Step-by-step build-up of ordered p–n heterojunctions at nanoscale for efficient light harvesting

Abstract

A simple “grafting from” strategy is reported to graft polymer brushes bearing thiophene groups onto TiO₂ nanowires via a surface initiated polymerization and FeCl₃ catalytic oxidized polymerization. Subsequently, the in situ CdS nanocrystals can be fabricated along the conductive polymer network, enabling highly ordered p–n heterojunctions to be prepared on the nanoscale. Polynorbornene bearing thiophene was grafted onto TiO₂ nanowires via surface-initiated ring-opening metathesis polymerization (SI-ROMP), and then the oxidization polymerization of the thiophene derivative was performed to prepare conductive conjugated polythiophene brushes. The carboxyl groups in conductive polymer brushes can be used as the source precursor to coordinate cadmium cations for templating CdS nanocrystal formation, meanwhile creating intimate interfacial contact between nanocrystals and conjugated polymers. This novel approach offered a superior platform to the step-by-step build-up of ordered heterojunctions on the nanometer scale and has great advantages in precisely controlling the detailed nano-morphology and dispersion of nanocrystals within conductive polymers.

---

1. Introduction

Recent advances in the synthesis and application of conductive polymers (CPs) have stimulated increasing interest in exploiting CPs for the development of organic/inorganic photovoltaic platforms.^1–9 The excellent properties of CPs such as their light weight, flexibility,¹⁰ excellent solution processability,¹¹ tailorable electrochemical properties and high charge-carrier mobility have greatly expanded their usage in diverse areas including photovoltaics, light-emitting diodes (LEDs) and thin film transistors.¹² Among these CPs, poly(3-hexylthiophene) (P3HT) is regarded as one of the most widely studied organic semiconductors for solar cells.^13,14 Other than dye-sensitized solar cells and silicon solar cells,^15–18 the emergence of semiconductor nanocrystals (NCs) with multiple exciton generation capability as the building blocks has afforded promising opportunities to deliver efficient solar energy conversion in next generation photovoltaic cells.¹⁹ Their quantum-confined nature allows us to manipulate the energy band gap of these NCs via size and shape control, in order to tune the photo response.²⁰

In order to combine the good mechanical flexibility and solution processability of CPs and the excellent photoactivity and high conductivity of NCs, organic/inorganic hybrid solar cells have gained widespread attention in recent years.^21–25 Usually, CP/NC nanocomposites are prepared by simply physically mixing these two components, which are mainly in the interconnected phase-separated structured domains over the sub-20 nm length-scale.^2,22,26 As such, it still remains a challenge to avoid severe nanoscale phase separation, which can lower the interfacial contact area and diminish the device performance, probably due to the interrupted charge transfer.²⁴ In practice, the exciton formed after light is harvested and the separated charges could only travel less than 20 nm before recombination in organic–inorganic hybrid solar cells.²⁷ An ideal configuration is the highly desirable bicontinuous and nanoscopic phase-separated mixture of CPs/NCs, limited to 20 nm, with strong electronic coupling between these two counterparts. Highly ordered orientation is key in achieving efficient charge separation and transport, but it is currently difficult to realize via the conventional blending approach due to the nature of the poorly defined interface. Thus, to solve microscopic phase separation problems is crucial in realizing efficient charge transfer and hence high photo conversion.²⁸ Several approaches have been proposed such as co-solvent mixtures and surface modification of QDs; the most promising approach is chemical bond linking between QDs and polymers.²⁹

Lin et al. have demonstrated P3HT linking with CdSe nanorods via the Heck coupling reaction for improved interfacial contact.³⁰ Besides CdSe nanocrystals, CdS also can be used in P3HT-based hybrid solar cells.^31–33 High density P3HT grafting silica via surface-initiated Kumada catalyst-transfer polycondensation,³⁴ and chemical bonded p–n heterojunctions by surface initiated electrodeposition of p-type conducting polymer inside TiO₂ nanotubes, both aimed at the improvement of interfacial adhesion.³⁵ In this paper, we report a novel approach to fabrication of ordered p–n heterojunctions on the nanoscale by combining surface initiated polymerization^36,37 with polymer brush templated nanocrystal fabrication to realize intimate contact nano-architecture, with the aim of efficiently promoting the charge transfer and light harvesting efficiency.

2. Experimental

TiO₂ nanowires (100 mg) were immersed in 1.5 mg mL⁻¹ solution (ultrapure water : ethanol = 4 : 1) of initiator and stirred for 18 h in the dark at room temperature. The modified TiO₂ nanowires were isolated and purified by repeated washing with ultrapure water and ethanol using centrifugation. Then the samples were dried under vacuum overnight at 40 °C. Firstly, the monomer of 5-norbornene-2-methanol-3-thiopheneacetic acid ester (NMTE) was polymerized on biomimetic initiator anchored TiO₂ nanowires via ring-opening metathesis polymerization (ROMP). The oxidation polymerization can be initiated from the active sites in each of the thiophene units of polymer chains via FeCl₃ catalysis to polymerize 3-hexylthiophene (and 3-thiopheneacetic acid, denoted as 3THA) monomers on the main polymer chain.³⁸ Furthermore, the carboxylic acid groups of functionalized 3THA can coordinate Cd²⁺ ions as a resource for fabrication of narrow size distributed CdS nanocrystals.³⁹

3. Results and discussion

Scheme 1 shows the fabrication process of conductive polymer modified TiO₂ nanowires and the nanocrystal/polymer hybrid materials. Firstly, we designed and prepared a novel thiophene derivative containing norbornene groups. ROMP can easily be performed on the TiO₂ surface via a catecholic initiator; then the oligo-polymers provide an active site to initiate the oxidation polymerization, to produce a branch-like shape on the main chains. At first, the universal poly(3-hexylthiophene) was grafted on TiO₂ nanowires, which yielded P3HT brushes of 55 nm via FeCl₃ catalytic polymerization, as illustrated in Fig. 1A. From Fig. 1B (a), TiO₂ nanowires have almost no photo response in the visible light region. The absorption of the pristine P3HT appears at about 450 nm. When P3HT was grafted onto TiO₂ nanowires, the absorption broadened into the visible region, and the peak was slightly blue-shifted to 436 nm, as shown in Fig. 1B (c), which is possibly due to the more coil-like P3HT polymer chains in the chemically linked environment, resulting in decreasing conjugation length of the repeating units on the chain.³⁸

Fig. 1 TEM images of (A) P3TH-TiO₂ nanowires, (B) UV-Vis spectra of (a) TiO₂ nanowires, (b) P3HT, (c) P3HT-TiO₂ nanowires.

Scheme 1 (A) Fabrication process of P3HT, poly(3-thiopheneacetic acid) modified TiO₂ nanowires and the CdS nanocrystal/conductive polymer hybrid nanocomposite materials. (B) Preparation of P3HT and poly(3-thiopheneacetic acid) modified TiO₂ nanowires.

In order to construct more complex ordered organic/inorganic p–n heterojunctions for expanding solar spectral coverage, a carboxyl group was introduced into the thiophene derivatives to replace the hexyl substituent, which possesses the cation exchange ability to immobilize Cd²⁺ ions onto polymer chains as a resource for CdS NC fabrication. The nano-confined environments produced by conductive polymer brushes can be used to fabricate well monodispersed CdS NCs. The highly ordered p–n heterojunction formed by the regular arrangement of CdS NCs and conducting polymers on the TiO₂ surface can realize intimate contact for facilitating charge separation and collection. The presence of the polymer layer and CdS NCs was first verified by TEM images. Fig. 2A shows the plain TiO₂ nanowires with relatively clear crystalline orientation. When poly(NMTE) was grafted onto TiO₂ nanowires via ROMP, the TiO₂ nanowires were homogeneously coated with a very uniform layer (about 15 nm) and the interface boundary is very clear, having a lighter color than the side wall of the TiO₂ nanowires in Fig. 2B. Fig. 2C shows that P3HTA brushes can reach approximately 80 nm via oxidation polymerization. Fig. 2D shows the CdS NCs formed on the polymer layer and TiO₂ nanowires; the color of the polymer layer is weaker than that of the TiO₂ nanowires and the CdS nanoparticles with a uniform size are well monodispersed on the polymer layer. In order to clearly observe the distribution of CdS NCs in the polymer layer, the STEM images are provided in Fig. 2E and we can clearly see the CdS nanoparticles with narrow monodispersity distributed in the polymer layer and adhered individually on the TiO₂ nanowires. The polymer/TiO₂ interface is clearly seen: the darker area on both sides is the polymer layer, and between the polymer layer is the TiO₂ nanowires, showing a three-dimensional effect. The presence of the polymer matrix stabilizes newly formed nanoparticles and prevents their further growth, resulting in narrowly distributed nanocrystals.⁴⁰ The nanoparticle sizes were measured at about 4–7 nm with good dispersivity, as shown in Fig. 2F.

Fig. 2 TEM images of (A) TiO₂ nanowires, (B) poly(NMTE)-TiO₂ nanowires, (C) P3HTA-TiO₂ nanowires, (D) CdS-P3HTA-TiO₂ nanowires, (E) STEM images of CdS-P3HTA-TiO₂ nanowires. (F) Histogram of size distribution of CdS nanoparticles.

Furthermore, the elemental composition of the CdS NC/polymer hybrid was investigated by EDX mapping images. Fig. 3 shows scanning TEM (STEM) images and several elemental mapping images; the TiO₂ nanowires and polymer layers possess obvious contrast and a clear interface from the STEM measurements. These elemental mapping images conform to the STEM image. The C, S, Cd element exists in both the TiO₂ nanowires and polymer layers with uniform distribution. Their mapping images show that the elemental distribution is relatively homogeneous. Because the content of the O species in the polymer layer is far less than in the TiO₂ nanowires, the O mapping image shows a strong O signal in the TiO₂ region; but the polymer layer on both sides of the TiO₂ nanowires shows a low O content, indicated by the light-colored mapping images, which is consistent with the above analysis. The width of the Ti mapping image is smaller compared to that of others, which also provides direct proof that the polymer layers are wrapped around the TiO₂ nanowires.

Fig. 3 STEM images of CdS nanocrystals-P3HTA-TiO₂ nanowires and the corresponding EDX mapping images of the C, O, Ti, S and Cd, respectively.

The ordered p–n heterojunction could easily be prepared on perpendicularly oriented TiO₂ nanotubes by the same procedure. The apparent nanoporous morphology of TiO₂ nanotubes is clear from the FESEM measurement, as shown in Fig. 4A. The pore size is about 100 nm and the tube wall exhibits a smooth topography, as shown in Fig. 4D. When P3THA brushes were prepared on TiO₂ nanotubes, the tube end was almost covered by precipitate, which blocked the tube, as represented in Fig. 4B. From the cross sectional view in Fig. 4E, the polymer layers are covering the tube walls, which makes the tube profile unclear. After the in situ loading of CdS nanocrystals in Fig. 4C, small nanoparticles appeared on the TiO₂ nanotube surface. As shown in the cross-section in Fig. 4F, the CdS nanocrystals with uniform size cover the nanotubes walls.

Fig. 4 FESEM images of (A) TiO₂ nanotubes, (B) poly(3THA)-TiO₂ nanotubes, (C) CdS-Poly (3THA)-TiO₂ nanotubes. Cross-section FESEM views of (D) TiO₂ nanotubes, (E) poly(3THA)-TiO₂ nanotubes, (F) CdS-poly(3THA)-TiO₂ nanotubes.

Successful polymer grafting and CdS nanocrystal uploading on TiO₂ nanowires were further ascertained by XPS. Fig. 5A displays the XPS survey spectra of TiO₂ nanowires, which are mainly composed of Ti and O, as well as a small amount of C contamination from adventitious hydrocarbon. Successful anchoring of the initiator on TiO₂ nanowires is indicated by the presence of N signals, which are not detected for TiO₂ nanowires (Fig. S2, ESI†). The surface chemical composition data of these surfaces are summarized in Table 1, ESI.† The presence of the strong S (2p) peak and S (2s) peak in Fig. 2A (c) confirms that P(NMTE) has been successfully grafted onto TiO₂ nanowires. After the oxidation polymerization of 3-thiopheneacetic acid, the S (2p, 2s) peaks still exist, and the S atomic concentration can rise to 5.27% from 3.05% in Table 1 due to more thiophene units in the polymers. When Cd²⁺ was immobilized onto the polymer chains via cation exchange, the Cd 3d peaks appear at binding energies of 405.4 eV and 412.2 eV (Fig. S2†). The formation of CdS nanocrystals is verified by the signals of the S (2p) peak at binding energies of 162.6 and 161.6 eV, as shown in Fig. 5A (d). Similarly, after poly(3-hexylthiophene) is grafted onto the TiO₂ nanowires, the S (2p) peak is also evident. Interestingly, from Table 1, during self-assembly of the initiator on the TiO₂ nanowires, ROMP, oxidation polymerization and CdS NC loading on the TiO₂ nanowires, the Ti signal was decreasing continuously, and while the P3THA polymers were being grafted on and the CdS NCs being formed, the Ti signal was negligible, which reveals that high density, thick polymers have been grafted onto the TiO₂ nanowires that block detection of the Ti signal.

Fig. 5 (A) XPS survey spectra of (a) TiO₂ nanowires, (b) poly(NMTE)-TiO₂ nanowires, (c) P3THA-TiO₂ nanowires, (d) CdS-P3THA-TiO₂ nanowires, (e) P3TH-TiO₂ nanowires. (B) TGA traces of (a) TiO₂ nanowires, (b) TiO₂ nanowires – initiator modified, (c) poly(NMTE)-TiO₂ nanowires, (d) Cd-P3THA-TiO₂ nanowires, (e) CdS-P3THA-TiO₂ nanowires, (f) P3HT-TiO₂ nanowires. (C) UV-Vis spectra of TiO₂ nanowires (a), P3THA-TiO₂ nanowires (b), CdS-P3THA-TiO₂ nanowires (c). (D) Current density vs. voltage characteristics of TiO₂ nanowires (a), P3THA-TiO₂ nanowires (b), CdS-P3THA-TiO₂ nanowires (c).

TGA was carried out to evaluate the change of the organic portion during polymer grafting and CdS NC loading in Fig. 5B. TGA of TiO₂ nanowires measured a 10.58% weight loss between 100 and 600 °C. Upon modification of the ROMP initiator, the composite was found to have 12.89% weight loss, so 2.31 wt% of the weight loss was from the initiator. After the ROMP of NMTE monomers, a relatively thin polymer film was coated on the TiO₂ surface with a weight loss of 21.65%, as illustrated in Fig. 5B (c). When oxidation polymerization and the immobilization of Cd²⁺ ions onto polymer chains were performed, the TGA curve in Fig. 5B (d) shows an overall weight loss of 45.40%. After CdS NC formation, the overall weight loss decreases to 24.12%, due to the formation of high density CdS nanoparticles (Fig. 5B (e)). The P3HT grafting TiO₂ nanowires shows an overall weight loss of 55.58%, corresponding to the thick polymers.

The photo absorption properties of the resulting hybrid nanocomposites were explored by UV-Vis measurement in chlorobenzene. The TiO₂ nanowires show no photoresponse in the visible region, as shown in Fig. 5C (a). After grafting P3THA, with its large thiophene skeleton conjugated structure, onto the TiO₂ nanowires, as shown in Fig. 5C (b), the spectral absorption of P3THA grafted TiO₂ nanowires extended to the visible light region; the absorption peak appears at 405 nm and the absorption is evident before 500 nm, which corresponds to the energy band gap of polythiophene derivatives.⁴¹ The side chains, substituent groups and regularity of the conductive polymers have a great influence on the energy band gap,³ so the energy band gap can be altered from 2.33 to 1.90 eV.⁴² When the CdS NCs are loaded, the absorption peak of the hybrid nanocomposites moves to 450 nm and the absorption is shifted to 600 nm. CdS materials with an energy band gap of 2.25 eV can extend the absorption to a maximum of 550 nm.⁴³ After coupling of the CdS nanocrystals to conductive polymers, the absorption can be extended to 600 nm, implying improved charge transfer between P3THA and CdS NCs, probably due to the intimate contact between these two constituents. Fig. 5D shows current density versus voltage in 0.1 M Na₂S under 150 mW cm⁻² illumination. The open circuit voltage (V_oc) and short current density (I_sc) of TiO₂ nanowires were 0.55V and 0.067 mA cm⁻², respectively, as shown in Fig. 5D (a). After the conductive polymers are grafted onto the TiO₂ nanowires, V_oc and I_sc can reach 0.68 V and 0.090 mA cm⁻², which could be attributed to the conductive polymers, which possess visible light harvesting ability. The CdS nanocrystal/polymer hybrids produced an increased V_oc and I_sc with 0.84 V and 0.137 mA cm⁻², compared to polymer/TiO₂ materials, which revealed that the ordered p–n heterojunction can be used as a promising building block for harvesting visible light.

4. Conclusions

In conclusion, surface initiated polymerization and oxidation polymerization were exploited to produce chemically bonded p–n heterojunctions and create ordered nanocrystal/conjugated polymer hybrid nanocomposites with a well-defined interface via a step-by-step strategy on the nanoscale. It allows conducting polymer brushes to tailor the size and dispersion of nanocrystals in order to further tune the photo properties of these hybrid materials. More importantly, the nanocrystals tightly wrapped the polymer chains, thereby affording direct interfacial contact, which is expected to greatly improve the photoelectric performance, enabling excellent charge separation and transfer with minimal carrier recombination; so high photo conversion efficiency is expected in this nano-architecture. Further studies to the structural engineering of the conjugated polymers and rational nanocrystals or nanocrystal combination design with tunable size and shape control are of utmost importance for the further development of organic/inorganic hybrid solar cells.

Acknowledgements

This work was supported by NSFC (20973188, 21125316) and the “Hundred Talents Program” of Chinese Academy of Sciences.

References

1. Y. Zai, Y. Han, K Lu, Y. J. Zhang and Z. X. Wei, RSC Adv., 2012, 2, 338–343.
2. C. R. McNeill, Energy Environ. Sci., 2012, 5, 5653–5667.
3. H. Zhou, L. Yang and W. You, Macromolecules, 2012, 45, 607–632.
4. A. J. Moulé, L. Chang, C. Thambidurai, R. Vidu and P. Stroeve, J. Mater. Chem., 2012, 22, 2351–2368.
5. F. He and L. Yu, J. Phys. Chem. Lett., 2011, 2, 3102–3113.
6. F. C. Krebs, T. Tromholt and M. Jorgensen, Nanoscale, 2010, 2, 873–886.
7. J. E. Slota, X. He and W. T. S. Huck, Nano Today, 2010, 5, 231–242.
8. Y. Liang, D. Feng, Y. Wu, S.-T. Tsai, G. Li, C. Ray and L. Yu, J. Am. Chem. Soc., 2009, 131, 7792–7799.
9. M. Brinkmann, D. Aldakov and F. Chandezon, Adv. Mater., 2007, 19, 3819–3823.
10. R. Po, M. Maggini and N. Camaioni, J. Phys. Chem. C, 2010, 114, 695–706.
11. J. Y. Kim, K. Lee, N. E. Coates, D. Moses, T. Q. Nguyen, M. Dante and A. J. Heeger, Science, 2007, 317, 222–225.
12. L.-M. Chen, Z. Hong, G. Li and Y. Yang, Adv. Mater., 2009, 21, 1434–1449.
13. G. K. Mor, S. Kim, M. Paulose, O. K. Varghese, K. Shankar, J. Basham and C. A. Grimes, Nano Lett., 2009, 9, 4250–4257.
14. K. Takamizu and K. Nomura, J. Am. Chem. Soc., 2012, 134, 7892–7895.
15. Y. Bai, J. Zhang, D. Zhou, Y. Wang, M. Zhang and P. Wang, J. Am. Chem. Soc., 2011, 133, 11442–11445.
16. D. Zhou, Q. Yu, N. Cai, Y. Bai, Y. Wang and P. Wang, Energy Environ. Sci., 2011, 2030–2034.
17. X. Shen, B. Sun, F. Yan, J. Zhao, F. Zhang, S. Wang, X. Zhu and S. Lee, ACS Nano, 2010, 4, 5869–5876.
18. X. Shen, B. Sun, D. Liu and S. T. Lee, J. Am. Chem. Soc., 2011, 133, 19408–19415.
19. P. V. Kamat, J. Phys. Chem. C, 2008, 112, 18737–18753.
20. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2008, 130, 4007–4015.
21. W. U. Huynh, J. J. Dittmer and A. P. Alivisatos, Science, 2002, 295, 2425–2427.
22. J. Bouclé, S. Chyla, M. S. P. Shaffer, J. R. Durrant, D. D. C. Bradley and J. Nelson, Adv. Funct. Mater., 2008, 18, 622–633.
23. M. D. Goodman, J. Xu, J. Wang and Z. Lin, Chem. Mater., 2009, 21, 934–938.
24. Z. Lin, Chem.–Eur. J., 2008, 14, 6294–6301.
25. J. Albero, Y. Zhou, M. Eck, F. Rauscher, P. Niyamakom, I. Dumsch, S. Allard, U. Scherf, M. Krüger and E. Palomares, Chem. Sci., 2011, 2, 2396–2401.
26. J. Xu, J. Hu, X. Liu, X. Qiu and Z. Wei, Macromol. Rapid Commun., 2009, 30, 1419–1423.
27. A. J. Moulé and K. Meerholz, Adv. Mater., 2008, 20, 240–245.
28. Y.-Y. Lin, T.-H. Chu, S.-S. Li, C.-H. Chuang, C.-H. Chang, W.-F. Su, C.-P. Chang, M.-W. Chu and C.-W. Chen, J. Am. Chem. Soc., 2009, 131, 3644–3649.
29. L. Zhao, X. Pang, R. Adhikary, J. W. Petrich and Z. Lin, Angew. Chem., Int. Ed., 2011, 50, 3958–3962.
30. L. Zhao, X. Pang, R. Adhikary, J. W. Petrich, M. Jeffries-EL and Z. Lin, Adv. Mater., 2011, 23, 2844–2849.
31. J. Qian, Q. S. Liu, G. Li, K. J. Jiang, L. M. Yang and Y. Song, Chem. Commun., 2011, 47, 6461–6463.
32. N. Balis, V. Dracopoulos, E. Stathatos, N. Boukos and P. Lianos, J. Phys. Chem. C, 2011, 115, 10911–10916.
33. H.-C. Liao, C.-C. Lin, Y.-W. Chen, T.-C. Liu and S.-Y. Chen, J. Mater. Chem., 2010, 20, 5429–5435.
34. V. Senkovskyy, R. Tkachov, T. Beryozkina, H. Komber, U. Oertel, M. Horecha, V. Bocharova, M. Stamm, S. A. Gevorgyan, F. C. Krebs and A. Kiriy, J. Am. Chem. Soc., 2009, 131, 16445–16453.
35. D. Wang, Q. Ye, B. Yu and F. Zhou, J. Mater. Chem., 2010, 20, 6910–6915.
36. A. M. Jonas, K. Glinel, R. Oren, B. Nysten and W. T. S. Huck, Macromolecules, 2007, 40, 4403–4405.
37. X. Zhou, X. Liu, Z. Xie and Z. Zheng, Nanoscale, 2011, 3, 4929–4939.
38. T. Xu, M. Yan, J. D. Hoefelmeyer and Q. Qiao, RSC Adv., 2012, 2, 854–862.
39. J. Yan, Q. Ye and F. Zhou, RSC Adv., 2012, 2, 3978–3985.
40. O. Azzaroni, A. A. Brown, N. Cheng, A. Wei, A. M. Jonas and W. T. S. Huck, J. Mater. Chem., 2007, 17, 3433–3439.
41. R. Kroon, M. Lenes, J. C. Hummelen, P. W. M. Blom and B. de Boer, Polym. Rev., 2008, 48, 531–582.
42. A. Dkhissi, F. Ouhib, A. Chaalane, R. C. Hiorns, C. Dagron-Lartigau, P. Iratcabal, J. Desbrieres and C. Pouchan, Phys. Chem. Chem. Phys., 2012, 14, 5613–5619.
43. C. X. Guo, H. B. Yang, Z. M. Sheng, Z. S. Lu, Q. L. Song and C. M. Li, Angew. Chem., Int. Ed., 2010, 49, 3014–3017.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c2ra21546b

---