Oligothiophene-based nanostructures: from solution to solid-state aggregates

Oligothiophene-based nanostructures: from solution to solid-state aggregates

Synthetic Metals 147 (2004) 67–72 Oligothiophene-based nanostructures: from solution to solid-state aggregates M. Surina , R. Lazzaronia , W.J. Feast...

245KB Sizes 0 Downloads 12 Views

Synthetic Metals 147 (2004) 67–72

Oligothiophene-based nanostructures: from solution to solid-state aggregates M. Surina , R. Lazzaronia , W.J. Feastb , A.P.H.J. Schenningc , E.W. Meijerc , Ph. Lecl`erea,c,∗ a

c

Service de Chimie des Mat´eriaux Nouveaux, Universit´e de Mons-Hainaut/Materia Nova, Place du Parc, 20, B 7000 Mons, Belgium b IRC in Polymer Science and Technology, Durham University, South Road, Durham, DH13LE, UK Laboratory of Macromolecular and Organic Chemistry, Eindhoven University of Technology, PO Box 513, NL 5600 MB, Eindhoven, The Netherlands Received 30 April 2004; received in revised form 4 May 2004; accepted 4 May 2004

Abstract The possibility to develop optoelectronic devices with improved properties by controlling the degree of organization at the molecular level of organic materials has been driving the design of new ␲-conjugated systems. In particular, the organization by self-assembling processes (␲–␲ interactions, hydrogen bonding) of well-defined oligomeric systems such as disubstituted oligothiophene derivatives has been demonstrated as a promising approach to conjugated materials with a high degree of structural order of the constituent building blocks. Here, tapping-mode atomic force microscopy is used to investigate the morphologies of (i) thin deposits made from assembly of thiophene-based oligomers starting from molecularly dissolved solutions or (ii) aggregates already formed in solution. In order to understand the results in terms of supramolecular organization, comparisons with molecular modeling simulations are performed. During the self-assembly processes, the interplay between the conjugated molecules, the solvent, and the substrate surface is of primary importance. Depending on the interactions between the molecules and the substrate, one-dimensional (nanowires) or two-dimensional (platelets) objects can be generated. The self-organization of conjugated building blocks in solution or in solid-state on surfaces represents a starting point for the construction of molecular electronics or even circuits, through surface patterning with nanometer-sized objects. © 2004 Elsevier B.V. All rights reserved. Keywords: Conjugated materials; Self-assembly; Scanning probe microscopy

1. Introduction Nanoscopic and mesoscopic (10–1000 nm) order in ␲conjugated systems is a topic of utmost importance because it determines the performance of the materials when used as components in organic electro-optical devices such as photovoltaic diodes [1], light emitting diodes (LEDs) [2] and field effect transistors (FETs) [3,4]. Well-defined ␲-conjugated oligomers play an important role in this field because their precise chemical structure and conjugation length gives rise to well-defined functional properties and facilitates control over their supramolecular organization [5]. Oligomers and ∗

Corresponding author. Tel.: +32 65 37 38 60; fax: +32 65 37 38 61. E-mail address: [email protected] (Ph. Lecl`ere).

0379-6779/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.synthmet.2004.05.022

polymers based on ␣,␣ -linked thiophenes are at the forefront of organic semiconductor materials with potential for applications in FETs and related structures [6]. In this application, high structural order of the semiconducting layer in the region very near the dielectric interface (up to few nm above it) is needed to achieve good charge transport properties [5d, e]. For example, the formation of supramolecular interactions via hydrogen bonding arrays in combination with ␲–␲ stacking has been used to self-assemble oligothiophenes into one-dimensional arrays on surfaces, generating a material with remarkably high charge carrier mobility [7]. In this way, the understanding of the self-organization of conjugated building blocks on surfaces, in combination with surface patterning with nanometer-sized objects, is a key step to evolve towards organic nanoelectronics or circuits [8–9].

68

M. Surin et al. / Synthetic Metals 147 (2004) 67–72

2.2. Sample preparation The samples for AFM were prepared by casting thin films onto substrates from THF, toluene or n-butanol solutions (typically 0.01 mg/ml). The substrates were freshly cleaved muscovite mica, silicon wafer, or graphite. The solvent was allowed to evaporate slowly at room temperature in air or in a solvent-saturated atmosphere. 2.3. Atomic force microscopy Tapping-mode atomic force microscopy was performed with a Nanoscope IIIa microscope from Digital Instruments/Veeco (operating in air at room temperature). Microfabricated silicon cantilevers were used with a spring constant of 30 N/m. Images of different areas of the sample were collected with the maximum available number of pixels (512) in each direction. The Nanoscope 5.12 image processing software was used for image analysis. Scheme 1. .

The control of molecular assembly to give well-defined structures on the nanoscale can be carried out via different complementary approaches: (i) conjugated oligomers can be sublimed and it is possible to follow their assembly as they form thin deposits on surfaces from individual molecules in the vapor phase; (ii) self-assembly can take place in solution (in relatively poor solvents) and (iii) if the compounds are soluble, one can generate thin deposits from conjugated compounds molecularly dispersed in a solution; aggregation takes place during the deposition and then depends on the interplay between the conjugated molecules, the solvent and the substrate surface. In this context, the present work focuses on the supramolecular organization of a series of soluble ␣,␣ -substituted oligothiophenes (Scheme 1). Thin deposits on atomically flat substrates are investigated by means of tapping-mode atomic force microscopy; the structures observed are deposits of aggregates already formed in solution (in a poor solvent) and from compounds molecularly dissolved in a solution. In the latest case, the influence of the substrate polarity on the supramolecular organization is also considered.

2. Experimental 2.1. Materials We have considered quinquethiophene, sexithiophene, and heptathiophene molecules (5T, 6T, and 7T, respectively), end-capped at the ␣ terminal positions by one or two short, flexible, and strongly hydrophilic segments of poly(ethylene oxide) (PEO) or an alkyl chain. The chemical structures of the oligothiophenes under investigation are described in Scheme 1. The synthesis of the compounds is described in Refs. [10–13].

3. Results and discussion 3.1. Aggregation in solution The UV–vis absorption and fluorescence spectra recorded for compound 1a and 2b in THF are typical for molecularly dissolved disubstituted sexithiophene chromophores, with a main band at λmax = 452 nm (Fig. 1a, shown for compound 1a) [14]. The fluorescence spectrum has a maximum at 528 nm and shows a characteristic vibronic fine structure, while the high intensity observed provides further evidence that the material is molecularly dissolved. In contrast, solutions of the same compound 1a in n-butanol at 20 ◦ C show a UV–vis absorption spectrum typical for aggregated sexithiophene derivatives [15] with a main band at λmax = 381 nm, blue-shifted by ca. λ = 71 nm compared to the THF solution (Fig. 1a). Similarly, for 2b in butanol, the UV–vis absorption indicates a main band at λmax = 403 nm, i.e., blue-shifted by λ∼50 nm compared the THF solution. These observations, together with evidence of aggregation in butanol from electrospray FT-ICR mass spectrometry measurements [16], strongly support the fact that these compounds aggregate in n-butanol. The shape and the size of these aggregates can be investigated by AFM measurements on deposits prepared from n-butanol solutions. The deposits of compound 2b on a mica surface show a microscopic morphology of small rods lying on the substrate, as shown in Fig. 1b; the height image clearly exhibits small aggregates of varying sizes (in bright) lying on the flat substrate (in dark). The phase image shows that the internal structure of those aggregates is made of a few rod-like objects. The average length and width of those rod-like objects is about 100 ± 15 nm and 50 ± 15 nm, respectively; their thickness is between 5 and 10 nm. These structures are likely to be related to the aggregates whose formation is deduced from solution spectroscopic studies.

M. Surin et al. / Synthetic Metals 147 (2004) 67–72

69

Fig. 1. (a) UV–vis absorption and fluorescence spectra of 1a in THF and n-butanol, (b) AFM height (left) and phase (right) images (2.0 ␮m × 2.0 ␮m) of 2b aggregates prepared from n-butanol solution.

3.2. Assembly from solution Using very dilute THF (or toluene) solutions allows the generation of thin deposits in the sub-monolayer range, which are particularly appropriate for high-resolution AFM studies. For all the studied molecules, the microscopic morphology of deposits on mica from THF or toluene solutions

(two good solvents) exhibits ultrathin regular layers lying on the substrate, i.e., islands with a very regular thickness of few nm, with lateral dimensions varying from few hundred nm to few ␮m. An illustration of this type of structure is shown in Fig. 2a (for compound 2b); in this case, the thickness of the layers is 2.7 ± 0.1 nm. Compared to the assembly in solution (vide supra), the deposits from a good solvent

Fig. 2. AFM height images (3.0 ␮m × 3.0 ␮m) of thin deposits of 2b on (a) mica and (b) on graphite.

70

M. Surin et al. / Synthetic Metals 147 (2004) 67–72

leads to layers presenting the same thickness whatever the preparation conditions (concentration, type of good solvent), but only depends on the chemical structure of the oligomer. In this case the molecules self-assemble on the surface while the solvent evaporates and not within the solution. In order to correlate the thickness values with the molecular architecture, molecular modeling is used to estimate the length of the various molecules in their fully extended configuration. The oligothiophene backbones are optimized with adjacent monomers in the S-anti configuration; the oligoethyleneoxide (EO) and hexyl groups are considered in the all-trans configuration [17]. The angle between the thiophene oligomer and the end-groups is optimized, which can lead to conformations with a bend between the oligothiophene and the end-groups. In all cases, the length of the fully extended molecule is given as the distance between the extremities of the molecule. The thickness of the first layer (as measured with AFM) is plotted versus the length of the fully extended molecules in Fig. 3. We observe that: (i) the thickness of the layers globally increases with the length of the molecule; (ii) its value is always lower than the length of the fully extended molecules (the solid line represents the 1:1 ratio). That means that the molecules are not fully extended perpendicular to the substrate. Comparing sets of molecules differing by a single structural parameter can provide further information on the molecular orientation within the monolayers:

and perpendicular to the substrate. Notice however that the increase in the thickness when going from 5T to 6T to 7T-containing molecules (0.4–0.5 nm) is very close to the length of one thiophene unit (about 0.4 nm). That suggests that the conjugated segment is indeed perpendicular to the substrate plane. The fact that in this series the thickness of the layers is only slightly larger than the length of the conjugated segment suggests that one PEO group of each molecule lies flat on the substrate (this is probably favored by the ionic character of the mica surface), while the other one is strongly tilted relative to the substrate normal. 2. When comparing sexithiophene systems symmetrically substituted with PEO segments of increasing length (compounds 1a, 1b, 1c), one observes a steady increase in the layer thickness (marked with the broken line in Fig. 3). Clearly, the conformation with the PEO groups flat on the surface and very strongly tilted versus the surface normal, which is proposed above for the compounds with the shortest PEO group (2b and 1a), does not hold for their counterpart with longer PEO groups. Most probably, the tilt angle of those groups relative to the surface normal becomes lower as their length increases. 3. The conformation of the hexyl groups could be inferred from the comparison of the compounds of series 3 and 4. In compound 3a the length of the fully extended molecule is 5.1 nm, and the mean thickness of the layers is 4.7 ± 0.1 nm. For compound 4a the layers are 4.0 ± 0.1 nm thick, with the length of the fully extended molecule being about 4.5 nm. Therefore, those two molecules stand nearly perpendicular to the substrate. The 0.7 nm increase in thickness of the layers from 4a to 3a is very close to the length of the hexyl group ˚ indicating that the hexyl end group is almost (8 A), perpendicular to the substrate. This is confirmed by the comparisons between molecules 3b and 4b and between molecules 3c and 4c: in each case, an increase in layer thickness between 0.6 and 0.8 ± 0.1 nm is observed. 4. Mono-substituted sexithiophene molecules seem to assemble very differently than disubstituted ones: compound 1a (and 2b), with two (EO)5 end groups, forms layers 2.7 ± 0.1 nm thick, whereas compound 4a, with only one (EO)5 end group, shows layers 4.0 ± 0.1 nm thick. The model of packing of the molecules of series 3 and 4 is therefore a quasi-perpendicular orientation of the full molecules on the substrate, with no strong tilt of the PEO groups.

1. In the symmetric molecules 2a, 2b, and 2c, the number of thiophene units is 5, 6 and 7, respectively, and the length of the ethyleneoxide groups is constant. The mean heights of the first layer in that series are 2.2, 2.7 and 3.1 nm (±0.1 nm in each case), respectively. The estimated total length of the molecules is about 6.2, 6.6 and 7.0 nm, with the length of the conjugated segments being close to 2.0, 2.4 and 2.8 nm, for 2a, 2b, and 2c, respectively. It is therefore clear that the molecules do not stand fully extended

Globally, these results indicate that: (i) the conjugated segments tend to orient perpendicular to the substrate plane, a behavior that is identical to that of the sexithiophene molecules in the vacuum-sublimed films [18], indicating that the same driving force for assembly, i.e., ␲–␲ interactions, is at work in both cases; (ii) non polar substituents, e.g., hexyl groups, also orient perpendicular to the substrate. Such orientation is consistent with the fact that the alkyl groups and the conjugated segments, being non-polar, do not have a strong tendency to

Fig. 3. Comparison between the thickness of the first layer on mica (as measured with AFM) and the length of the molecules in their fully extended conformation (as calculated with molecular modeling). The grey circle actually belongs to two different series of black circles and empty circles. Numbers refer to the molecules detailed in Scheme 1.

M. Surin et al. / Synthetic Metals 147 (2004) 67–72

interact with the polar mica surface; (iii) the behavior of the PEO groups is more complex. The shortest PEO segments appear to be able to interact with the mica surface, whereas the orientation of longer segments is different. To further investigate the influence of the substrate on the supramolecular organization, thin deposits were generated onto graphite, which is an apolar surface. For symmetric molecules (series 1 and 2), straight fibrils are formed; this is illustrated in Fig. 2b, corresponding to deposits of 2b on graphite. Fig. 2b shows bundle of fibrils. The single fibrils are straight, from few hundred nm to ␮m long, typically about 1 nm-high and their width correlates well with the computed length of the fully extended molecule (the measured width is 7 ± 1 nm and the length of the fully extended molecule is 6.6 nm). This suggests that the thin fibrils are one-moleculewide ␲-stacks. The width of the large fibrils is a multiple of the length of the molecule in its fully extended configuration. They therefore correspond to the lateral aggregation of a few thin fibrils. From the AFM data in combination with the molecular modeling simulations, we have suggested that the fibrillar structures are made of a conjugated core of parallel ␲-stacked ˚ with ethyoligothiophenes (interchain distance: 3.7–3.9 A) lene oxide groups on each side of the fibrillar axis [19]. Therefore, conjugated chains are perpendicular to the fibril axis, edge-on over the substrate. The length of the ribbons indicates that such ␲-stacking can extend over micrometers, i.e., thousands of molecules. This type of 1D assembly is radically different from the morphology observed for the same compounds on mica. It is interesting to note that the ribbons tend to be aligned along three directions at an angle of 120◦ to each other (Fig. 2b). This arrangement is reminiscent of the three-fold symmetry of the graphite surface, indicating that interactions with the substrate also play an important role in the chain assembly. Note that, for deposits of asymmetric molecules (series 3 and 4) on graphite, a different morphology has been observed, i.e., monolayers with the molecules lying parallel, with the conjugated segments face-on over the substrate [20]. In terms of surface polarity, the naturally oxidized surface of silicon is probably intermediate between strongly polar, ionic mica and apolar graphite. The strong interactions involving the PEO substituents with mica on one hand and the conjugated units with graphite on the other hand, might therefore not play such a strong role in defining the morphology of the deposits formed on silicon. Indeed, symmetric compounds with short PEO groups are found to organize into fibrillar assemblies (instead of islands or flat monolayers). Interestingly, while achiral compounds (such as 1a) form flat featureless fibrils (Fig. 4a), the assemblies built from their chiral counterpart (here compound 2b) all appear as lefthanded helical objects (Fig. 4b). Clearly, the presence of a stereocenter in the molecule induces the formation of chiral assemblies in the solid-state. Somehow surprisingly, we have recently observed that the compound with inverted chirality (i.e., the R,R counterpart of 2b, which is S,S ) assembles into

71

Fig. 4. AFM phase images on naturally oxidized silicon of (a) 1a (5.0 ␮m × 5.0 ␮m) and (b) 2b (1.0 ␮m × 1.0 ␮m). Only 2b shows lefthanded helical aggregates.

helical fibrils that are also left-handed [19]. This suggests that the formation of those chiral objects is not only due to the chiral nature of the building blocks, but also to some other (surface) effect that has to be understood.

4. Conclusion Even though the structural information obtained from the AFM data is quite qualitative, it is nevertheless important because the transport processes in organic FETs take place in a very thin region in contact with the dielectric (up to few nm above). Information on the molecular orientation within that region is therefore essential for the understanding of the device performances. The present studies of the self-assembly of oligothiophenes end-capped with oligoethyleneoxide tails have shown that the interchain interactions (␲–␲ stacking between conjugated segments) and the molecule-substrate interactions are both at work, and lead to well-defined nanostructures, from 2D ultrathin regular layers (on mica) to 1D fibrillar objects (on graphite). Moreover, left-handed helical

72

M. Surin et al. / Synthetic Metals 147 (2004) 67–72

fibrils are found for thin deposits of chiral molecule 2b on the naturally oxidized silicon surface.

Acknowledgements

[6] [7]

The collaboration between Mons, Eindhoven, and Durham has been conducted within the European Commission Training and Mobility of Researchers Network LAMINATE (Large Area Molecular electronics Involving a Novel Approach to Training and Education – Contract Number HPRN-CT-2000-00135). Research in Mons has been conducted within the framework of the Belgian Science Policy Interuniversity Attraction Poles Program (PAI V/3) and the European Science Foundation “Structuring, Manipulation, Analysis and Reactive Transformation of Nanostructures (SMARTON)” program. Research in Mons is also supported by the European Commission, the Government of the R´egion Wallonne (Phasing Out – Hainaut) and the Belgian National Fund for Scientific Research FNRS/FRFC. M.S. acknowledges the F.R.I.A. (Belgium) for a doctoral scholarship. Financial support from the Engineering and Physical Sciences Research Council and the University of Durham for, interalia, infrastructure, spectroscopic and analytical facilities is gratefully acknowledged.

[8]

[9]

[10]

[11] [12] [13] [14] [15]

References [1] N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Science 258 (1992) 1474; G. Yu, J.C. Hummelen, F. Wudl, A.J. Heeger, Science 270 (1995) 1789; J.J.M. Halls, C.A. Walsh, N.C. Greenham, E.A. Marseglia, R.H. Friend, Nature 376 (1995) 498. [2] J.H. Bourroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. MacKay, R.H. Friend, P.L. Burn, A.B. Holmes, Nature 347 (1990) 539. [3] F. Garnier, R. Hajlaoui, A. Yassar, P. Srivastava, Science 265 (1994) 1684; Y.-Y. Lin, D.J. Gundlach, S.F. Nelson, T.N. Jackson, IEEE Trans. Electron Dev. 44 (1997) 1325. [4] A.R. Brown, A. Pomp, C.M. Hart, D.M. De Leeuw, Science 270 (1995) 972; H. Sirringhaus, N. Tessler, R.H. Friend, Science 280 (1998) 1741; H. Sirringhaus, P.J. Brown, R.H. Friend, M.N. Nielsen, K. Bechgaard, B.M.W. Langeveld-Voss, A.J.H. Spiering, R.A.J. Janssen, E.W. Meijer, P. Herwig, D.M. De Leeuw, Nature 401 (1999) 685. [5] K. M¨ullen, G. Wegner, Electronic Materials: The Oligomer Approach, VCH, Weinheim, 1998; G. Horowitz, Adv. Mater. 10 (1998) 365;

[16]

[17]

[18]

[19]

[20]

A.J. Lovinger, H.E. Katz, A. Dodabalapur, Chem. Mater. 10 (1998) 3275; C.D. Dimitrakopoulos, P.R.L. Malenfant, Adv. Mater. 14 (2002) 99; H.E. Katz, Z. Bao, J. Phys. Chem. B 104 (2000) 671. D. Fichou (Ed.), Handbook of Oligo- and Polythiophenes, WileyVCH, 1999. A. Gesqui`ere, M.M.S. Abdel-Mottaleb, S. De Feyter, F.C. De Schryver, F.S. Schoonbeek, J.H. van Esch, R.M. Kellog, B.L. Feringa, A. Calderone, R. Lazzaroni, J.L. Br´edas, Langmuir 16 (2000) 10385; D.B.A. Rep, R. Roelfsema, J.H. van Esch, F.S. Schoonbeek, R.M. Kellog, B.L. Feringa, T.T.M. Palstra, T.M. Klapwijk, Adv. Mater. 12 (2000) 563. J.S. Moore, J. Zhang, J. Angew. Chem. Int. Ed. Engl. 31 (1992) 922; J.S. Moore, J. Zhang, Angew. Chem. Int. Ed. 38 (1999) 1393; W.B. Davies, W.A. Svec, M.A. Ratner, M.R. Wasielewski, Nature 396 (1998) 60. D.M. Kolb, G.E. Engelmann, Angew. Chem. Int. 39 (2000) 922; H. Engelkamp, S. Middelbeek, R.M.J. Nolte, Science 284 (1999) 785. A.P.H.J. Schenning, A.F.M. Kilbinger, F. Biscarini, M. Cavallini, H.J. Cooper, P.J. Derrick, W.J. Feast, R. Lazzaroni, Ph. Lecl`ere, L.A. McDonnel, E.W. Meijer, S.C.J. Meskers, J. Am. Chem. Soc. 124 (2002) 1269. A.F.M. Kilbinger, W.J. Feast, J. Mater. Chem. 10 (2000) 1777. O. Henze, D. Parker, W.J. Feast, J. Mater. Chem. 13 (2003) 1269. A.F.M. Kilbinger, A.P.H.J. Schenning, F. Goldoni, W.J. Feast, E.W. Meijer, J. Am. Chem. Soc. 122 (2000) 1820. K. Fa¨ıd, M. Fr´echette, M. Ranger, L. Mazerolle, I. L´evesque, M. Leclerc, Chem. Mater. 7 (1995) 1390. A. Yassar, G. Horowitz, P. Valat, V. Wintgens, M. Hmyene, F. Deloffre, P. Srivastava, P. Lang, F. Garnier, J. Phys. Chem. 99 (1995) 9155 (See, for a solid state spectrum of ␣-6T) ; M. Muccini, E. Lunedei, C. Taliani, D. Beljonne, J. Cornil, J.L. Br´edas, J. Chem. Phys. 109 (1998) 10513. A.F.M. Kilbinger, H.J. Cooper, L.A. Mc Donnel, W.J. Feast, P.J. Derrick, A.P.H.J. Schenning, E.W. Meijer, Chem. Comm. 10 (2000) 383. Molecular Modeling has been performed using the Universal Force Field available in the Cerius2 package by Accelrys. This force field accurately describes the geometry of oligothiophenes in the solidstate and the geometry and torsional behavior of saturated chains such as alkyl or PEO. The optimization procedure was carried out using Conjugate Gradient method, with RMS force criterion set to ˚ 10−3 kcal/mol A. F. Biscarini, R. Zamboni, P. Samor`ı, P. Ostoja, C. Taliani, Phys. Rev. B 52 (1995) 14868; A.J. Lovinger, D.D. Davis, A. Dodabalapur, H.E. Katz, Chem. Mater. 8 (1996) 2836. Ph. Lecl`ere, M. Surin, R. Lazzaroni, A.F.M. Kilbinger, O. Henze, P. Jonkheijm, F. Biscarini, M. Cavallini, W.J. Feast, E.W. Meijer, A.P.H.J. Schenning, J. Mater. Chem. 14 (2004) 1959. Ph. Lecl`ere, M. Surin, P. Viville, R. Lazzaroni, A.F.M. Kilbinger, O. Henze, M. Cavallini, F. Biscarini, W.J. Feast, A.P.H.J. Schenning, E.W. Meijer, Chem. Mater. (Special Issue on Organic Electronics), in press.