Home
Search
Collections
Journals
About
Contact us
My IOPscience
Incorporation of carbon nanotubes in a hierarchical porous photoanode of tandem quantum dot sensitized solar cells
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 345402 (http://iopscience.iop.org/0957-4484/25/34/345402) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 129.108.9.184 This content was downloaded on 14/08/2014 at 15:21
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 345402 (10pp)
doi:10.1088/0957-4484/25/34/345402
Incorporation of carbon nanotubes in a hierarchical porous photoanode of tandem quantum dot sensitized solar cells Mohammad Reza Golobostanfard1, Hossein Abdizadeh1,2 and Shamsoddin Mohajerzadeh3 1
School of Metallurgy and Materials Engineering, College of Engineering, University of Tehran, PO Box 14395-553, Tehran, Iran 2 Center of Excellence for High Performance Materials, University of Tehran, Tehran, Iran 3 School of Electrical and Computer Engineering, College of Engineering, University of Tehran, PO Box 14395-553, Tehran, Iran E-mail: [email protected] Received 14 March 2014, revised 29 May 2014 Accepted for publication 4 July 2014 Published 7 August 2014 Abstract
The incorporation of multi-walled carbon nanotubes (MWCNT) in quantum dot (QD) sensitized solar cells (QDSC) based on CdSe QDs and quantum rods (QRs) is investigated. The composite hierarchical porous photoanode of titania/CNT is synthesized by sol–gel induced phase separation and QDs/QRs are prepared by the modified solvothermal method. The QDs and QRs form a tandem structure on the hierarchical porous photoanode after deposition by the electrophoretic method. Incorporation of MWCNT in the QDSC photoanode in optimum content (0.32 wt%) causes appreciable enhancement in cells efficiency (about 41% increase). This improvement in efficiency mainly emerges from the beneficial role of MWCNTs in charge injection and collection. The MWCNTs result in longer electron lifetime and higher electron diffusion length, which is confirmed by electrochemical impedance spectroscopy. Keywords: quantum dot sensitized solar cells, quantum rod, tandem structure, solvothermal method, hierarchical porous photoanode (Some figures may appear in colour only in the online journal) 1. Introduction
DSSCs. The major obstacles for increasing the efficiency in these devices include (1) photoanode pore size, which should be large enough for diffusion of larger size QDs compared with molecular dye [11], (2) back reaction recombination at the TiO2/QD/electrolyte interface due to low surface coverage of QDs and short electron lifetime in TiO2 due to the relatively long regeneration time compared to injection time [12], (3) finding suitable electrolyte with high stability and high QD regeneration efficiency without affecting the open-circuit voltage [13], (4) proper counter electrode commensurate to the electrolyte [14], (5) improving TiO2/QD attachment and interface [15], and (6) short exciton lifetime in QDs during diffusion to the interface of TiO2/QD [16]. The hierarchical porous photoanodes with large macrochannels guarantee the diffusion of QDs into inner parts, and hence large cross sectional sensitization could be achieved.
Dye sensitized solar cells (DSSC)s, quantum dot sensitized solar cells (QDSC)s, and organic bulk heterojunction solar cells are three major promising photovoltaic technologies, which offer adequate efficiency besides cost effectiveness [1–3]. Among them, QDSCs with the advantages of tunable band gap, high extinction coefficient, large molecular dipole, sensitization to diffuse light, ability to fabricate tandem structure, generation of multiple excitons, and utilization of hot electrons, have been subjected to large efforts for introducing next generation photovoltaic cells [4–8]. Although the efficiency of more than 5% has been reported for QDSCs [9, 10], a greater improvement in the performance of these devices is still required to make them competitive with their more developed counterparts, i.e. 0957-4484/14/345402+10$33.00
1
© 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 345402
M R Golobostanfard et al
On the other hand, the mesopores in these structures provide large surface area required for high sensitization [17, 18]. The back reaction recombination and short electron lifetime in these devices were improved by applying CdS buffer and ZnS blocking layers [19, 20]. Moreover, the polysulfide and corresponding Cu2S counter electrode were proved to be the suitable choice for overcoming the electrolyte and counter electrode problems in QDSCs due to their high stability and diffusibility as well as compatibility with sulfide and selenosulfide species [21, 22]. Thus, the main bottleneck that should be carefully addressed is the short electron and exciton lifetimes in these devices. Moreover, extending the vis-NIR absorption range is a very crucial method for improving the efficiency, which could be obtained by the tandem structure of different semiconductors (AlInP/GaInP/AlGaAs/GaAs) [23], different alloyed QDs (CdSe1 − xSx) [24], or different dyes [25]. Carbon nanotubes (CNTs) with outstanding electrical and chemical properties as well as relatively appropriate thermal property can be a promising candidate in charge separation and charge transfer in optoelectronic devices [26–28]. CNTs have been utilized in QDSCs as a counter electrode [29, 30], an acceptor in direct contact with QDs [31, 32], an assisting layer for charge separation between the transparent conductive oxide and TiO2 layer [33, 34], and a charge transport route in the photoanode [35]. The cells performance was improved in all CNT incorporated QDSCs. However, incorporation of CNTs in a photoanode of QDSCs with a proper interface has never been reported. The incorporation of CNTs in a photoanode with a suitable interface is preferred due to its beneficial role in charge injection from TiO2 to CNT, reducing the deficient trap states on CNT sidewalls, and increasing the ballistic charge transport. In this research, the incorporation of CNTs in a hierarchical meso/macroporous photoanode of tandem structured QDSC with selective positioning of QDs and quantum rods (QRs) is investigated. The electrophoretic deposition causes QDs with larger band gaps to be placed on inner layers and QRs with smaller band gaps on outer layers, which form a tandem structure in QDSC. Tandem structured QDSC based on various CdSe shapes could be advantageous in the case of efficiency enhancement. Sol–gel induced phase separation synthesized hierarchical porous photoanodes were utilized in all cells with low organic residues and ability to generate inter-connected tunable pore sizes. QDs and QRs were synthesized through the solvothermal method which offers the advantages of more diversity on shape and low levels of organic ligands.
(H2SO4), cadmium oxide powder (CdO), cadmium nitrate (Cd(NO3)2.4H2O), selenium chloride (SeCl4), triethylene tetramine (TETA), sulfur powder (S), sodium sulfide (Na2S), sodium hydroxide (NaOH), zinc acetate dehydrate (ZAD), and copper chloride (CuCl2) were all in reagent grade and purchased from Merck. Deionized water (DIW, 18.2 MΩ) was used in all experiments. The fluorine doped tin oxide conductive glass substrates (FTO, 15 Ω per square) and Surlynionomer were purchased from Dyesol. Multi-walled carbon nanotubes (MWCNT, outer diameter 40–60 nm, 97% purity, 130–160 m2 g−1) were provided from Shenzen Nanotech port. All chemicals were used as received without further purification. 2.2. CdSe quantum dot and rod synthesis
CdSe QDs and QRs were synthesized by a modified homemade solvothermal reactor with the stirring system and controllable internal pressure, as reported elsewhere [36]. Typically, 1 mmol of CdO powder and 1 mmol of SeCl4 were dissolved in 20 mL TETA in the solvothermal reactor. The reactants were heated to 220 °C for 24 h under continuous stirring and external pressure of 400 kPa. The synthesis time of 2 h at 220 °C was also considered for comparison. Then, the precipitates were separated, centrifuged, and redispersed in ethanol. The centrifugation and dispersion processes were repeated several times to ensure removal of organic residues. Finally, stepwise size selective centrifugation was applied to obtain CdSe QDs. 2.3. CNT functionalization
CNT functionalization was performed according to the previous research [37]. Typically, 0.5 g of MWCNT was sonicated in concentrated HNO3: H2SO4 (3 : 1 vol%) for 10 min. Then, the mixture was refluxed at 80 °C for 1 h. After dilution, the mixture was washed with DIW and centrifuged several times to ensure removal of anions. The product was dried at 60 °C and kept dried until use. 2.4. Photoanode preparation
2. Material and methods
The FTO substrates were cleaned ultrasonically in detergent solution, DIW, and EtOH for 10 min. Sol preparation and deposition of blocking layer were discussed elsewhere [38]. Then, the film was calcined at 450 °C for 1 h. This layer acts as a blocking layer in the device. The preparation method for the hierarchical porous photoanode has also been reported elsewhere [39]. The final composition of the sol was TBT: DEA:DIW = 1:1:1 with 0.75 mol L−1 concentration and 0.32 wt% CNT. CNT contents of 0.16 and 0.64 wt% were also considered.
2.1. Material
2.5. QDSC fabrication
Ethanol (EtOH), 1-propanol (1PrOH), tetrapropyl ortotitanate (TTiP), tetrabutyl ortotitanate (TBT), diethanol amine (DEA), polyethylene glycol (PEG, average molecular weight 1000), nitric acid (HNO3), hydrochloric acid (HCl), sulfuric acid
A very thin CdS layer was deposited on the porous photoanode by SILAR. The 0.05 molar Cd(NO3)2 in EtOH and 0.05 molar Na2S in DIW and MeOH were used as the cationic and anionic solutions, respectively. Each SILAR cycle 2
Nanotechnology 25 (2014) 345402
M R Golobostanfard et al
and FTO/MWCNT/Cu2S cathode and sealed by Surlynionomer. The electrolyte was injected into the cells by evacuation through a predrilled hole. Then, the hole was sealed by Surlyn and microscope glass slide. The active area of the TiO2 photoanode was 0.16 cm2. 2.6. Characterization
X-ray diffraction (XRD), Philips X-pert pro PW1730, Cu-kα, was utilized to study the phase structure of the QDs and photoanodes. High resolution transmission electron microscope (HRTEM), Philips CM30, 300 kV, equipped with a selected area electron diffraction pattern (SAED) was employed to study the QD and CNT morphologies. Field emission scanning electron microscopy (FESEM), Hitachi S4160, was used to investigate the morphology of porous films. Atomic force microscopy (AFM), Dual Scope DS 95–200/50, was performed in contact mode to achieve surface roughness of the films. The infrared spectra were recorded using a Fourier-transformed infrared (FTIR) spectrophotometer, Bruker TENSOR27, in transmittance mode at 400–4000 cm−1 with KBr as blank. Raman spectra were recorded using a BRUKER (SETERRA, spectral resolution
Search
Collections
Journals
About
Contact us
My IOPscience
Incorporation of carbon nanotubes in a hierarchical porous photoanode of tandem quantum dot sensitized solar cells
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 345402 (http://iopscience.iop.org/0957-4484/25/34/345402) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 129.108.9.184 This content was downloaded on 14/08/2014 at 15:21
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 345402 (10pp)
doi:10.1088/0957-4484/25/34/345402
Incorporation of carbon nanotubes in a hierarchical porous photoanode of tandem quantum dot sensitized solar cells Mohammad Reza Golobostanfard1, Hossein Abdizadeh1,2 and Shamsoddin Mohajerzadeh3 1
School of Metallurgy and Materials Engineering, College of Engineering, University of Tehran, PO Box 14395-553, Tehran, Iran 2 Center of Excellence for High Performance Materials, University of Tehran, Tehran, Iran 3 School of Electrical and Computer Engineering, College of Engineering, University of Tehran, PO Box 14395-553, Tehran, Iran E-mail: [email protected] Received 14 March 2014, revised 29 May 2014 Accepted for publication 4 July 2014 Published 7 August 2014 Abstract
The incorporation of multi-walled carbon nanotubes (MWCNT) in quantum dot (QD) sensitized solar cells (QDSC) based on CdSe QDs and quantum rods (QRs) is investigated. The composite hierarchical porous photoanode of titania/CNT is synthesized by sol–gel induced phase separation and QDs/QRs are prepared by the modified solvothermal method. The QDs and QRs form a tandem structure on the hierarchical porous photoanode after deposition by the electrophoretic method. Incorporation of MWCNT in the QDSC photoanode in optimum content (0.32 wt%) causes appreciable enhancement in cells efficiency (about 41% increase). This improvement in efficiency mainly emerges from the beneficial role of MWCNTs in charge injection and collection. The MWCNTs result in longer electron lifetime and higher electron diffusion length, which is confirmed by electrochemical impedance spectroscopy. Keywords: quantum dot sensitized solar cells, quantum rod, tandem structure, solvothermal method, hierarchical porous photoanode (Some figures may appear in colour only in the online journal) 1. Introduction
DSSCs. The major obstacles for increasing the efficiency in these devices include (1) photoanode pore size, which should be large enough for diffusion of larger size QDs compared with molecular dye [11], (2) back reaction recombination at the TiO2/QD/electrolyte interface due to low surface coverage of QDs and short electron lifetime in TiO2 due to the relatively long regeneration time compared to injection time [12], (3) finding suitable electrolyte with high stability and high QD regeneration efficiency without affecting the open-circuit voltage [13], (4) proper counter electrode commensurate to the electrolyte [14], (5) improving TiO2/QD attachment and interface [15], and (6) short exciton lifetime in QDs during diffusion to the interface of TiO2/QD [16]. The hierarchical porous photoanodes with large macrochannels guarantee the diffusion of QDs into inner parts, and hence large cross sectional sensitization could be achieved.
Dye sensitized solar cells (DSSC)s, quantum dot sensitized solar cells (QDSC)s, and organic bulk heterojunction solar cells are three major promising photovoltaic technologies, which offer adequate efficiency besides cost effectiveness [1–3]. Among them, QDSCs with the advantages of tunable band gap, high extinction coefficient, large molecular dipole, sensitization to diffuse light, ability to fabricate tandem structure, generation of multiple excitons, and utilization of hot electrons, have been subjected to large efforts for introducing next generation photovoltaic cells [4–8]. Although the efficiency of more than 5% has been reported for QDSCs [9, 10], a greater improvement in the performance of these devices is still required to make them competitive with their more developed counterparts, i.e. 0957-4484/14/345402+10$33.00
1
© 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 345402
M R Golobostanfard et al
On the other hand, the mesopores in these structures provide large surface area required for high sensitization [17, 18]. The back reaction recombination and short electron lifetime in these devices were improved by applying CdS buffer and ZnS blocking layers [19, 20]. Moreover, the polysulfide and corresponding Cu2S counter electrode were proved to be the suitable choice for overcoming the electrolyte and counter electrode problems in QDSCs due to their high stability and diffusibility as well as compatibility with sulfide and selenosulfide species [21, 22]. Thus, the main bottleneck that should be carefully addressed is the short electron and exciton lifetimes in these devices. Moreover, extending the vis-NIR absorption range is a very crucial method for improving the efficiency, which could be obtained by the tandem structure of different semiconductors (AlInP/GaInP/AlGaAs/GaAs) [23], different alloyed QDs (CdSe1 − xSx) [24], or different dyes [25]. Carbon nanotubes (CNTs) with outstanding electrical and chemical properties as well as relatively appropriate thermal property can be a promising candidate in charge separation and charge transfer in optoelectronic devices [26–28]. CNTs have been utilized in QDSCs as a counter electrode [29, 30], an acceptor in direct contact with QDs [31, 32], an assisting layer for charge separation between the transparent conductive oxide and TiO2 layer [33, 34], and a charge transport route in the photoanode [35]. The cells performance was improved in all CNT incorporated QDSCs. However, incorporation of CNTs in a photoanode of QDSCs with a proper interface has never been reported. The incorporation of CNTs in a photoanode with a suitable interface is preferred due to its beneficial role in charge injection from TiO2 to CNT, reducing the deficient trap states on CNT sidewalls, and increasing the ballistic charge transport. In this research, the incorporation of CNTs in a hierarchical meso/macroporous photoanode of tandem structured QDSC with selective positioning of QDs and quantum rods (QRs) is investigated. The electrophoretic deposition causes QDs with larger band gaps to be placed on inner layers and QRs with smaller band gaps on outer layers, which form a tandem structure in QDSC. Tandem structured QDSC based on various CdSe shapes could be advantageous in the case of efficiency enhancement. Sol–gel induced phase separation synthesized hierarchical porous photoanodes were utilized in all cells with low organic residues and ability to generate inter-connected tunable pore sizes. QDs and QRs were synthesized through the solvothermal method which offers the advantages of more diversity on shape and low levels of organic ligands.
(H2SO4), cadmium oxide powder (CdO), cadmium nitrate (Cd(NO3)2.4H2O), selenium chloride (SeCl4), triethylene tetramine (TETA), sulfur powder (S), sodium sulfide (Na2S), sodium hydroxide (NaOH), zinc acetate dehydrate (ZAD), and copper chloride (CuCl2) were all in reagent grade and purchased from Merck. Deionized water (DIW, 18.2 MΩ) was used in all experiments. The fluorine doped tin oxide conductive glass substrates (FTO, 15 Ω per square) and Surlynionomer were purchased from Dyesol. Multi-walled carbon nanotubes (MWCNT, outer diameter 40–60 nm, 97% purity, 130–160 m2 g−1) were provided from Shenzen Nanotech port. All chemicals were used as received without further purification. 2.2. CdSe quantum dot and rod synthesis
CdSe QDs and QRs were synthesized by a modified homemade solvothermal reactor with the stirring system and controllable internal pressure, as reported elsewhere [36]. Typically, 1 mmol of CdO powder and 1 mmol of SeCl4 were dissolved in 20 mL TETA in the solvothermal reactor. The reactants were heated to 220 °C for 24 h under continuous stirring and external pressure of 400 kPa. The synthesis time of 2 h at 220 °C was also considered for comparison. Then, the precipitates were separated, centrifuged, and redispersed in ethanol. The centrifugation and dispersion processes were repeated several times to ensure removal of organic residues. Finally, stepwise size selective centrifugation was applied to obtain CdSe QDs. 2.3. CNT functionalization
CNT functionalization was performed according to the previous research [37]. Typically, 0.5 g of MWCNT was sonicated in concentrated HNO3: H2SO4 (3 : 1 vol%) for 10 min. Then, the mixture was refluxed at 80 °C for 1 h. After dilution, the mixture was washed with DIW and centrifuged several times to ensure removal of anions. The product was dried at 60 °C and kept dried until use. 2.4. Photoanode preparation
2. Material and methods
The FTO substrates were cleaned ultrasonically in detergent solution, DIW, and EtOH for 10 min. Sol preparation and deposition of blocking layer were discussed elsewhere [38]. Then, the film was calcined at 450 °C for 1 h. This layer acts as a blocking layer in the device. The preparation method for the hierarchical porous photoanode has also been reported elsewhere [39]. The final composition of the sol was TBT: DEA:DIW = 1:1:1 with 0.75 mol L−1 concentration and 0.32 wt% CNT. CNT contents of 0.16 and 0.64 wt% were also considered.
2.1. Material
2.5. QDSC fabrication
Ethanol (EtOH), 1-propanol (1PrOH), tetrapropyl ortotitanate (TTiP), tetrabutyl ortotitanate (TBT), diethanol amine (DEA), polyethylene glycol (PEG, average molecular weight 1000), nitric acid (HNO3), hydrochloric acid (HCl), sulfuric acid
A very thin CdS layer was deposited on the porous photoanode by SILAR. The 0.05 molar Cd(NO3)2 in EtOH and 0.05 molar Na2S in DIW and MeOH were used as the cationic and anionic solutions, respectively. Each SILAR cycle 2
Nanotechnology 25 (2014) 345402
M R Golobostanfard et al
and FTO/MWCNT/Cu2S cathode and sealed by Surlynionomer. The electrolyte was injected into the cells by evacuation through a predrilled hole. Then, the hole was sealed by Surlyn and microscope glass slide. The active area of the TiO2 photoanode was 0.16 cm2. 2.6. Characterization
X-ray diffraction (XRD), Philips X-pert pro PW1730, Cu-kα, was utilized to study the phase structure of the QDs and photoanodes. High resolution transmission electron microscope (HRTEM), Philips CM30, 300 kV, equipped with a selected area electron diffraction pattern (SAED) was employed to study the QD and CNT morphologies. Field emission scanning electron microscopy (FESEM), Hitachi S4160, was used to investigate the morphology of porous films. Atomic force microscopy (AFM), Dual Scope DS 95–200/50, was performed in contact mode to achieve surface roughness of the films. The infrared spectra were recorded using a Fourier-transformed infrared (FTIR) spectrophotometer, Bruker TENSOR27, in transmittance mode at 400–4000 cm−1 with KBr as blank. Raman spectra were recorded using a BRUKER (SETERRA, spectral resolution
