- Email: [email protected]
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
Nano Energy (2015) 12, 501–509
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nanoenergy
RAPID COMMUNICATION
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes Jia Lianga,b, Gengmin Zhanga,c,n, Wentao Suna, Pei Dongb a
Key Laboratory for the Physics and Chemistry of Nanodevices and Department of Electronics, Peking University, Beijing 100871, China b Department of Materials Science and NanoEngineering, Rice University, 6100 Main Street, Houston, TX 77005, United States c SIP-UCLA Institute for Technology Advancement, Suzhou 215123, Jiangsu Province, China Received 3 September 2014; received in revised form 9 January 2015; accepted 10 January 2015 Available online 21 January 2015
KEYWORDS
Abstract
Multi-working electrodes; Flexible; TiO2 nanotube arrays; Dye-sensitized solar cells
Novel flexible fiber-type dye-sensitized solar cells (FF-DSSCs) with multi-working electrodes (MWFFDSSCs) have been developed. In each MWFF-DSSC, all the components are assembled into a flexible plastic capillary tube. A Pt microwire along the axis of the tube is used as the sole counter electrode and a number of Ti microwires surrounding it, which are all covered with highly ordered titanium dioxide (TiO2) nanotube arrays, are jointly used as the working electrodes. This new configuration brings about good flexibility, capability of harvesting light from all directions and a conversion efficiency competitive with those of the conventional DSSCs. The photovoltaic performances of the MWFF-DSSC with six working electrodes (MWFF-DSSC(6)) are better than those of the MWFF-DSSCs with two to five working electrodes (MWFF-DSSC(x), x=2, 3, 4, or 5) and the FF-DSSC with a single working electrode (SWFF-DSSC). When the as-prepared TiO2 nanotube arrays are used as the working electrodes, a 6.6% conversion efficiency is obtained from an MWFF-DSSC(6). When the TiO2 nanotube arrays are treated in the niobium isopropoxide solution, the conversion efficiency is further raised to 9.1% and only suffers a 6.6% relative decrease even under a 1801 bending. & 2015 Elsevier Ltd. All rights reserved.
n Corresponding author at: Key Laboratory for the Physics and Chemistry of Nanodevices and Department of Electronics, Peking University, Beijing 100871, China Tel.: +86 10 62751773; fax: +86 10 62762999. E-mail address: [email protected] (G. Zhang).
http://dx.doi.org/10.1016/j.nanoen.2015.01.023 2211-2855/& 2015 Elsevier Ltd. All rights reserved.
Since its invention in 1991, the dye-sensitized solar cell (DSSC) has attracted significant attention as a promising photovoltaic device due to its low cost and convenience in fabrication [1]. So far, tremendous efforts have been devoted to the fundamental and experimental researches on the DSSCs for improving their power conversion efficiencies and compatibilities to practical applications [2]. A conventional
502 DSSC is made up of a photoanode (also termed as a working electrode), a counter electrode and an electrolyte filled in the space between them. Generally, the photoanode is a fluorine-doped tin oxide (FTO) glass substrate covered with a dye-sensitized titanium dioxide (TiO2) film and the counter electrode is a platinized conductive glass [3]. The eventual performance of a DSSC is a synergistic effect of all these components [4–11]. Recently, researchers have been increasingly interested in the flexible DSSCs, which feature lightweight, low cost rollto-roll process and extensive application perspectives. Such substrates as conducting plastics, metal sheets, and metal wires are used in the flexible DSSCs [6,12–19]. Among them, the metal wires are considered to be more promising than the relatively poorly conducting flexible plastics and opaque metal sheets. Thus the flexible fiber-type dye-sensitized solar cells (FF-DSSCs) based on metal wires are very attractive [20– 28]. Although the FF-DSSCs have already exhibited good performances, there is still room for further improvement. First, the use of a platinum (Pt) microwire as the counter electrode sacrifices the advantage of low cost of an FF-DSSC. Second, in the calculation of the conversion efficiency of an FF-DSSC, the product of the diameter and effective length of the working electrode is usually used to represent the cross section of the incident light [6,17–28]. That is, only the light incident to the flank side of the working electrode is taken into consideration and the light incident to other part of an FF-DSSC is ignored. Because the light incident to the part other than the working electrode, which is not utilized effectively, is not included in the computation, the denominator in computing the conversion efficiency is unreasonably underestimated. In contrast, in a conventional DSSC, the photoanode receives all the light incident to the cell. Therefore, for a more reasonable comparison with the conventional DSSCs, the product of the diameter and effective length of the capillary tube, rather than the working electrode, should be used to represent the cross section of the incident light [29]. Accordingly, if the Pt counter electrode can be utilized more efficiently, i.e., larger photocurrent can be extracted from a single Pt microwire, and all the light incident to an FFDSSC can be utilized, the performances of the FF-DSSCs will probably be further improved. In this context, FF-DSSCs with multi-working electrodes (MWFF-DSSC(x), x=2, 3, 4, 5, and 6) were designed. As shown in Fig. 1, an MWFF-DSSC(6) contained a Pt microwire counter electrode and six Ti microwire working electrodes. The Ti microwires had been submitted to two-step anodization beforehand and were thus covered with highly ordered titanium dioxide (TiO2) nanotube arrays. Then they were sensitized with dye molecules. All the electrodes were inserted into a flexible plastic capillary tube with the sole Pt microwire counter electrode surrounded by the six Ti microwire working electrodes. Subsequently electrolyte was injected into the plastic tube. When the Ti microwires covered with as-prepared TiO2 nanotube arrays were used as the working electrodes, an MWFF-DSSC(6) exhibited obviously higher conversion efficiency than other MWFF-DSSC(x)s (x=2, 3, 4, or 5) and the FF-DSSC with single-working electrode (SWFF-DSSC) in this work. When the TiO2 nanotube arrays were treated in niobium isopropoxide solution, the conversion efficiency was further improved. Moreover, the high conversion efficiency did not suffer considerable loss under bending.
J. Liang et al. Similar to the previous works in this lab, TiO2 nanotube arrays with good orderliness and alignment were obtained by the two-step anodization process [22,30]. Fig. 2a and b shows the top and side view scanning electron microscopy (SEM) images of a typical TiO2 nanotube array. The nanotube array has a neat and tidy surface and nanotubes in it are uniform in size and alignment. As described in the previous work in this lab, this good orderliness played a favorable role in guaranteeing the good performance of the DSSCs [22]. In the X-ray photoelectron spectroscopy (XPS) results shown in Fig. 2c, the C 1s photoelectron line, which is found in the spectra of all the samples and generally believed to have arisen from adventitious carbon caused by exposure to the atmosphere, is used as the standard for calibration during data processing. In accordance with the previous results in this lab, with the energy of the C 1s line set at 284.6 eV, the energies of the Ti 2p1/2 and 2p3/2 lines are identified [22,31]. The oxygen-related lines can also be easily identified. The spectra of both the as-prepared samples and the samples treated in niobium isopropoxide are dominated by the Ti and O lines, confirming that elemental titanium and oxygen were the major elements on all the surfaces of the working electrodes. An analysis on the chemical shifts of the Ti lines suggests that the Ti element existed in the form of TiO2. In the spectrum acquired from the TiO2 nanotube array treated with niobium isopropoxide, two lines at 209.7 and 206.8 eV are assignable to Nb 3d3/2 and Nb 3d5/2, respectively, indicating that the surface of the TiO2 nanotube arrays were covered with niobium pentoxide (Nb2O5) after the niobium isopropoxide treatment [22,31]. In the X-ray diffraction (XRD) patterns shown in Fig. 2d, all the diffraction peaks are consistent with the anatase phase of TiO2 (JCPDS no. 21-1272). The absence of Nb2O5 peak in the sample treated with niobium isopropoxide is attributed to the thinness of the Nb2O5 barrier layer [22,31]. Three types of FF-DSSCs were assembled: (i) an SWFF-DSSC based on an as-prepared TiO2 nanotube array, (ii) MWFF-DSSC (x)s (x=2, 3, 4, 5, and 6) based on as-prepared TiO2 nanotube arrays and (iii) an MWFF-DSSC(6) based on Nb2O5-covered TiO2 nanotube arrays. They were all 1.7 cm in length. The SWFFDSSC and MWFF-DSSC(6), in either the straight or the bending form, are shown in Fig. 3. As shown in Fig. 3a and b, the working and counter electrodes were extracted at the two ends of the cell in both the SWFF-DSSC and MWFF-DSSC(6). The difference between them was that the SWFF-DSSC contained only one Ti microwire working electrode and the MWFF-DSSC(6) contained six. In the measurement of their photovoltaic performances, the six working electrodes in the MWFF-DSSC(6) were connected together outside the cell using soldering tin, so that the photoelectrons generated on all the working electrodes were utilized. Moreover, as the sole Pt counter electrode was shared by the six Ti working electrodes in the MWFF-DSSC(6), larger photocurrent could be extracted from the Pt counter electrode in the MWFF-DSSC(6) than from that in the SWFF-DSSC. In this sense, the cost on Pt material consumed by per unit current was reduced. As shown in Fig. 3c and d, the MWFF-DSSC(6) was as flexible as the SWFF-DSSC and they were both competitive with the previously reported FF-DSSCs in terms of flexibility [19–22]. The relations between the photocurrent densities (J) and the voltages (V) of the SWFF-DSSC and Nb2O5-free MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) were measured and are given in Fig. 4a.
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
503
Fig. 1 Schematic diagram and equivalent circuit of an MWFF-DSSC(6) with six working electrodes and one counter electrode.
Fig. 2 The TiO2 nanotube arrays fabricated in the second-step anodization on the Ti microwire electrode surfaces. (a) A top view SEM image; (b) a side view SEM image; (c) XPS results and (d) XRD patterns. (The upper graphs in (c) and (d) were acquired from the sample after the treatment in niobium isopropoxide; the lower graphs from the untreated sample. Letter “A” in (d) means “anatase”).
The photovoltaic properties of the FF-DSSCs are characterized by the four most important parameters, namely, the shortcircuit current density (JSC), open-circuit voltage (VOC), fill factor (FF), and the power conversion efficiency (η). Their vaules are summarized in Table 1. As mentioned in the beginning of this article, in most of the previous works, the conversion efficiencies of the SWFF-DSSCs were calculated with only the light incident to the working electrode taken into consideration. That is, the product of the diameter and effective length of the working electrode was used as the cross section of the incident light [6,17–28]. In this article, in contrast, the product of the diameter and effective length of the capillary tube, instead of the working electrode, is used as the cross
section of the incident light, so that the light received by the whole DSSC is taken into consideration and a direct comparison with the conventional DSSCs can be made. All the Nb2O5-free MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) had a much larger JSC and a slightly lower VOC than the SWFF-DSSC. Meanwhile, as the number of the working electrodes increased, the JSC increased and the VOC decreased, reaching their respective maximum and minimum in the MWFF-DSSC(6). The disparities in JSC and VOC are attributed to the difference in the number of the working electrodes. The TiO2 surface area increased with the number of the working electrodes, raising both the number of loaded dye molecules and the number of electron recombination events. Therefore, JSC was raised at the expense of VOC. Because the
504
J. Liang et al.
Fig. 3 Pictures of the two types of FF-DSSCs. (a) a straight SWFF-DSSC; (b) a straight MWFF-DSSC(6); (c) an SWFF-DSSC under bending; and (d) an MWFF-DSSC(6) under bending.
Fig. 4 Photovoltaic properties of the 1.7 cm long FF-DSSCs. (a) J–V curves of the SWFF-DSSC and Nb2O5-free MWFF-DSSC(x)s (x= 2, 3, 4, 5, and 6); (b) J–V curves of individual working electrodes disconnected with each other in an MWFF-DSSC(6).
increase in JSC was much more remarkable than the decrease in VOC, the η of the MWFF-DSSCs was still considerably improved. For understanding the role of each working electrode in an MWFF-DSSC(6), a cell was made with the six working electrodes disconnected with each other. The outputs of all the six electrodes were measured separately and the results are shown in Fig. 4b and Table 2. As shown in the inset of Fig. 4b, the electrodes can be grouped into three pairs according to their relative positions with respect to the incident light: electrodes F1 & F2 were on the “front” surface of the cell that directly received the illumination; electrodes S1 & S2 were on the “side” surface that the light illuminated approximately along the tangent direction; and electrodes B1 & B2 were on the “back” surface of the cell that received hardly any direct illumination. Fig. 4b and Table 2 show that all the six working electrodes could output photovoltages and photocurrents. Each pair of electrodes, i.e., F1 & F2, S1 & S2, and B1 & B2, behaved quite alike, obviously due to the similarity in their positions under illumination. As can be readily expected, the JSCs and ηs obtained from electrodes F1 & F2 were the highest and those from electrodes B1 & B2 the lowest. Since the dye molecules around electrodes F1 & F2 were under direct illumination, most of the photoelectrons could reach the electrodes without traveling a long distance, thus the illumination to electrodes F1 & F2 was most effective in generating photocurrent. The dye molecules around
electrodes S1 & S2 received much less direct illumination and those around electrodes B1 & B2 even received very little direct illumination. However, due to the scattering of the incident light inside the cell, some photoelectrons could still be generated around these four electrodes. Therefore, electrodes S1 & S2 and B1 & B2 also delivered photocurrents under illumination. Noteworthy is that the sum of the JSCs of all the individual working electrodes shown in Table 2, 18.4 mA cm 2, is larger than the JSC of the MWFF-DSSC (6) shown in Table 1, 15.6 mA cm 2. This disparity is quite understandable. First, the results shown in the last but one row of Table 1 and in Table 2 were obtained from two different samples. Second and more importantly, the conditions around a working electrode, e.g., the electric field distribution and the recombination probabilities of the photoelectrons, would be different when the working electrode was working alone and working in parallel with other five working electrodes simultaneously. Moreover, as shown in Fig. 5a, the performance of the MWFF-DSSC(6) was remarkably boosted by treating the TiO2 nanotube arrays in niobium isopropoxide solution. As shown in the last two rows of Table 1, the VOC increased from 0.61 to 0.66 V, the JSC from 15.6 to 18.8 mA cm 2, the FF from 0.69 to 0.73 and the η from 6.6% to 9.1%. This improvement is ascribable to the formation of a Nb2O5 energy barrier on the surface of the TiO2 nanotube arrays, which resulted from the treatment in niobium isopropoxide [22,32–34]. The
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
Table 1
Photovoltaic properties of the FF-DSSCs.
FF-DSSC type
VOC (V)
FF
η (%)
6.1 9.4 11.6 13.6 15.2 15.6
0.67 0.66 0.65 0.64 0.63 0.61
0.70 0.59 0.63 0.57 0.57 0.69
2.8 3.6 4.7 4.9 5.4 6.6
18.8
0.66
0.73 9.1
JSC (mA cm
SWFF-DSSC MWFF-DSSC(2) MWFF-DSSC(3) MWFF-DSSC(4) MWFF-DSSC(5) MWFF-DSSC(6) without Nb2O5 MWFF-DSSC(6) with Nb2O5
2
)
Table 2 Photovoltaic properties of individual working electrodes disconnected with each other in an MWFFDSSC(6). Electrodes
JSC (mA cm
F1 F2 S1 S2 B1 B2
4.2 4.1 3.6 3.4 1.6 1.5
2
)
VOC (V)
FF
η (%)
0.61 0.61 0.62 0.60 0.57 0.57
0.63 0.66 0.65 0.65 0.78 0.77
1.6 1.7 1.4 1.3 0.7 0.7
conduction band edge potential of Nb2O5 is about 100 mV more negative than that of TiO2, thus the conduction band edge of Nb2O5 locates between the conduction band edge of TiO2 and the excited states of the dye molecules. Therefore, the Nb2O5 layer served as an energy barrier to the electrons in the conduction band of TiO2. This energy barrier slowed the recombination at the TiO2/dye/electrolyte interface, rendering more electrons to accumulate in the conduction band of TiO2. Consequently, more photoexcited electrons could flow via the external circuit and a larger JSC resulted. Moreover, because the recombination at the TiO2/dye/electrolyte interface is the major loss mechanism of VOC, which is determined by the difference between the electron quasi-Fermi level in the TiO2 film under illumination and the redox electrolyte level, the existence of the Nb2O5 energy barrier also helped to maintain the VOC at a higher level [31–35]. The Nyquist plots of the MWFF-DSSC(6), either with or free from Nb2O5, are given in Fig. 5b. The semicircles are attributable to the electron recombination at the TiO2/dye/electrolyte interface. The recombination resistance in the MWFF-DSSC (6) with Nb2O5 was larger than that in the MWFF-DSSC (6) free from Nb2O5, indicating that the Nb2O5 barrier layer did weaken the electron recombination process [22,31]. For the purpose of testing the reproducibility of the above results, five more MWFF-DSSC(6) with Nb2O5 were fabricated. The photovoltaic performances of these cells were quite consistent with those given in Fig. 5a and Table 1 (Fig. S1 and Table S1, Supporting information).
505
A comparison can be made between the MWFF-DSSC(6) in this work and the conventional DSSCs with P25 films on FTO substrates used as their photoanodes. The effective areas and conversion efficiencies of the conventional DSSCs are usually between 0.1 cm2 to 0.3 cm2 and 5% to 12%, respectively (Table S2, Supporting information) [2,36–39]. In this study, the effective area of the MWFF-DSSC(6) was 1.7 0.08=0.14 cm2 and the conversion efficiency of the MWFF-DSSC(6) with Nb2O5 reached 9.1%, suggesting that the MWFF-DSSC(6) were competitive with the conventional DSSCs based on P25 films. (It might be necessary to emphasize that the 0.14 cm2 was the flank area of the whole cell instead of the area of the Ti microwire only and all the light illuminating this area was taken into consideration in computing the conversion efficiencies.) Furthermore, to the authors’ best knowledge, a conversion efficiency larger than 9% has not been achieved from any conventional DSSCs based on TiO2 nanotube array photoanodes (Table S2, Supporting information) [40–42]. In this sense, this study has shown that increasing the number of working electrodes is an effective strategy for improving the performances of the DSSCs. For the purpose of investigating how the length of an MWFF-DSSC(6) influenced the conversion efficiency, a 3.2 cm long MWFF-DSSC(6) with Nb2O5, which was almost twice as long as the 1.7 cm long cell described above, was fabricated and its photovoltaic properties were measured (Fig. S2, Supporting information). The area under illumination then became 3.2 0.08=0.26 cm2 and the conversion efficiency dropped to 8.1%. Because the space between the six working electrodes and one counter electrode was rather narrow, the longer the tube was, the more difficult it was to fully fill the space with the electrolyte owing to the viscous resistance to the electrolyte from the electrodes and the capillary tube. Nonetheless, a comparison shows that this drop was not very serious and a prolonged MWFF-DSSC(6) can still be considered as competitive with the conventional DSSCs. An obvious advantage of the MWFF-DSSC(6) against conversional DSSCs is that the MWFF-DSSC(6) can harvest light from all directions perpendicular to the fiber. That is, incident light from any direction perpendicular to the fiber can generate electricity. In this work, as indicated in Fig. 6a, light was allowed to illuminate the 1.7 cm long MWFF-DSSC (6) with Nb2O5 respectively from the three directions indicted by points A, B, and C. Three J–V curves were respectively acquired from the cell under illumination of the three incident directions and they are collectively shown in Fig. 6b. The corresponding photovoltaic performance parameters are listed in Table 3. The first three rows of Table 3 show that the performances of the MWFF-DSSC (6) were almost the same when the incident light illuminated the cell from different directions. In Fig. 6b, the three curves almost coincide with each other. Analogously, although the Nb2O5-free MWFF-DSSC(6) had lower conversion efficiencies than the MWFF-DSSC(6) with Nb2O5, the Nb2O5-free MWFFDSSC(6) also had the advantage that the light illuminating the cell from different directions led to quite similar outputs (Fig. S3 and Table S3, Supporting information). When fewer than six working electrodes were used in the Nb2O5-free MWFF-DSSCs, the capabilities of the cells harvesting light were obviously dependent on the incident direction of the light (Fig. S4, Supporting information). In the measurements of the photovoltaic properties under bending, the MWFF-DSSC(6) in this work appeared to be more
506
J. Liang et al.
Fig. 5 Photovoltaic properties of the MWFF-DSSC(6)s either with or free from Nb2O5. (a) J–V curves (The curve of the Nb2O5-free MWFF-DSSC(6) is similar to the one in Fig. 4a); (b) electrical impedance spectroscopy (EIS) results.
Fig. 6 Photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 under illumination from different directions. (a) The three vertical directions of the incident light (indicated by points A, B, and C, respectively); (b) the three J–V curves acquired with the cell respectively illuminated from the three vertical incident directions.
Table 3 Dependence of the photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 on the incident direction of the light and bending angle of the cell. Bending angles (1)
Point JSC (mA cm
0 0 0 30 90 180
A B C A A A
18.8 19.1 18.6 18.3 18.0 17.8
2
)
VOC (V)
FF
η (%)
0.66 0.65 0.66 0.68 0.65 0.66
0.73 0.72 0.74 0.71 0.72 0.72
9.1 9.0 9.1 8.8 8.4 8.5
flexible than the conventional flexible DSSCs. The effective length of the wire subjected to illumination was 1.7 cm. It is assumed that the bending was uniform along the whole MWFF-DSSC(6), i.e., the curvature radius was equal at every point. As shown in Fig. 7a, the bending angle is defined as the central angle subtended by the fiber under bending. Three J–V curves that were obtained from the MWFF-DSSC(6) with Nb2O5 under 301, 901 and 1801 bending angles, respectively,
are collectively shown in Fig. 7b. The corresponding photovoltaic parameters are listed in Table 3. As shown in the last three rows of Table 3, on the one hand, the conversion efficiency suffered a lost when the cell was under bending, probably due to the partial damage of the highly ordered TiO2 nanotube arrays; on the other hand, this lost was not very considerable. For instance, even under a 1801 bending, the cell retained about 93% of its conversion efficiency in comparison with the straight form (η=8.5% vs. η=9.1%). Accordingly, the three curves are hardly distinguishable in Fig. 7b. The behavior of the Nb2O5-free MWFF-DSSC(6) under bending was quite similar to that of the MWFF-DSSC(6) with Nb2O5 (Fig. S5 and Table S3, Supporting information). In summary, new-type FF-DSSCs with multi-working electrodes, namely MWFF-DSSC(x)s (x =2, 3, 4, 5, and 6), have been designed and realized. In each MWFF-DSSC(6), the working electrodes were six Ti microwires covered with highly ordered TiO2 nanotube arrays that surrounded a Pt microwire. Importantly, for a more reasonable comparison with the conventional DSSCs, the flank cross section of the whole cell, instead of that of only a Ti microwire, was used as the cross section area of the incident light in the computation of the conversion efficiency. Even when the TiO2 nanotube arrays on the working electrodes were not
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
507
Fig. 7 Photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 under bending. (a) Schematic of the MWFF-DSSC (6) under different bending angles; (b) J–V curves.
covered with Nb2O5, the conversion efficiency of the MWFFDSSC(6) reached 6.6%, which was higher than those of the MWFF-DSSCs with two to five working electrodes and the SWFF-DSSC (using the computation method put forward in this article). The conversion efficiency was further raised to 9.1% by covering the TiO2 nanotube array surfaces with Nb2O5 layers. This value, within the authors’ knowledge, was the highest among all the reported FF-DSSCs and was also competitive with the conventional DSSCs. Furthermore, the MWFF-DSSC(6) could largely retain the good performances under bending. For example, the conversion efficiency of the MWFF-DSSC(6) with Nb2O5 only suffered a 6.6% decrease when the fiber was under a 1801 bending. Therefore, the feasibility of improving the performance of the FFDSSCs by employing more working electrodes has been proved. Following this strategy, a considerable expansion of the application scope of the FF-DSSCs can be expected.
Experimental methods Synthesis of photoanodes Highly ordered TiO2 nanotube arrays were fabricated in a modified anodization process [30]. The primary component of the electrolyte employed in anodizing a Ti microwire (0.25 mm in diameter, 99.7% in purity, Alfa aesar) was an ethylene glycol (C2H6O2) solution containing NH4F (0.25% in mass) and H2O (0.6% in volume). In each fabrication, the Ti microwire was cleaned ultrasonically in turn in acetone, de-ionized water and ethanol for 15 min each. Then, the Ti microwire was attached to the positive electrode of a DC power source. A piece of graphite was connected to the negative electrode and immersed in the electrolyte approximately 5 cm away from the Ti microwire. All the experiments were done at room temperature. For a better ordering and alignment of the TiO2 nanotube arrays, the Ti microwire was anodized twice. The first anodization step served to provide a regular pattern on the Ti surface, which worked as a starting template for the second anodization. In the first-step anodization, the Ti microwire was anodized under a 60 V voltage for 6 h. Then, the nanotube layer was removed ultrasonically in deionized water. In the second-step anodization, the Ti microwire, on whose surface a regular pattern remained, was again submitted to anodization at 60 V for 12 h. A highly ordered TiO2 nanotube array was obtained in this
second-step anodization. Finally, for crystallizing this TiO2 nanotube array, the sample was annealed in air at 500 1C for 30 min with a ramp rate of 2 1C min 1 and cooled to room temperature naturally. For the purpose of introducing a Nb2O5 layer on its surface, the Ti microwire covered with the TiO2 nanotube array was dipped into a 5 mM niobium isopropoxide solution at room temperature for 30 s. Then, the microwire was flushed with isopropanol, dried and finally annealed again under the same condition as described above.
Device fabrication The working electrodes, either with or without Nb2O5 layers, were immersed into 0.3 mM cis-diisothiocyanato-bis (2,20-bipyridyl-4,40-dicarboxylato) ruthenium(II) bis(tetrabutylammonium) (N719, Dalian HeptaChroma SolarTech, China) solution for 24 h after they cooled down from sintering at about 80 1C. Then they were rinsed with ethanol to remove physically adsorbed dye molecules. The counter electrode in each solar cell was a Pt wire (0.25 mm in diameter, 99.9% in purity, Alfa aesar). The FF-DSSCs were assembled in two ways, respectively. First, the method introduced in previous literatures was used [6,17–28]. One Ti microwire and one Pt microwire were inserted into a flexible plastic capillary tube, 0.8 mm in diameter, so that an FF-DSSC with only one working electrode, namely an SWFF-DSSC, was made. Its appearance is shown in Fig. 3a. Second, a number of Ti microwires and a Pt microwire surrounded by them were collectively inserted into a flexible plastic capillary tube, also 0.8 mm in diameter, so that an MWFF-DSSC(x) was made, where x = 2, 3, 4, 5, or 6 denotes the number of the working electrodes. The configuration and appearance of an MWFF-DSSC(6) are given in Figs. 1 and 3b, respectively. Then I /I3 based liquid electrolyte was injected into the capillary tube. At last, the two ends of the capillary tube were sealed with sealing glue to prevent evaporation of the electrolyte.
Characterization techniques The morphologies and microstructures of the samples were observed using a field-emission scanning electron microscope (FESEM). The chemical compositions and crystallinities of the samples were investigated using XPS and XRD, respectively. The
508
J. Liang et al.
photovoltaic performance of the DSSCs was measured under the illumination of an artificial sunlight with the standard air mass 1.5 (AM 1.5) spectrum and a 100 mW cm 2 irradiance (Oriel Solar Simulator, Model 91160). The current–voltage characteristics of the cells were recorded by an electrochemical analyzer. Following the practice of most researchers, the backward sweeping direction was followed in the measurements in this work. That is, the voltages on the two terminals of the cells were swept from 0.9 to 0 V. As suggested by Yang et al. [43], if the forward direction, i.e., from 0 to 0.9 V, had been followed, poorer performances would have been obtained (Fig. S2, Supporting information). The EIS was acquired using an electrochemical workstation with the frequency range from 0.1 Hz to 100 kHz in the dark condition. The effective areas of the SWFF-DSSC and the MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) were both 0.14 cm2, which was the product of the diameter of the capillary tube, 0.8 mm, and the effective length of the microwires subjected to illumination, 1.7 cm.
Acknowledgment This work was supported by the National Natural Science Foundation of China (nos. 61171023, 61306079 and 61376059).
Appendix A.
Supporting information
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2015.01.023.
References [1] B. Oregan, M. Gratzel, Nature 353 (1991) 737–740. [2] A. Yella, H.-W. Lee, H.N. Tsao, C. Yi, A.K. Chandiran, M.K. Nazeeruddin, E.W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Graetzel, Science 334 (2011) 629–634. [3] M. Gratzel, Nature 414 (2001) 338–344. [4] W.-Q. Wu, Y.-F. Xu, H.-S. Rao, C.-Y. Su, D.-B. Kuang, J. Am. Chem. Soc. 136 (2014) 6437–6445. [5] X.A. Jeanbourquin, X. Li, C. Law, P.R.F. Barnes, R. HumphryBaker, P. Lund, M.I. Asghar, B.C. O’Regan, J. Am. Chem. Soc. 136 (2014) 7286–7294. [6] W.X. Guo, C. Xu, G. Zhu, C.F. Pan, C.J. Lin, Z.L. Wang, Nano Energy 1 (2012) 176–182. [7] J. Liang, G. Zhang, H. Xia, W. Sun, RSC Adv. 4 (2014) 12649–12652. [8] B. Liu, E.S. Aydil, J. Am. Chem. Soc. 131 (2009) 3985–3990. [9] J.R. Jennings, A. Ghicov, L.M. Peter, P. Schmuki, A.B. Walker, J. Am. Chem. Soc. 130 (2008) 13364–13372. [10] J. Khamwannah, Y.Y. Zhang, S.Y. Noh, H. Kim, C. Frandsen, S.D. Kong, S. Jin, Nano Energy 1 (2012) 411–417. [11] L. Li, T.Y. Zhai, Y. Bando, D. Golberg, Nano Energy 1 (2012) 91–106. [12] Y. Peng, J. Zhong, K. Wang, B.F. Xue, Y.B. Cheng, Nano Energy 2 (2013) 235–240. [13] H.C. Weerasinghe, F.Z. Huang, Y.B. Cheng, Nano Energy 2 (2013) 174–189. [14] H. Sun, Z. Yang, X. Chen, L. Qiu, X. You, P. Chen, H. Peng, Angew. Chem. Int. Ed. 52 (2013) 8276–8280. [15] S. Ito, N.-L.C. Ha, G. Rothenberger, P. Liska, P. Comte, S.M. Zakeeruddin, P. Pechy, M.K. Nazeeruddin, M. Graetzel, Chem. Commun. (2006) 4004–4006.
[16] V. Galstyan, A. Vomiero, E. Comini, G. Faglia, G. Sberveglieri, RSC Adv. 1 (2011) 1038–1044. [17] X. Fan, Z. Chu, F. Wang, C. Zhang, L. Chen, Y. Tang, D. Zou, Adv. Mater. 20 (2008) 592–595. [18] Y.P. Fu, M. Peng, Z.B. Lv, X. Cai, S.C. Hou, H.W. Wu, X. Yu, H. Kafafy, D.C. Zou, Nano Energy 2 (2013) 537–544. [19] Z. Liu, M. Misra, ACS Nano 4 (2010) 2196–2200. [20] M.R. Lee, R.D. Eckert, K. Forberich, G. Dennler, C.J. Brabec, R.A. Gaudiana, Science 324 (2009) 232–235. [21] W. Wang, Q. Zhao, H. Li, H. Wu, D. Zou, D. Yu, Adv. Funct. Mater. 22 (2012) 2775–2782. [22] J. Liang, J. Yang, G. Zhang, W. Sun, Electrochem. Commun. 37 (2013) 80–83. [23] F.J. Cai, T. Chen, H.S. Peng, J. Mater. Chem. 22 (2012) 14856–14860. [24] S.W. Pan, Z.B. Yang, P.N. Chen, J. Deng, H.P. Li, H.S. Peng, Angew. Chem. Int. Ed. 53 (2014) 6110–6114. [25] Z.B. Yang, H. Sun, T. Chen, L.B. Qiu, Y.F. Luo, H.S. Peng, Angew. Chem. Int. Ed. 52 (2013) 7545–7548. [26] S.W. Pan, Z.B. Yang, H.P. Li, L.B. Qiu, H. Sun, H.S. Peng, J. Am. Chem. Soc. 135 (2013) 10622–10625. [27] X. Fang, Z.B. Yang, L.B. Qiu, H. Sun, S.W. Pan, J. Deng, Y.F. Luo, H.S. Peng, Adv. Mater. 26 (2014) 1694–1698. [28] Z.B. Yang, J. Deng, X.M. Sun, H.P. Li, H.S. Peng, Adv. Mater. 26 (2014) 2643–2647. [29] P. Ball, Nature (2001). http://dx.doi.org/10.1038/news010215-9. [30] S. Li, G. Zhang, D. Guo, L. Yu, W. Zhang, J. Phys. Chem. C 113 (2009) 12759–12765. [31] J. Liang, G. Zhang, W. Sun, RSC Adv. 4 (2014) 6746–6751. [32] K. Sayama, H. Sugihara, H. Arakawa, Chem. Mater. 10 (1998) 3825–3832. [33] S.G. Chen, S. Chappel, Y. Diamant, A. Zaban, Chem. Mater. 13 (2001) 4629–4634. [34] A. Zaban, S.G. Chen, S. Chappel, B.A. Gregg, Chem. Commun. (2000) 2231–2232. [35] N. Kopidakis, N.R. Neale, A.J. Frank, J. Phys. Chem. B 110 (2006) 12485–12489. [36] S. Li, L. Qiu, C. Shi, X. Chen, F. Yan, Adv. Mater. 26 (2014) 1266–1271. [37] J. Yang, P. Ganesan, J. Teuscher, T. Moehl, Y.J. Kim, C. Yi, P. Comte, K. Pei, T.W. Holcombe, M.K. Nazeeruddin, J. Hua, S.M. Zakeeruddin, H. Tian, M. Graetzel, J. Am. Chem. Soc. 136 (2014) 5722–5730. [38] J. Luo, M. Xu, R. Li, K.-W. Huang, C. Jiang, Q. Qi, W. Zeng, J. Zhang, C. Chi, P. Wang, J. Wu, J. Am. Chem. Soc. 136 (2014) 265–272. [39] J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T. Bessho, A. Marchioro, E. Ghadiri, J.-E. Moser, C. Yi, M.K. Nazeeruddin, M. Graetzel, Nat. Commun. 3 (2012). [40] F. Zhuge, J. Qiu, X. Li, X. Gao, X. Gan, W. Yu, Adv. Mater. 23 (2011) 1330–1334. [41] M. Ye, D. Zheng, M. Lv, C. Chen, C. Lin, Z. Lin, Adv. Mater. 25 (2013) 3039–3044. [42] M. Ye, X. Xin, C. Lin, Z. Lin, Nano Lett. 11 (2011) 3214–3220. [43] X.D. Yang, M. Yanagida, L.Y. Han, Energy Environ. Sci. 6 (2013) 54–56. Jia Liang received his B.S. degree in 2010 from Southeast University, Nanjing, China. He is a Ph.D. candidate of Key Laboratory for Physics and Chemistry of Nanodevices in the Department of Electronics, School of Electrical Engineering and Computer Science at Peking University. He joined Dr. Jun Lou’s research group in Rice University as a visiting student in the first half of 2014. His main research interest is synthesis of energy nanomaterials and their applications in solar cells and catalyses.
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
Gengmin Zhang received his B.Sc. and Ph.D. in Science degrees both from Peking University, China, in 1991 and 1996, respectively. His thesis for the Ph.D. degree was on the photoemission from the oxide-coated cathodes under laser illumination. Supported by a Sino-Japanese intergovernmental project, he studied the behaviors of organic molecules on the graphite surface in Osaka University, Japan, from 1998 to 2000. His current research is focused on the photovoltaic, ferroelectric and field emission properties of nanomaterials. Dr. Wentao Sun received her B.S. and Ph.D. degrees from Central South University of Technology and Technical Institute of Physics and Chemistry, CAS in 1999 and 2005, respectively. Then she was a Postdoctoral Research Fellow in Peking University from 2005 to 2008. In 2008, she joined Department of Electronics of Peking University as an Assistant Professor. In 2011, she was promoted as an Associate Professor. Dr. Sun’s current research interests mainly focus on nanomaterial synthesis and nanoenergy devices.
509
Dr. Pei Dong received her Ph.D. in Mechanical Engineering and Materials Science from Rice University in 2013, under the guidance of Prof. Jun Lou. She is currently a joint Postdoctoral Research Fellow in Prof. Pulickel M. Ajayan’s and Prof. Jun Lou’s groups in Department of Materials Science and NanoEngineering at Rice University. Her research mainly focuses on nanomaterials enabled dye-sensitized solar cells and understanding self-stiffening mechanisms in nanocomposites.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nanoenergy
RAPID COMMUNICATION
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes Jia Lianga,b, Gengmin Zhanga,c,n, Wentao Suna, Pei Dongb a
Key Laboratory for the Physics and Chemistry of Nanodevices and Department of Electronics, Peking University, Beijing 100871, China b Department of Materials Science and NanoEngineering, Rice University, 6100 Main Street, Houston, TX 77005, United States c SIP-UCLA Institute for Technology Advancement, Suzhou 215123, Jiangsu Province, China Received 3 September 2014; received in revised form 9 January 2015; accepted 10 January 2015 Available online 21 January 2015
KEYWORDS
Abstract
Multi-working electrodes; Flexible; TiO2 nanotube arrays; Dye-sensitized solar cells
Novel flexible fiber-type dye-sensitized solar cells (FF-DSSCs) with multi-working electrodes (MWFFDSSCs) have been developed. In each MWFF-DSSC, all the components are assembled into a flexible plastic capillary tube. A Pt microwire along the axis of the tube is used as the sole counter electrode and a number of Ti microwires surrounding it, which are all covered with highly ordered titanium dioxide (TiO2) nanotube arrays, are jointly used as the working electrodes. This new configuration brings about good flexibility, capability of harvesting light from all directions and a conversion efficiency competitive with those of the conventional DSSCs. The photovoltaic performances of the MWFF-DSSC with six working electrodes (MWFF-DSSC(6)) are better than those of the MWFF-DSSCs with two to five working electrodes (MWFF-DSSC(x), x=2, 3, 4, or 5) and the FF-DSSC with a single working electrode (SWFF-DSSC). When the as-prepared TiO2 nanotube arrays are used as the working electrodes, a 6.6% conversion efficiency is obtained from an MWFF-DSSC(6). When the TiO2 nanotube arrays are treated in the niobium isopropoxide solution, the conversion efficiency is further raised to 9.1% and only suffers a 6.6% relative decrease even under a 1801 bending. & 2015 Elsevier Ltd. All rights reserved.
n Corresponding author at: Key Laboratory for the Physics and Chemistry of Nanodevices and Department of Electronics, Peking University, Beijing 100871, China Tel.: +86 10 62751773; fax: +86 10 62762999. E-mail address: [email protected] (G. Zhang).
http://dx.doi.org/10.1016/j.nanoen.2015.01.023 2211-2855/& 2015 Elsevier Ltd. All rights reserved.
Since its invention in 1991, the dye-sensitized solar cell (DSSC) has attracted significant attention as a promising photovoltaic device due to its low cost and convenience in fabrication [1]. So far, tremendous efforts have been devoted to the fundamental and experimental researches on the DSSCs for improving their power conversion efficiencies and compatibilities to practical applications [2]. A conventional
502 DSSC is made up of a photoanode (also termed as a working electrode), a counter electrode and an electrolyte filled in the space between them. Generally, the photoanode is a fluorine-doped tin oxide (FTO) glass substrate covered with a dye-sensitized titanium dioxide (TiO2) film and the counter electrode is a platinized conductive glass [3]. The eventual performance of a DSSC is a synergistic effect of all these components [4–11]. Recently, researchers have been increasingly interested in the flexible DSSCs, which feature lightweight, low cost rollto-roll process and extensive application perspectives. Such substrates as conducting plastics, metal sheets, and metal wires are used in the flexible DSSCs [6,12–19]. Among them, the metal wires are considered to be more promising than the relatively poorly conducting flexible plastics and opaque metal sheets. Thus the flexible fiber-type dye-sensitized solar cells (FF-DSSCs) based on metal wires are very attractive [20– 28]. Although the FF-DSSCs have already exhibited good performances, there is still room for further improvement. First, the use of a platinum (Pt) microwire as the counter electrode sacrifices the advantage of low cost of an FF-DSSC. Second, in the calculation of the conversion efficiency of an FF-DSSC, the product of the diameter and effective length of the working electrode is usually used to represent the cross section of the incident light [6,17–28]. That is, only the light incident to the flank side of the working electrode is taken into consideration and the light incident to other part of an FF-DSSC is ignored. Because the light incident to the part other than the working electrode, which is not utilized effectively, is not included in the computation, the denominator in computing the conversion efficiency is unreasonably underestimated. In contrast, in a conventional DSSC, the photoanode receives all the light incident to the cell. Therefore, for a more reasonable comparison with the conventional DSSCs, the product of the diameter and effective length of the capillary tube, rather than the working electrode, should be used to represent the cross section of the incident light [29]. Accordingly, if the Pt counter electrode can be utilized more efficiently, i.e., larger photocurrent can be extracted from a single Pt microwire, and all the light incident to an FFDSSC can be utilized, the performances of the FF-DSSCs will probably be further improved. In this context, FF-DSSCs with multi-working electrodes (MWFF-DSSC(x), x=2, 3, 4, 5, and 6) were designed. As shown in Fig. 1, an MWFF-DSSC(6) contained a Pt microwire counter electrode and six Ti microwire working electrodes. The Ti microwires had been submitted to two-step anodization beforehand and were thus covered with highly ordered titanium dioxide (TiO2) nanotube arrays. Then they were sensitized with dye molecules. All the electrodes were inserted into a flexible plastic capillary tube with the sole Pt microwire counter electrode surrounded by the six Ti microwire working electrodes. Subsequently electrolyte was injected into the plastic tube. When the Ti microwires covered with as-prepared TiO2 nanotube arrays were used as the working electrodes, an MWFF-DSSC(6) exhibited obviously higher conversion efficiency than other MWFF-DSSC(x)s (x=2, 3, 4, or 5) and the FF-DSSC with single-working electrode (SWFF-DSSC) in this work. When the TiO2 nanotube arrays were treated in niobium isopropoxide solution, the conversion efficiency was further improved. Moreover, the high conversion efficiency did not suffer considerable loss under bending.
J. Liang et al. Similar to the previous works in this lab, TiO2 nanotube arrays with good orderliness and alignment were obtained by the two-step anodization process [22,30]. Fig. 2a and b shows the top and side view scanning electron microscopy (SEM) images of a typical TiO2 nanotube array. The nanotube array has a neat and tidy surface and nanotubes in it are uniform in size and alignment. As described in the previous work in this lab, this good orderliness played a favorable role in guaranteeing the good performance of the DSSCs [22]. In the X-ray photoelectron spectroscopy (XPS) results shown in Fig. 2c, the C 1s photoelectron line, which is found in the spectra of all the samples and generally believed to have arisen from adventitious carbon caused by exposure to the atmosphere, is used as the standard for calibration during data processing. In accordance with the previous results in this lab, with the energy of the C 1s line set at 284.6 eV, the energies of the Ti 2p1/2 and 2p3/2 lines are identified [22,31]. The oxygen-related lines can also be easily identified. The spectra of both the as-prepared samples and the samples treated in niobium isopropoxide are dominated by the Ti and O lines, confirming that elemental titanium and oxygen were the major elements on all the surfaces of the working electrodes. An analysis on the chemical shifts of the Ti lines suggests that the Ti element existed in the form of TiO2. In the spectrum acquired from the TiO2 nanotube array treated with niobium isopropoxide, two lines at 209.7 and 206.8 eV are assignable to Nb 3d3/2 and Nb 3d5/2, respectively, indicating that the surface of the TiO2 nanotube arrays were covered with niobium pentoxide (Nb2O5) after the niobium isopropoxide treatment [22,31]. In the X-ray diffraction (XRD) patterns shown in Fig. 2d, all the diffraction peaks are consistent with the anatase phase of TiO2 (JCPDS no. 21-1272). The absence of Nb2O5 peak in the sample treated with niobium isopropoxide is attributed to the thinness of the Nb2O5 barrier layer [22,31]. Three types of FF-DSSCs were assembled: (i) an SWFF-DSSC based on an as-prepared TiO2 nanotube array, (ii) MWFF-DSSC (x)s (x=2, 3, 4, 5, and 6) based on as-prepared TiO2 nanotube arrays and (iii) an MWFF-DSSC(6) based on Nb2O5-covered TiO2 nanotube arrays. They were all 1.7 cm in length. The SWFFDSSC and MWFF-DSSC(6), in either the straight or the bending form, are shown in Fig. 3. As shown in Fig. 3a and b, the working and counter electrodes were extracted at the two ends of the cell in both the SWFF-DSSC and MWFF-DSSC(6). The difference between them was that the SWFF-DSSC contained only one Ti microwire working electrode and the MWFF-DSSC(6) contained six. In the measurement of their photovoltaic performances, the six working electrodes in the MWFF-DSSC(6) were connected together outside the cell using soldering tin, so that the photoelectrons generated on all the working electrodes were utilized. Moreover, as the sole Pt counter electrode was shared by the six Ti working electrodes in the MWFF-DSSC(6), larger photocurrent could be extracted from the Pt counter electrode in the MWFF-DSSC(6) than from that in the SWFF-DSSC. In this sense, the cost on Pt material consumed by per unit current was reduced. As shown in Fig. 3c and d, the MWFF-DSSC(6) was as flexible as the SWFF-DSSC and they were both competitive with the previously reported FF-DSSCs in terms of flexibility [19–22]. The relations between the photocurrent densities (J) and the voltages (V) of the SWFF-DSSC and Nb2O5-free MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) were measured and are given in Fig. 4a.
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
503
Fig. 1 Schematic diagram and equivalent circuit of an MWFF-DSSC(6) with six working electrodes and one counter electrode.
Fig. 2 The TiO2 nanotube arrays fabricated in the second-step anodization on the Ti microwire electrode surfaces. (a) A top view SEM image; (b) a side view SEM image; (c) XPS results and (d) XRD patterns. (The upper graphs in (c) and (d) were acquired from the sample after the treatment in niobium isopropoxide; the lower graphs from the untreated sample. Letter “A” in (d) means “anatase”).
The photovoltaic properties of the FF-DSSCs are characterized by the four most important parameters, namely, the shortcircuit current density (JSC), open-circuit voltage (VOC), fill factor (FF), and the power conversion efficiency (η). Their vaules are summarized in Table 1. As mentioned in the beginning of this article, in most of the previous works, the conversion efficiencies of the SWFF-DSSCs were calculated with only the light incident to the working electrode taken into consideration. That is, the product of the diameter and effective length of the working electrode was used as the cross section of the incident light [6,17–28]. In this article, in contrast, the product of the diameter and effective length of the capillary tube, instead of the working electrode, is used as the cross
section of the incident light, so that the light received by the whole DSSC is taken into consideration and a direct comparison with the conventional DSSCs can be made. All the Nb2O5-free MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) had a much larger JSC and a slightly lower VOC than the SWFF-DSSC. Meanwhile, as the number of the working electrodes increased, the JSC increased and the VOC decreased, reaching their respective maximum and minimum in the MWFF-DSSC(6). The disparities in JSC and VOC are attributed to the difference in the number of the working electrodes. The TiO2 surface area increased with the number of the working electrodes, raising both the number of loaded dye molecules and the number of electron recombination events. Therefore, JSC was raised at the expense of VOC. Because the
504
J. Liang et al.
Fig. 3 Pictures of the two types of FF-DSSCs. (a) a straight SWFF-DSSC; (b) a straight MWFF-DSSC(6); (c) an SWFF-DSSC under bending; and (d) an MWFF-DSSC(6) under bending.
Fig. 4 Photovoltaic properties of the 1.7 cm long FF-DSSCs. (a) J–V curves of the SWFF-DSSC and Nb2O5-free MWFF-DSSC(x)s (x= 2, 3, 4, 5, and 6); (b) J–V curves of individual working electrodes disconnected with each other in an MWFF-DSSC(6).
increase in JSC was much more remarkable than the decrease in VOC, the η of the MWFF-DSSCs was still considerably improved. For understanding the role of each working electrode in an MWFF-DSSC(6), a cell was made with the six working electrodes disconnected with each other. The outputs of all the six electrodes were measured separately and the results are shown in Fig. 4b and Table 2. As shown in the inset of Fig. 4b, the electrodes can be grouped into three pairs according to their relative positions with respect to the incident light: electrodes F1 & F2 were on the “front” surface of the cell that directly received the illumination; electrodes S1 & S2 were on the “side” surface that the light illuminated approximately along the tangent direction; and electrodes B1 & B2 were on the “back” surface of the cell that received hardly any direct illumination. Fig. 4b and Table 2 show that all the six working electrodes could output photovoltages and photocurrents. Each pair of electrodes, i.e., F1 & F2, S1 & S2, and B1 & B2, behaved quite alike, obviously due to the similarity in their positions under illumination. As can be readily expected, the JSCs and ηs obtained from electrodes F1 & F2 were the highest and those from electrodes B1 & B2 the lowest. Since the dye molecules around electrodes F1 & F2 were under direct illumination, most of the photoelectrons could reach the electrodes without traveling a long distance, thus the illumination to electrodes F1 & F2 was most effective in generating photocurrent. The dye molecules around
electrodes S1 & S2 received much less direct illumination and those around electrodes B1 & B2 even received very little direct illumination. However, due to the scattering of the incident light inside the cell, some photoelectrons could still be generated around these four electrodes. Therefore, electrodes S1 & S2 and B1 & B2 also delivered photocurrents under illumination. Noteworthy is that the sum of the JSCs of all the individual working electrodes shown in Table 2, 18.4 mA cm 2, is larger than the JSC of the MWFF-DSSC (6) shown in Table 1, 15.6 mA cm 2. This disparity is quite understandable. First, the results shown in the last but one row of Table 1 and in Table 2 were obtained from two different samples. Second and more importantly, the conditions around a working electrode, e.g., the electric field distribution and the recombination probabilities of the photoelectrons, would be different when the working electrode was working alone and working in parallel with other five working electrodes simultaneously. Moreover, as shown in Fig. 5a, the performance of the MWFF-DSSC(6) was remarkably boosted by treating the TiO2 nanotube arrays in niobium isopropoxide solution. As shown in the last two rows of Table 1, the VOC increased from 0.61 to 0.66 V, the JSC from 15.6 to 18.8 mA cm 2, the FF from 0.69 to 0.73 and the η from 6.6% to 9.1%. This improvement is ascribable to the formation of a Nb2O5 energy barrier on the surface of the TiO2 nanotube arrays, which resulted from the treatment in niobium isopropoxide [22,32–34]. The
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
Table 1
Photovoltaic properties of the FF-DSSCs.
FF-DSSC type
VOC (V)
FF
η (%)
6.1 9.4 11.6 13.6 15.2 15.6
0.67 0.66 0.65 0.64 0.63 0.61
0.70 0.59 0.63 0.57 0.57 0.69
2.8 3.6 4.7 4.9 5.4 6.6
18.8
0.66
0.73 9.1
JSC (mA cm
SWFF-DSSC MWFF-DSSC(2) MWFF-DSSC(3) MWFF-DSSC(4) MWFF-DSSC(5) MWFF-DSSC(6) without Nb2O5 MWFF-DSSC(6) with Nb2O5
2
)
Table 2 Photovoltaic properties of individual working electrodes disconnected with each other in an MWFFDSSC(6). Electrodes
JSC (mA cm
F1 F2 S1 S2 B1 B2
4.2 4.1 3.6 3.4 1.6 1.5
2
)
VOC (V)
FF
η (%)
0.61 0.61 0.62 0.60 0.57 0.57
0.63 0.66 0.65 0.65 0.78 0.77
1.6 1.7 1.4 1.3 0.7 0.7
conduction band edge potential of Nb2O5 is about 100 mV more negative than that of TiO2, thus the conduction band edge of Nb2O5 locates between the conduction band edge of TiO2 and the excited states of the dye molecules. Therefore, the Nb2O5 layer served as an energy barrier to the electrons in the conduction band of TiO2. This energy barrier slowed the recombination at the TiO2/dye/electrolyte interface, rendering more electrons to accumulate in the conduction band of TiO2. Consequently, more photoexcited electrons could flow via the external circuit and a larger JSC resulted. Moreover, because the recombination at the TiO2/dye/electrolyte interface is the major loss mechanism of VOC, which is determined by the difference between the electron quasi-Fermi level in the TiO2 film under illumination and the redox electrolyte level, the existence of the Nb2O5 energy barrier also helped to maintain the VOC at a higher level [31–35]. The Nyquist plots of the MWFF-DSSC(6), either with or free from Nb2O5, are given in Fig. 5b. The semicircles are attributable to the electron recombination at the TiO2/dye/electrolyte interface. The recombination resistance in the MWFF-DSSC (6) with Nb2O5 was larger than that in the MWFF-DSSC (6) free from Nb2O5, indicating that the Nb2O5 barrier layer did weaken the electron recombination process [22,31]. For the purpose of testing the reproducibility of the above results, five more MWFF-DSSC(6) with Nb2O5 were fabricated. The photovoltaic performances of these cells were quite consistent with those given in Fig. 5a and Table 1 (Fig. S1 and Table S1, Supporting information).
505
A comparison can be made between the MWFF-DSSC(6) in this work and the conventional DSSCs with P25 films on FTO substrates used as their photoanodes. The effective areas and conversion efficiencies of the conventional DSSCs are usually between 0.1 cm2 to 0.3 cm2 and 5% to 12%, respectively (Table S2, Supporting information) [2,36–39]. In this study, the effective area of the MWFF-DSSC(6) was 1.7 0.08=0.14 cm2 and the conversion efficiency of the MWFF-DSSC(6) with Nb2O5 reached 9.1%, suggesting that the MWFF-DSSC(6) were competitive with the conventional DSSCs based on P25 films. (It might be necessary to emphasize that the 0.14 cm2 was the flank area of the whole cell instead of the area of the Ti microwire only and all the light illuminating this area was taken into consideration in computing the conversion efficiencies.) Furthermore, to the authors’ best knowledge, a conversion efficiency larger than 9% has not been achieved from any conventional DSSCs based on TiO2 nanotube array photoanodes (Table S2, Supporting information) [40–42]. In this sense, this study has shown that increasing the number of working electrodes is an effective strategy for improving the performances of the DSSCs. For the purpose of investigating how the length of an MWFF-DSSC(6) influenced the conversion efficiency, a 3.2 cm long MWFF-DSSC(6) with Nb2O5, which was almost twice as long as the 1.7 cm long cell described above, was fabricated and its photovoltaic properties were measured (Fig. S2, Supporting information). The area under illumination then became 3.2 0.08=0.26 cm2 and the conversion efficiency dropped to 8.1%. Because the space between the six working electrodes and one counter electrode was rather narrow, the longer the tube was, the more difficult it was to fully fill the space with the electrolyte owing to the viscous resistance to the electrolyte from the electrodes and the capillary tube. Nonetheless, a comparison shows that this drop was not very serious and a prolonged MWFF-DSSC(6) can still be considered as competitive with the conventional DSSCs. An obvious advantage of the MWFF-DSSC(6) against conversional DSSCs is that the MWFF-DSSC(6) can harvest light from all directions perpendicular to the fiber. That is, incident light from any direction perpendicular to the fiber can generate electricity. In this work, as indicated in Fig. 6a, light was allowed to illuminate the 1.7 cm long MWFF-DSSC (6) with Nb2O5 respectively from the three directions indicted by points A, B, and C. Three J–V curves were respectively acquired from the cell under illumination of the three incident directions and they are collectively shown in Fig. 6b. The corresponding photovoltaic performance parameters are listed in Table 3. The first three rows of Table 3 show that the performances of the MWFF-DSSC (6) were almost the same when the incident light illuminated the cell from different directions. In Fig. 6b, the three curves almost coincide with each other. Analogously, although the Nb2O5-free MWFF-DSSC(6) had lower conversion efficiencies than the MWFF-DSSC(6) with Nb2O5, the Nb2O5-free MWFFDSSC(6) also had the advantage that the light illuminating the cell from different directions led to quite similar outputs (Fig. S3 and Table S3, Supporting information). When fewer than six working electrodes were used in the Nb2O5-free MWFF-DSSCs, the capabilities of the cells harvesting light were obviously dependent on the incident direction of the light (Fig. S4, Supporting information). In the measurements of the photovoltaic properties under bending, the MWFF-DSSC(6) in this work appeared to be more
506
J. Liang et al.
Fig. 5 Photovoltaic properties of the MWFF-DSSC(6)s either with or free from Nb2O5. (a) J–V curves (The curve of the Nb2O5-free MWFF-DSSC(6) is similar to the one in Fig. 4a); (b) electrical impedance spectroscopy (EIS) results.
Fig. 6 Photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 under illumination from different directions. (a) The three vertical directions of the incident light (indicated by points A, B, and C, respectively); (b) the three J–V curves acquired with the cell respectively illuminated from the three vertical incident directions.
Table 3 Dependence of the photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 on the incident direction of the light and bending angle of the cell. Bending angles (1)
Point JSC (mA cm
0 0 0 30 90 180
A B C A A A
18.8 19.1 18.6 18.3 18.0 17.8
2
)
VOC (V)
FF
η (%)
0.66 0.65 0.66 0.68 0.65 0.66
0.73 0.72 0.74 0.71 0.72 0.72
9.1 9.0 9.1 8.8 8.4 8.5
flexible than the conventional flexible DSSCs. The effective length of the wire subjected to illumination was 1.7 cm. It is assumed that the bending was uniform along the whole MWFF-DSSC(6), i.e., the curvature radius was equal at every point. As shown in Fig. 7a, the bending angle is defined as the central angle subtended by the fiber under bending. Three J–V curves that were obtained from the MWFF-DSSC(6) with Nb2O5 under 301, 901 and 1801 bending angles, respectively,
are collectively shown in Fig. 7b. The corresponding photovoltaic parameters are listed in Table 3. As shown in the last three rows of Table 3, on the one hand, the conversion efficiency suffered a lost when the cell was under bending, probably due to the partial damage of the highly ordered TiO2 nanotube arrays; on the other hand, this lost was not very considerable. For instance, even under a 1801 bending, the cell retained about 93% of its conversion efficiency in comparison with the straight form (η=8.5% vs. η=9.1%). Accordingly, the three curves are hardly distinguishable in Fig. 7b. The behavior of the Nb2O5-free MWFF-DSSC(6) under bending was quite similar to that of the MWFF-DSSC(6) with Nb2O5 (Fig. S5 and Table S3, Supporting information). In summary, new-type FF-DSSCs with multi-working electrodes, namely MWFF-DSSC(x)s (x =2, 3, 4, 5, and 6), have been designed and realized. In each MWFF-DSSC(6), the working electrodes were six Ti microwires covered with highly ordered TiO2 nanotube arrays that surrounded a Pt microwire. Importantly, for a more reasonable comparison with the conventional DSSCs, the flank cross section of the whole cell, instead of that of only a Ti microwire, was used as the cross section area of the incident light in the computation of the conversion efficiency. Even when the TiO2 nanotube arrays on the working electrodes were not
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
507
Fig. 7 Photovoltaic performance of the 1.7 cm long MWFF-DSSC(6) with Nb2O5 under bending. (a) Schematic of the MWFF-DSSC (6) under different bending angles; (b) J–V curves.
covered with Nb2O5, the conversion efficiency of the MWFFDSSC(6) reached 6.6%, which was higher than those of the MWFF-DSSCs with two to five working electrodes and the SWFF-DSSC (using the computation method put forward in this article). The conversion efficiency was further raised to 9.1% by covering the TiO2 nanotube array surfaces with Nb2O5 layers. This value, within the authors’ knowledge, was the highest among all the reported FF-DSSCs and was also competitive with the conventional DSSCs. Furthermore, the MWFF-DSSC(6) could largely retain the good performances under bending. For example, the conversion efficiency of the MWFF-DSSC(6) with Nb2O5 only suffered a 6.6% decrease when the fiber was under a 1801 bending. Therefore, the feasibility of improving the performance of the FFDSSCs by employing more working electrodes has been proved. Following this strategy, a considerable expansion of the application scope of the FF-DSSCs can be expected.
Experimental methods Synthesis of photoanodes Highly ordered TiO2 nanotube arrays were fabricated in a modified anodization process [30]. The primary component of the electrolyte employed in anodizing a Ti microwire (0.25 mm in diameter, 99.7% in purity, Alfa aesar) was an ethylene glycol (C2H6O2) solution containing NH4F (0.25% in mass) and H2O (0.6% in volume). In each fabrication, the Ti microwire was cleaned ultrasonically in turn in acetone, de-ionized water and ethanol for 15 min each. Then, the Ti microwire was attached to the positive electrode of a DC power source. A piece of graphite was connected to the negative electrode and immersed in the electrolyte approximately 5 cm away from the Ti microwire. All the experiments were done at room temperature. For a better ordering and alignment of the TiO2 nanotube arrays, the Ti microwire was anodized twice. The first anodization step served to provide a regular pattern on the Ti surface, which worked as a starting template for the second anodization. In the first-step anodization, the Ti microwire was anodized under a 60 V voltage for 6 h. Then, the nanotube layer was removed ultrasonically in deionized water. In the second-step anodization, the Ti microwire, on whose surface a regular pattern remained, was again submitted to anodization at 60 V for 12 h. A highly ordered TiO2 nanotube array was obtained in this
second-step anodization. Finally, for crystallizing this TiO2 nanotube array, the sample was annealed in air at 500 1C for 30 min with a ramp rate of 2 1C min 1 and cooled to room temperature naturally. For the purpose of introducing a Nb2O5 layer on its surface, the Ti microwire covered with the TiO2 nanotube array was dipped into a 5 mM niobium isopropoxide solution at room temperature for 30 s. Then, the microwire was flushed with isopropanol, dried and finally annealed again under the same condition as described above.
Device fabrication The working electrodes, either with or without Nb2O5 layers, were immersed into 0.3 mM cis-diisothiocyanato-bis (2,20-bipyridyl-4,40-dicarboxylato) ruthenium(II) bis(tetrabutylammonium) (N719, Dalian HeptaChroma SolarTech, China) solution for 24 h after they cooled down from sintering at about 80 1C. Then they were rinsed with ethanol to remove physically adsorbed dye molecules. The counter electrode in each solar cell was a Pt wire (0.25 mm in diameter, 99.9% in purity, Alfa aesar). The FF-DSSCs were assembled in two ways, respectively. First, the method introduced in previous literatures was used [6,17–28]. One Ti microwire and one Pt microwire were inserted into a flexible plastic capillary tube, 0.8 mm in diameter, so that an FF-DSSC with only one working electrode, namely an SWFF-DSSC, was made. Its appearance is shown in Fig. 3a. Second, a number of Ti microwires and a Pt microwire surrounded by them were collectively inserted into a flexible plastic capillary tube, also 0.8 mm in diameter, so that an MWFF-DSSC(x) was made, where x = 2, 3, 4, 5, or 6 denotes the number of the working electrodes. The configuration and appearance of an MWFF-DSSC(6) are given in Figs. 1 and 3b, respectively. Then I /I3 based liquid electrolyte was injected into the capillary tube. At last, the two ends of the capillary tube were sealed with sealing glue to prevent evaporation of the electrolyte.
Characterization techniques The morphologies and microstructures of the samples were observed using a field-emission scanning electron microscope (FESEM). The chemical compositions and crystallinities of the samples were investigated using XPS and XRD, respectively. The
508
J. Liang et al.
photovoltaic performance of the DSSCs was measured under the illumination of an artificial sunlight with the standard air mass 1.5 (AM 1.5) spectrum and a 100 mW cm 2 irradiance (Oriel Solar Simulator, Model 91160). The current–voltage characteristics of the cells were recorded by an electrochemical analyzer. Following the practice of most researchers, the backward sweeping direction was followed in the measurements in this work. That is, the voltages on the two terminals of the cells were swept from 0.9 to 0 V. As suggested by Yang et al. [43], if the forward direction, i.e., from 0 to 0.9 V, had been followed, poorer performances would have been obtained (Fig. S2, Supporting information). The EIS was acquired using an electrochemical workstation with the frequency range from 0.1 Hz to 100 kHz in the dark condition. The effective areas of the SWFF-DSSC and the MWFF-DSSC(x)s (x=2, 3, 4, 5, and 6) were both 0.14 cm2, which was the product of the diameter of the capillary tube, 0.8 mm, and the effective length of the microwires subjected to illumination, 1.7 cm.
Acknowledgment This work was supported by the National Natural Science Foundation of China (nos. 61171023, 61306079 and 61376059).
Appendix A.
Supporting information
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2015.01.023.
References [1] B. Oregan, M. Gratzel, Nature 353 (1991) 737–740. [2] A. Yella, H.-W. Lee, H.N. Tsao, C. Yi, A.K. Chandiran, M.K. Nazeeruddin, E.W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Graetzel, Science 334 (2011) 629–634. [3] M. Gratzel, Nature 414 (2001) 338–344. [4] W.-Q. Wu, Y.-F. Xu, H.-S. Rao, C.-Y. Su, D.-B. Kuang, J. Am. Chem. Soc. 136 (2014) 6437–6445. [5] X.A. Jeanbourquin, X. Li, C. Law, P.R.F. Barnes, R. HumphryBaker, P. Lund, M.I. Asghar, B.C. O’Regan, J. Am. Chem. Soc. 136 (2014) 7286–7294. [6] W.X. Guo, C. Xu, G. Zhu, C.F. Pan, C.J. Lin, Z.L. Wang, Nano Energy 1 (2012) 176–182. [7] J. Liang, G. Zhang, H. Xia, W. Sun, RSC Adv. 4 (2014) 12649–12652. [8] B. Liu, E.S. Aydil, J. Am. Chem. Soc. 131 (2009) 3985–3990. [9] J.R. Jennings, A. Ghicov, L.M. Peter, P. Schmuki, A.B. Walker, J. Am. Chem. Soc. 130 (2008) 13364–13372. [10] J. Khamwannah, Y.Y. Zhang, S.Y. Noh, H. Kim, C. Frandsen, S.D. Kong, S. Jin, Nano Energy 1 (2012) 411–417. [11] L. Li, T.Y. Zhai, Y. Bando, D. Golberg, Nano Energy 1 (2012) 91–106. [12] Y. Peng, J. Zhong, K. Wang, B.F. Xue, Y.B. Cheng, Nano Energy 2 (2013) 235–240. [13] H.C. Weerasinghe, F.Z. Huang, Y.B. Cheng, Nano Energy 2 (2013) 174–189. [14] H. Sun, Z. Yang, X. Chen, L. Qiu, X. You, P. Chen, H. Peng, Angew. Chem. Int. Ed. 52 (2013) 8276–8280. [15] S. Ito, N.-L.C. Ha, G. Rothenberger, P. Liska, P. Comte, S.M. Zakeeruddin, P. Pechy, M.K. Nazeeruddin, M. Graetzel, Chem. Commun. (2006) 4004–4006.
[16] V. Galstyan, A. Vomiero, E. Comini, G. Faglia, G. Sberveglieri, RSC Adv. 1 (2011) 1038–1044. [17] X. Fan, Z. Chu, F. Wang, C. Zhang, L. Chen, Y. Tang, D. Zou, Adv. Mater. 20 (2008) 592–595. [18] Y.P. Fu, M. Peng, Z.B. Lv, X. Cai, S.C. Hou, H.W. Wu, X. Yu, H. Kafafy, D.C. Zou, Nano Energy 2 (2013) 537–544. [19] Z. Liu, M. Misra, ACS Nano 4 (2010) 2196–2200. [20] M.R. Lee, R.D. Eckert, K. Forberich, G. Dennler, C.J. Brabec, R.A. Gaudiana, Science 324 (2009) 232–235. [21] W. Wang, Q. Zhao, H. Li, H. Wu, D. Zou, D. Yu, Adv. Funct. Mater. 22 (2012) 2775–2782. [22] J. Liang, J. Yang, G. Zhang, W. Sun, Electrochem. Commun. 37 (2013) 80–83. [23] F.J. Cai, T. Chen, H.S. Peng, J. Mater. Chem. 22 (2012) 14856–14860. [24] S.W. Pan, Z.B. Yang, P.N. Chen, J. Deng, H.P. Li, H.S. Peng, Angew. Chem. Int. Ed. 53 (2014) 6110–6114. [25] Z.B. Yang, H. Sun, T. Chen, L.B. Qiu, Y.F. Luo, H.S. Peng, Angew. Chem. Int. Ed. 52 (2013) 7545–7548. [26] S.W. Pan, Z.B. Yang, H.P. Li, L.B. Qiu, H. Sun, H.S. Peng, J. Am. Chem. Soc. 135 (2013) 10622–10625. [27] X. Fang, Z.B. Yang, L.B. Qiu, H. Sun, S.W. Pan, J. Deng, Y.F. Luo, H.S. Peng, Adv. Mater. 26 (2014) 1694–1698. [28] Z.B. Yang, J. Deng, X.M. Sun, H.P. Li, H.S. Peng, Adv. Mater. 26 (2014) 2643–2647. [29] P. Ball, Nature (2001). http://dx.doi.org/10.1038/news010215-9. [30] S. Li, G. Zhang, D. Guo, L. Yu, W. Zhang, J. Phys. Chem. C 113 (2009) 12759–12765. [31] J. Liang, G. Zhang, W. Sun, RSC Adv. 4 (2014) 6746–6751. [32] K. Sayama, H. Sugihara, H. Arakawa, Chem. Mater. 10 (1998) 3825–3832. [33] S.G. Chen, S. Chappel, Y. Diamant, A. Zaban, Chem. Mater. 13 (2001) 4629–4634. [34] A. Zaban, S.G. Chen, S. Chappel, B.A. Gregg, Chem. Commun. (2000) 2231–2232. [35] N. Kopidakis, N.R. Neale, A.J. Frank, J. Phys. Chem. B 110 (2006) 12485–12489. [36] S. Li, L. Qiu, C. Shi, X. Chen, F. Yan, Adv. Mater. 26 (2014) 1266–1271. [37] J. Yang, P. Ganesan, J. Teuscher, T. Moehl, Y.J. Kim, C. Yi, P. Comte, K. Pei, T.W. Holcombe, M.K. Nazeeruddin, J. Hua, S.M. Zakeeruddin, H. Tian, M. Graetzel, J. Am. Chem. Soc. 136 (2014) 5722–5730. [38] J. Luo, M. Xu, R. Li, K.-W. Huang, C. Jiang, Q. Qi, W. Zeng, J. Zhang, C. Chi, P. Wang, J. Wu, J. Am. Chem. Soc. 136 (2014) 265–272. [39] J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T. Bessho, A. Marchioro, E. Ghadiri, J.-E. Moser, C. Yi, M.K. Nazeeruddin, M. Graetzel, Nat. Commun. 3 (2012). [40] F. Zhuge, J. Qiu, X. Li, X. Gao, X. Gan, W. Yu, Adv. Mater. 23 (2011) 1330–1334. [41] M. Ye, D. Zheng, M. Lv, C. Chen, C. Lin, Z. Lin, Adv. Mater. 25 (2013) 3039–3044. [42] M. Ye, X. Xin, C. Lin, Z. Lin, Nano Lett. 11 (2011) 3214–3220. [43] X.D. Yang, M. Yanagida, L.Y. Han, Energy Environ. Sci. 6 (2013) 54–56. Jia Liang received his B.S. degree in 2010 from Southeast University, Nanjing, China. He is a Ph.D. candidate of Key Laboratory for Physics and Chemistry of Nanodevices in the Department of Electronics, School of Electrical Engineering and Computer Science at Peking University. He joined Dr. Jun Lou’s research group in Rice University as a visiting student in the first half of 2014. His main research interest is synthesis of energy nanomaterials and their applications in solar cells and catalyses.
High efficiency flexible fiber-type dye-sensitized solar cells with multi-working electrodes
Gengmin Zhang received his B.Sc. and Ph.D. in Science degrees both from Peking University, China, in 1991 and 1996, respectively. His thesis for the Ph.D. degree was on the photoemission from the oxide-coated cathodes under laser illumination. Supported by a Sino-Japanese intergovernmental project, he studied the behaviors of organic molecules on the graphite surface in Osaka University, Japan, from 1998 to 2000. His current research is focused on the photovoltaic, ferroelectric and field emission properties of nanomaterials. Dr. Wentao Sun received her B.S. and Ph.D. degrees from Central South University of Technology and Technical Institute of Physics and Chemistry, CAS in 1999 and 2005, respectively. Then she was a Postdoctoral Research Fellow in Peking University from 2005 to 2008. In 2008, she joined Department of Electronics of Peking University as an Assistant Professor. In 2011, she was promoted as an Associate Professor. Dr. Sun’s current research interests mainly focus on nanomaterial synthesis and nanoenergy devices.
509
Dr. Pei Dong received her Ph.D. in Mechanical Engineering and Materials Science from Rice University in 2013, under the guidance of Prof. Jun Lou. She is currently a joint Postdoctoral Research Fellow in Prof. Pulickel M. Ajayan’s and Prof. Jun Lou’s groups in Department of Materials Science and NanoEngineering at Rice University. Her research mainly focuses on nanomaterials enabled dye-sensitized solar cells and understanding self-stiffening mechanisms in nanocomposites.