnO nanostructures. On the contrary, keff in the branched ZnO nanowire DSSC is smaller than the bare nanowire one to cause the smaller effective diffusion length [22] (Deff = (Rk/Rw)LF2keff ) in branched ZnO nanowire DSSC. Since keff is related to reaction rate, the electron lifetime, τ = 1/keff, could be prolonged by the additional transport distance between branches and conductive electrode (backbone nanowire).
Fig. 4-9 Nyquist plots of the bare ZnO nanowires and branched ZnO nanowires DSSCs under AM1.5
illumination. The solid lines are the fitting results.
In general, the current density for a DSSC is determined by the initial number of photogenerated carriers, the electron injection efficiency from dye molecules to semiconductor, and the recombination rate between the injected electrons and oxidized dye or redox species in the electrolyte. Base on the assumption of the same injection efficiency and recombination rate for the given ZnO DSSC systems, it is reasonable that the initial number of photogenerated carriers may be significantly affected by the variation in the light-harvesting capability of different-structured photoanodes. Figure 4-10 displays the wavelength distribution of incident monochromatic photon to current conversion efficiency (IPCE). The photocurrent peaks occurring at approximately 400 nm are due to direct light harvesting by ZnO semiconductor (see Fig. 4-11), in which the photogenerated electrons diffuse through ZnO and the holes in the valence band are replenished directly by charge transfer from the I3-/I-electrolyte [20]. The maxima of IPCE in the visible region contributed by the dye absorption are located at approximately 525 nm (see Fig. 4-12), corresponding to the visible t2Æπ* metal-to-ligand charge transfer (MLCT). The IPCE obtained for the branched ZnO nanowire DSSC is almost 1.5 times that of the bare ZnO nanowire. The improvement in the IPCE suggests that the high energy conversion efficiency results predominantly from sufficient dye-loading by branched ZnO nanowires, which enlarge internal surface area within the photoelectrode. This
concept is anticipated to be equally applicable to other semiconductor photoelectrodes in DSSCs and organic–inorganic hybrid solar cells, despite that the filling factor of the ZnO nanostructures are not optimized at this point, as previously shown in the SEM images (see Fig. 4-7). Further improvement of energy conversion efficiency could be implemented through adjusting denser and longer branches to fill the interstitial voids between backbone nanowires, which will substantially improve the light harvesting and the current density.
400 450 500 550 600 650 700
0
Fig. 4-10 The incident monochromatic photon to current conversion efficiency (IPCE) of the bare
ZnO nanowires and the branched ZnO nanowires DSSCs.
Fig. 4-11 The photoluminescence spectra of ZnO under 325 nm He-Cd Laser excitation.
Fig. 4-12 The absorbance spectra of N719 [28]