Polymer chain/nanocrystal ordering in thin films of regioregular poly(3-hexylthiophene) and blends with a soluble fullerene

Abstract

Here we report a study of the polymer chain/nanocrystal ordering in thin films (nanolayers) of regioregular poly(3-hexylthiophene) (P3HT) and blends of P3HT with a soluble fullerene derivative. A detailed analysis has been made of two dimensional (2D) grazing incidence X-ray diffraction (GIXRD) measurements with synchrotron radiation. P3HT samples with three different levels of regioregularity (RR) were synthesized and used to investigate the influence of RR on the chain ordering in thin films. Blend films were also prepared to investigate the influence of fullerene molecule addition on chain ordering. For the analysis, one dimensional (1D) GIXRD patterns were extracted from the 2D images for varying azimuthal angles, allowing information to be obtained for chain ordering in both the out-of-plane (OOP) and in-plane (IP) directions. These results show that the degree of P3HT chain ordering is strongly affected by RR, and that thermal annealing improves chain ordering in the OOP direction. This observation is in good agreement with high resolution transmission electron microscope measurements of film nanomorphology.

---

Introduction

Regioregular poly(3-hexylthiophene) (P3HT) has attracted keen interest for application in polymer solar cells^1–8 and field effect transistors^9–11 following its first synthesis in 1993.¹² This interest can be attributed to the resulting enhanced charge carrier mobility^9,13 and extended light absorption (up to 650 nm).^4,14 These improved properties are known to have an origin in the strong tendency of regioregular P3HT chains to self-organize within thin film samples. Orientation of the P3HT thiophene backbone ring plane normal to the substrate plane is also often observed, particularly with low surface energy substrates.^8,9,12 In addition, the P3HT chain ordering becomes much pronounced upon thermal annealing.⁸

P3HT chain ordering has also been observed for blend films with varying fractions of a soluble fullerene, [6,6]-phenyl C61-butyric acid methyl ester (PCBM), that are prepared using chloroform,¹⁵ chlorobenzene,⁸ and 1,2-dichlorobenze¹⁶ as solvents. The degree of ordering was, however, found to be dependent on the solvent type. We reported further that changes in the degree of regioregularity have a strong influence on the performance of polymer solar cells based on P3HT–PCBM blends.⁸ However, those studies focused on more general observations concerning chain ordering–electronic property relationships rather than on the detailed conformation of the P3HT chains at the molecular level.

In this work, we have undertaken a detailed analysis of the degree of P3HT chain orientation as a function of azimuthal angle (the extrema correspond to the out-of-plane (OOP) and in-plane (IP) substrate directions (see Fig. 1)). In real samples we don't expect a perfect orientation of the thiophene ring plane in the OOP direction and it is then important to consider the distribution of orientations that arises. P3HT samples with three different degrees of regioregularity (90.7%, 93%, and 95.2% RR) were used to examine the influence of regioregularity (RR) on chain ordering again as a function of azimuthal angle. Two-dimensional (2D) images were measured using a synchrotron radiation grazing incidence X-ray diffraction (GIXRD) apparatus, from which one-dimensional (1D) GIXRD patterns (intensity versus diffraction angle) at varying azimuthal angles (between the OOP and IP extrema) were extracted. The nanomorphology of films was additionally evaluated using a high resolution field emission gun-transmission electron microscope (FEG-TEM) in order to correlate the GIXRD deduced P3HT chain ordering with P3HT chain/nanocrystal spacings in the IP direction.

Fig. 1 (a) Schematic illustration of the synchrotron GIXRD measurement for a P3HT film (nanolayer) on a quartz substrate where the incidence angle (θ_IN) of the synchrotron X-ray beam (I_IN) is 0.2°: I_IP, I_OOP, 2θ, and α denote the diffracted X-ray intensity for the in-plane direction, the diffracted X-ray intensity for the out-of-plane direction, the diffraction angle, and the azimuthal angle, respectively. (b) Example of a 2D GIXRD image (P3HT film (95.2% RR)) with diffraction orders (100), (200), (300), and (010) as indicated. (c) Schematic illustration of the X-ray diffraction (dashed arrows) direction with respect to the ordering direction (azimuthal angle) for theoretical P3HT nanocrystals constructed using an energy minimized structure of 11 P3HT repeat units (assuming 100% RR): the incident X-ray beam is assumed to impinge from a direction that is nearly normal to the plane of the page with an upward tilt angle of 0.2°, whilst the ‘a’ and ‘c’ axes are as previously described.⁸

Experimental

The regioregular P3HT samples were synthesized by Merck Chemicals using their standard synthetic route and were end-capped with hydrogen.¹⁷ The weight average molecular weights (M_w), polydispersity indices (PDI), and melting points (T_m) were as follows: for 95.2% RR P3HT, M_w = 2.19 × 10⁴, PDI = 1.57, T_m = 212 °C; for 93% RR P3HT, M_w = 3.19 × 10⁴, PDI = 1.79, T_m = 211 °C; for 90.7% RR P3HT, M_w = 4.59 × 10⁴, PDI = 1.94, T_m = 203 °C. All P3HT samples were purified in the same manner to give typical metallic impurities Ni < 2 µg g⁻¹ and Mg < 50µg g⁻¹. PCBM was synthesized at the University of Groningen (The Netherlands) and used as received.¹⁸ Pristine P3HT solutions and blend solutions (P3HT : PCBM = 1 : 1 by weight) were prepared using chlorobenzene (CB) at a solid concentration of 30 mg ml⁻¹ and 60 mg ml⁻¹, respectively. The pristine and blend solutions were spin-coated onto quartz substrates (spectrosil B). Additional films were made by spin-coating the same solutions on top of poly(3,4-ethylenedioxythiophene)–poly(styrenesulfonate) (PEDOT–PSS) (Baytron P standard grade, HC Stark) thin films that had been spin-coated on to indium–tin oxide (ITO) coated glass substrates. The latter constitutes the same substrate structures that are used for polymer solar cell fabrication. Thermal annealing of spin-coated films was carried out at 140 °C for 2 h inside a nitrogen filled glove box.

2D GIXRD images of the P3HT and P3HT–PCBM blend films were measured using a high power X-ray beam (photon flux (I_IN) ≈ 10¹¹ photons s⁻¹ mrad⁻¹ per 0.1%, beam size ≤ 0.5 mm²) from a synchrotron radiation source (4C2 beamline, Pohang Accelerator Laboratory, South Korea) and a detection system equipped with a 2D X-ray detector (PI-SCX4300-165/2, Princeton Instruments) (see Fig. 1a for the experimental geometry). For all of these measurements the incidence angle (θ_IN) was fixed as 0.2°. The azimuthal angle dependent 1D GIXRD patterns were extracted from corresponding 2D GIXRD images using a custom program based on Microsoft Visual C⁺⁺. The in-plane (IP) nanomorphology of the P3HT and P3HT–PCBM blend films was measured using a FEG-TEM (JEM-2100F, JEOL, Japan).

Results and discussion

As shown in Fig. 1a, the 2D GIXRD measurement provides information on the P3HT chain orientation for the directions that lie between the OOP and IP directions. This information can be obtained by extracting intensity versus diffraction angle plots as a function of azimuthal angle (α) from the corresponding 2D GIXRD image (see Fig. 1b for a representative 2D GIXRD image). This image clearly shows the dependence of the diffracted X-ray intensity on the azimuthal angle: the (100), (200), (300), and (010) diffraction orders vary in strength as α varies. On the basis of the previous reports,^8,9,12 we can decouple the degree of P3HT chain ordering as shown in Fig. 1c: the a-axis⁸ chain ordering occurs solely in the OOP direction in the case of α = 0°, whilst it occurs solely in the IP direction for α = ±90°; otherwise (0° < α < 90°), the a-axis chain ordering should occur in a direction that lies between the OOP and IP directions (for example, see the two cases for α = ±45° in Fig. 1c). The two azimuthal angles (+α and −α) would need to be separately considered if the chain ordering were asymmetric with respect to the OOP direction (axis). However, as seen from this and other images (Fig. 2), most of the 2D GIXRD images of P3HT films studied in this work show symmetric shapes, indicating no particular tilted chain ordering with respect to the OOP axis. Therefore we will present only the (half space) data for positive (+) azimuthal angles.

Fig. 2 1D GIXRD intensity profiles (2D images inset) for pristine P3HT films parametric in the azimuthal angle (lines from top to bottom in each plot: α = 0°, 18.44°, 45°, 63.45°, and 90°). The different plots are: (left top) as-spun 95.2% RR, (right top) annealed 95.2% RR, (left middle) as-spun 93% RR, (right middle) annealed 93% RR, (left bottom) as-spun 90.7% RR, and (right bottom) annealed 90.7% RR.

Fig. 2 shows the 1D GIXRD patterns of pristine P3HT films for varying azimuthal angles. As the azimuthal angle increases, each of the characteristic peaks of the P3HT film samples decreases irrespective of the polymer RR and/or thermal annealing (see Fig. 2a). Note that the (010) peak will not be further discussed here because it is overlapped by a substrate reflection peak. The dominant diffraction (100) peak clearly shows a gradual reduction in intensity as the azimuthal angle increases, but the peak is still apparent for all films even when α = 90°. This strongly suggests that there is at least some a-axis chain ordering within the IP direction for all RR P3HT films (cf. the far-right schematic diagram (α = ±90°) in Fig. 1c). The (200) and (300) peaks, however, almost disappear at α = 18.44° for both the 90.7% and the 93% RR P3HT films, whereas they still exist for the 95.2% RR P3HT film. This trend is true for both as-spun (not-annealed) and annealed P3HT films. In addition, the (300) peak is almost negligible for as-spun 90.7% and 93% RR P3HT films, whilst it is observable for the corresponding annealed films. The intensity of the (300) peak is much weaker for annealed 90.7% and 93% RR P3HT films than for the annealed 95.2% RR P3HT film. These results indicate a correlation between the degree of RR and the intensity of the higher order diffraction peaks that relate to higher degrees of crystal and/or chain ordering.¹⁹ In addition, the higher order diffraction peaks, (200) and (300), completely disappear for α > 45° irrespective of RR and thermal annealing (these peaks exist for α ≈ 27° for the 95.2% RR P3HT films though the data are not shown here). This indicates that most of the longer range P3HT chain orientation occurs for directions that lie within a cone centred on the OOP direction.

As shown in Fig. 3, compared to the situation for pristine P3HT films, the as-spun blend films show an abrupt intensity decrease for the (100) peak for azimuthal angles between α = 0° and α = 18.44°. This is considered to be a result of the influence of PCBM molecules that disrupt the P3HT chain ordering in blend films. This disruption is partially recovered by thermal annealing but the effect is still more pronounced for the annealed blend films than for the annealed pristine films. In addition, the (300) peak is negligibly small for the blend films irrespective of the degree of RR and for all azimuthal angles, even after the films have been thermally annealed. This means that the highest degrees of chain ordering could not be achieved in P3HT–PCBM blend films even upon thermal annealing. The different dark current density versus voltage (J–V) characteristics observed for devices made with P3HT and P3HT–PCBM blend films are considered to relate closely to this situation: lower dark current densities are found for devices with blend films.⁸

Fig. 3 1D GIXRD intensity profiles (2D images inset) for P3HT:PCBM blend films parametric in the azimuthal angle (lines from top to bottom in each plot: α = 0°, 18.44°, 45°, 63.45°, and 90°). The different plots are: (left top) as-spun 95.2% RR, (right top) annealed 95.2% RR, (left middle) as-spun 93% RR, (right middle) annealed 93% RR, (left bottom) as-spun 90.7% RR, and (right bottom) annealed 90.7% RR.

Using these 1D GIXRD patterns, the intensity change of the (100) peak as a function of azimuthal angle can be plotted as shown in Fig. 4 and 5 to allow detailed studies of the effects of RR and thermal annealing. Fig. 4 shows that the (100) intensity decreases for all films as the azimuthal angle increases, exhibiting two distinctive stages with rapid and slow decays. In particular, the intensity remains non-zero even at α = 90°, indicating the existence of a-axis chain ordering even for the IP direction. This confirms that the a-axis chain ordering occurs over the entire azimuthal angle range, although predominantly in the OOP direction. For both as-spun and annealed P3HT films the steep drop in intensity occurs over an angular spread Δα > 15° whilst for P3HT–PCBM blend films Δα < 15°.

Fig. 4 Normalized X-ray diffraction intensity as a function of azimuthal angle for P3HT and P3HT–PCBM blend films. Parametric in regioregularity (left top: as-spun P3HT films; left bottom: annealed P3HT films; right top: as-spun P3HT:PCBM films; right bottom: annealed P3HT:PCBM films): 95.2% RR (filled squares), 93% RR (filled circles), 90.7% RR (filled triangles). Arrows denote the corresponding inflection points between OOP and IP direction a-axis chain ordering.

Fig. 5 Normalized X-ray diffraction intensity as a function of azimuthal angle for P3HT and P3HT:PCBM blend films. Parametric in thermal annealing (P3HT (left panels) and P3HT–PCBM blend (right panels) films with top: 95.2% RR; middle: 93% RR; bottom: 90.7% RR before (filled squares) and after (open circles) thermal annealing). Arrows denote the corresponding inflection points between OOP and IP direction a-axis chain ordering.

It is interesting to note that for the as-spun P3HT films, the higher the degree of RR the lower the (100) intensity in the slow decay stage (α > 15°), leading to more pronounced a-axis chain ordering in the OOP direction (less pronounced a-axis ordering in the IP direction) with an increasing degree of RR. This trend is similarly observed for the annealed P3HT films, although the (100) intensity difference in the slow decay stage with respect to the degree of RR is much smaller for the annealed P3HT films than for the as-spun P3HT films. We note, however, that for both as-spun and annealed pristine P3HT films the 95.2% RR film showed a little higher intensity in the rapid decay stage (0° < α < 15°) compared to the other RR films. This confirms again that a-axis chain ordering in the OOP direction becomes more pronounced with increasing RR. In the case of blend films this trend with RR is also observed for both rapid and slow decay stages though the intensity difference in the slow decay stage varies less with RR for the P3HT–PCBM blend films than for the pristine P3HT films. This discrepancy between pristine and blend films is attributed to the influence of the PCBM molecules in the blend as mentioned above.

In order to investigate the detailed effect of thermal annealing, the 1D GIXRD patterns for as-spun and annealed films were separately compared for each degree of regioregularity (Fig. 5). Thermal annealing reduced the a-axis chain ordering in the IP direction for all of the pristine films irrespective of their RR, but its effect becomes more pronounced as the RR decreases. Here it is noteworthy that the intensity of the a-axis chain ordering at α = 90° is close to zero for the 95.2% RR P3HT sample, whereas it remains greater than 0.1 for the 90.7% RR P3HT sample even after thermal annealing (>0.2 for the as-spun P3HT film). Therefore, assuming that the inflection point between the rapid and slow intensity decays as a function of azimuthal angle represents a boundary in terms of OOP ordering changes, we can deduce that the degree of a-axis chain ordering in the IP direction is more than 10% (∼20% at the inflection point) and 20% (∼35% at the inflection point) for annealed and as-spun 90.7% RR P3HT films, respectively, whilst the degree is close to 0% for both as-spun and annealed 95.2% RR P3HT films. This remarkable difference in the degree of IP direction a-axis chain ordering (0% for the 95.2% RR P3HT film versus 20% for the 90.7% RR P3HT film) is surprising considering that it arises from only a 4.5% RR difference between 95.2% and 90.7% RR P3HTs.

In contrast to the pristine P3HT films that exhibit the reduced intensity in both rapid and slow decay stages upon thermal annealing irrespective of the degree of RR (except α = 0°), the P3HT–PCBM blend films show quite different trends: the intensity in both rapid and slow decay stages was increased upon thermal annealing for the 95.2% RR P3HT–PCBM blend film, whereas it was nearly the same or marginally decreased for the 93% and 90.7% RR P3HT–PCBM blend films. This difference can be attributed to the influence of the PCBM molecules on the P3HT chain ordering during thermal annealing. The influence of PCBM is considered to be responsible for the higher intensity of the a-axis chain ordering at α = 90° for the blend films than for the pristine films irrespective of RR. Applying the same assumption for the inflection point between the rapid and slow decay stages as a function of α as a boundary between OOP and IP direction chain ordering, we can also deduce the degree of a-axis chain orientation in the IP direction as ∼10% and ∼20% for the 95.2% RR and 90.7% RR P3HT–PCBM blend films, respectively. This result confirms that a-axis chain ordering in the IP direction is dominant for lower regioregularity P3HT films, leading in part to the lower performance observed for polymer solar cells made with lower RR P3HT.⁸

Considering the above detailed analysis of the P3HT a-axis chain ordering, it can be anticipated that the IP direction nanomorphology of the pristine and blend films differs depending on the RR of the P3HT sample. In this regard, the nanomorphology of the films was investigated using a FEG-TEM as shown in Fig. 6a–e. Since the incidence electron direction (see Fig. 6f) used here is normal to the substrate (film plane), these measurements give us information on the lateral (IP) direction nanomorphology of the studied films. In more detail, if all a-axis P3HT nanocrystals are perfectly ordered in the OOP direction, we can expect to see a very fine-grained nanomorphology resulting from the lowest possible spacing between P3HT chains and/or nanocrystals: this spacing will asymptotically approach the b-axis (interplane) spacing as the structural perfection improves. If perfect OOP ordering does not occur then the IP direction spacing will tend to become larger and larger as the degree of OOP a-axis chain ordering decreases.

Fig. 6 FEG-TEM images of annealed P3HT and P3HT–PCBM blend films: (a) 95.2% RR P3HT film, (b) 90.7% RR P3HT film, (c) 95.2% RR P3HT–PCBM blend film, and (d) 90.7% RR P3HT–PCBM blend film. For comparison the image (e) of an as-spun 90.7% RR P3HT–PCBM blend film is shown in the right bottom panel. A schematic illustration of the TEM geometry is shown in the left bottom panel (f). The scale bar size is the same for all images.

As shown in Fig. 6a and 6b for the pristine P3HT films, the overall nanomorphology is finer-grained for the higher RR sample than for the lower RR sample though the difference is not very pronounced. This is consistent with the above analysis based on the azimuthal angle dependence of the 1D GIXRD patterns: lower degrees of RR result in higher degrees of a-axis chain ordering in the IP direction (lower degrees of a-axis chain ordering in the OOP direction). This leads to a greater chain–crystal spacing in the IP direction. The same trend is also observed for the blend films (see Fig. 6c and 6d), an effect that might result in the different device performance found for samples with different RR.⁸ In addition, it should also be noted that thermal annealing leads to blend films (see Fig. 6d and 6e) with a smaller spacing between nanocrystals, in good agreement with the GIXRD results as a function of azimuthal angle.⁸

Conclusions

The P3HT chain ordering in P3HT and P3HT–PCBM blend films was analyzed using azimuthal angle dependent 1D GIXRD patterns. For each of the three degrees of RR studied here the higher order diffraction peaks for the P3HT films completely disappear at α ≈ 30°, indicating that long range crystal ordering with neighbouring crystal domains is only effective for angles α ≤ 30° from the OOP direction. The degree of a-axis chain ordering in the OOP direction was higher for the higher RR P3HT samples than for the lower RR samples. Thermal annealing improves the degree of a-axis chain ordering in the OOP direction, an effect that is especially pronounced for the lower RR P3HT samples. Assuming that the inflection point between the rapid and slow intensity decays as a function of azimuthal angle represents a boundary in terms of OOP ordering changes, most of the annealed P3HT films show ≈80% a-axis chain ordering in the OOP direction whilst only ≈65% a-axis chain ordering in the OOP direction is deduced for the as-spun 90.7% RR P3HT film. In contrast to the P3HT films, the P3HT–PCBM blend films show ≈80% or more a-axis chain ordering in the OOP direction independent of annealing. These chain ordering effects correlate well with the film nanomorphology probed with FEG-TEM measurements. The latter showed that the higher the P3HT RR the smaller the spacing between nanocrystals.

Acknowledgements

The Imperial College London authors thank British Petroleum International for financial support via the OSCER project. M. R. and C. S. H. thank the National Research Laboratory programs [polymer synthesis & physics lab. (M. R.) and nanoinformation materials lab. (C. S. H.)] of South Korea for financial support.

References

1. F. Padinger, R. S. Ritterberger and N. S. Sariciftci, Adv. Funct. Mater., 2003, 13, 85.
2. C. J. Brabec, Sol. Energy Mater. Sol. Cells, 2004, 83, 273.
3. D. Chirvase, J. Parisi, J. C. Hummelen and V. Dyakonov, Nanotechnology, 2004, 15, 1317.
4. Y. Kim, S. A. Choulis, J. Nelson, D. D. C. Bradley, S. Cook and J. R. Durrant, J. Mater. Sci., 2005, 40, 1371.
5. Y. Kim, S. A. Choulis, J. Nelson, D. D. C. Bradley, S. Cook and J. R. Durrant, Appl. Phys. Lett., 2005, 86, 063502.
6. M. Reyes-Reyes, K. Kim and D. L. Carroll, Appl. Phys. Lett., 2005, 87, 083506.
7. G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery and Y. Yang, Nat. Mater., 2005, 4, 864.
8. Y. Kim, S. Cook, S. M. Tuladhar, S. A. Choulis, J. Nelson, J. R. Durrant, D. D. C. Bradley, M. Giles, I. McCulloch, C. S. Ha and M. Ree, Nat. Mater., 2006, 5, 197.
9. H. Sirringhaus, P. J. Brown, R. H. Friend, M. M. Nielsen, K. Bechgaard, B. M. W. Langeveld-Voss, A. J. H. Spiering, R. A. J. Janssen, E. W. Meijer, P. Herwing and D. M. de Leeuw, Nature, 1999, 401, 685.
10. H. Sirringhaus, N. Tessler and R. H. Friend, Science, 1998, 280, 1741.
11. L. Bürgi, T. J. Richards, R. H. Friend and H. Sirringhaus, J. Appl. Phys., 2003, 94, 6129.
12. R. D. McCullough, S. Tristram-Nagle, S. P. Williams, R. D. Lowe and M. Jayaraman, J. Am. Chem. Soc., 1993, 115, 4910.
13. S. A. Choulis, Y. Kim, J. Nelson, D. D. C. Bradley, M. Giles, M. Shkunov and I. McCulloch, Appl. Phys. Lett., 2004, 85, 3890.
14. P. J. Brown, D. S. Thomas, A. Köhler, J. S. Wilson, J. S. Kim, C. M. Ramsdale, H. Sirringhaus and R. H. Friend, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 67, 064203.
15. T. Erb, U. Zhokhavets, G. Gobsch, S. Raleva, B. Stühn, P. Schilinsky, C. Waldauf and C. J. Brabec, Adv. Funct. Mater., 2005, 15, 1193.
16. Y. Kim, J. Nelson, J. R. Durrant and D. D. C. Bradley, Adv. Funct. Mater., 2006 , to be submitted.
17. G. Koller, B. Falk, C. Weller, M. Giles and I. McCulloch, Process of preparing regioregular poly(3-substituted thiophenes), PCT Int. Appl., WO2005014691, 2005.
18. J. C. Hummelen, P. W. Knight, F. LePeg, F. Wudl, J. Yao and C. L. Wilkins, J. Org. Chem., 1995, 60, 532.
19. C. Hammond, The basics of crystallography and diffraction, Oxford Science Publications, Oxford University Press, Oxford, 2nd edn, 2001.

---

Footnote

† The HTML version of this article has been enhanced with colour images.

---