Introduction
Aggregation of pi-conjugated molecules is relevant because the functional properties and the electronic interactions between building blocks can easily be modulated by varying the temperature, solvent polarity, and concentration [1]. Self-assembling molecules can be exploited to generate ordered aggregates, which is relevant for both fundamental and applied research. For example, the performance of organic semiconducting molecules in optoelectronic applications depends on the functional properties of the individual molecules and on their mutual orientations in the solid-state, which can be tuned in solution, during the early stages of aggregation.
Small molecules and polymers have pros and cons in regard to their characterization and applications. Small molecules present smaller variability between batches and are easier to purify and characterize, however, polymers generate larger conjugation lengths. Therefore, small molecules represent a better system to study H- and J-aggregation, while polymeric molecules present better properties for some optoelectronic applications. Indeed, H- and J-aggregates were firstly studied in dye assemblies, which often form these aggregates depending on the relative alignment of the transition dipole moments in adjacent molecules. In an H-aggregate, the intramolecular stacking is predominantly face-to-face, while in J-aggregates the stacking is predominantly head-to-tail [2]. J-aggregates were originally exploited in photographic processes or to modulate light signals in optical communication devices [3]. Currently the ultimate goal is tuning the solid-state functional properties of molecules and their mutual orientations [2]. Thus, the study of H-J aggregation contributes to understanding the role of molecular packing and effect on the materials photovoltaic performance. H- and J-aggregation strongly modify the optical absorption and fluorescence features, which has important consequences for the oscillator strengths of the transitions from the ground to the excited states (S0->S1 transitions), and the energies thereof [2]. H-aggregates exhibit blue-shifted absorption spectra in respect to the absorption of the monomer, and are subradiant. On the other hand J-aggregates exhibit the opposite behavior, red-shifted absorption spectra (in respect to the monomer) and are superradiant [4].
The concept of H-J aggregation was expanded by Spano et al. to analyze films of polythiophenes, in order to perform structure-function studies [5]. Particularly for polythiophenes, H- and J-aggregates coexist in the form of “H-J aggregates”, and the contribution of each mode differs in every practical situation [5]. For structured absorption-fluorescence spectra, the ratio of the first two vibronic peak intensities provides further information, with H(J)-aggregates showing a decrease (increase) in this ratio with increasing excitonic coupling, while the ratio of the 0–0 to 0–1 emission intensities (decreases) with disorder and increases (decreases) with increasing temperature. In absorption and emission spectra, values smaller than 1 in the A1/A2 oscillator strength ratios indicate significant inter-chain coupling (characteristic of H-aggregates) in the aggregates [4]. Indeed, these emission ratios are limiting cases that provide a framework which allow to interpret absorption/emission in more complex morphologies, such as herringbone packing in oligo(phenylene vinylene)s, oligothiophenes and polyacene crystals, as well as the polymorphic packing arrangements observed in carotenoids [4]. Zhu et al. [6] used this concept to study the molecular ordering in solution, of a hydrophilic, thermo-responsive polythiophene, with ethylene oxide side groups, using absorption in solution and synchrotron X-ray scattering to track co-facial stacking (i.e. [0 1 0] ordering) and cofacial molecular stacking (i.e. [98] ordering). The well-defined structuring of both absorption and fluorescence allowed comparing the 0–0/0–1 ratio in order to estimate the [0 1 0] ordering.
Structured spectra is also generated by nanofibers or thin films, in which case it is possible to gain understanding on the exciton coupling present (i.e. intra- or inter-chain), as shown in previous studies on poly-3-hexylthiophene (P3HT), one of the most studied polymers for organic solar cells applications [7], [8], [9].
Besides absorption and fluorescence spectra, the excitation spectrum also provides information on H-J like aggregation of water-soluble polymers and small molecules. However, to the best of our knowledge, this method has been scarcely reported in literature. Deng et al., [10] observed that an increase in the concentration in aqueous solutions of lignosulfonates generates a distortion in the fluorescence excitation spectrum, without modifying the fluorescence emission spectrum. In an analogous study, we used the excitation spectra to study the solution concentration-driven aggregation of cationic polythiophenes (CPTs) with hydrogen-bonding (H-bonding) capabilities, as a function of the side-chain length and the polarity and H-bonding capacity of the solvent [11].
The excitation spectra has shown to be informative on H-J aggregation of small molecules, because it is capable of detecting the spectral response to pi-pi stacking of aromatic groups [10]. This criterion has also been used in studies using small molecules, such as a near-infrared dye, as a function of concentration and solvent [12].
In the solid-state, the analysis of the morphology and/or fluorescence of films deposited onto mica (using atomic force microscopy (AFM) and fluorescence microscopy), are a useful approaches to study the impact of solvent dependent, solution- and solid-state properties, of cationic molecules, as shown previously for small [13], [3], [14], [15], [16] and polymeric molecules, either unconjugated [17] or –conjugated [18]. From these, the study by Yao et al. [13] is particularly relevant for the present work, since it deals with the tuning of J-aggregation of a pseudoisocyanine dye at mica/water interfaces due to addition of 5% of an organic solvent (either 1-propanol or DI) in aqueous solutions. AFM and fluorescence microscopy showed that the morphology and fluorescence of films deposited onto mica, indeed correlate with the spectroscopic data.
Besides the morphology and fluorescence of films deposited onto mica, the surface free energy (SFE) has proven to correlate with solid-state properties of spin-coated films, and devices including them. For example, the SFE impacts the morphology, miscibility and segregation between adjacent layers, or layers and electrodes in organic solar cells (OSCs) [19], [20], [21]. For example, a difference of around 10 mN/m in the SFE between layers (29.1 and 41.1 mN/m) promotes a poor miscibility, producing a slightly larger phase-separated film morphology [20], [22], [23]. However, when this difference decreases to around 2.5 mN/m (29.1 and 31.6 mN/m) penetration and diffusion of [6,6]-Phenyl-C71-butyric acid methyl ester (PC70BM) into the polymer region is promoted [19], [20], [22], [23]. SFE also relates to the adhesive properties of the constituent layers of an OSC, impacting the mechanical stability of the device [24], and it is also known to impact on the short circuit current and fill factor of these devices [25]. SFE has been used specifically to co-optimize the adhesion and power conversion efficiency by performing surface treatments of the buffer layer [26].
For semiconducting polymeric films, the SFE (together with energy level and electrical conductivity) can be modified (i) by means of molecular structure, e.g. by changing the polymer backbone and lengths of alkyl side chains [27]; (ii) by doping processes, e.g. increasing the SFE of poly(3-hexylthiophene) (P3HT) films by doping [28]; and, in an easier way, (iii) by doing a Judicious selection of the polarity of the solvent mixture, which allows modulation of self-assembled aggregates (e.g. vesicles, rods etc.), as well as the optical properties of conjugated polymers and CPEs, as reviewed by Houston et al. [19]. Variations in solvent polarity modify the relation between polarity and rigidity of both backbone and side chains of CPEs in solution, inducing conformational changes [29].
Also, co-solvents allow gaining information on H-bonding interactions. For example, methanol–dimethylformamide (DMF) mixtures interfer with polymer-polymer and polymer-solvent H-bonding interactions, generating a nanoribbon morphology of poly(ethylene oxide) (PEO) [30]. This occurs because mixed solvents generate preferential solvation of certain parts of the polymer, such as backbone and attached functional group, in certain component of the binary mixture [31]. Also, as reviewed by McDowell et al., [32] co-solvents (also known as “additives” in the field of OSCs) provide an extra level of control over the two main parameters that control the OSC formation during solution processing: (i) thermodynamic parameters in solution, such as the solubility of donor and acceptor materials in the solvent(s), ease of crystallization/aggregation, and the mutual interactions between the solvents and the donor and acceptor solutes, and (ii) drying kinetics parameters, such as the vapor pressure of the solvents, and the deposition conditions that collectively define the drying kinetics of the mixture [32]. In our previous contribution [33] we used this approach, when analyzing the effect of imidazolium methylation on the SFE (estimated by means of contact angle goniometry) of imidazolium CPTs spin-coated onto plasma-activated glass, using water or a 50:50 v/v 1,4-dioxane-water (W-DI) mixture as processing solvents. It was observed that imidazolium methylation decreases the total SFE (γS) in ≈1mN/m, probably due to a more ordered structure, as suggested by previous studies on pentacene films which showed, by means of contact angle goniometry, that decreased film order increases γS in less than 1 mN/m [34]. It is important to highlight that this result of SFE correlated with results from X-ray diffraction (GIXD), synchrotron X-ray diffraction (XRD) and FTIR). In our previous work it was also observed that DI decreases γS in 0.2–0.4 mN/m, increasing the polar contribution (γSp) and decreasing the dispersive contribution (γSd) in 1–2 mN/m [33]. This information was discussed in terms of solvation and polymeric conformation within the films. Despite the cited contributions, and others using Kelvin probe force microscopy (KPFM) or ultraviolet photoelectron spectroscopy (UPS) (e.g. [35] and its references), there are not yet available clear guidelines with respect to the structure of CPTs for designing high performance polythiophene-based interfacial layer materials [19].
This work presents a study on the enhancement of J-like aggregation in solution- and solid-state, of a CPT due to the presence of 1,4-dioxane as cosolvent, in solution and solid-state.
The CPT, labelled PT1, is functionalized with isothiouronium units, which provide charge-assisted H-bonding (CAHB) capabilities, and a high sensitivity to the polarity and H-bonding capacity of the solvent. Water or a 1,4-dioxane-water 50:50 v/v mixture (W-DI) were used either as media or as processing solvent for deposition, because of their clearly different polarity/H-bonding capacity.
In solution, J-like aggregation enhancement of PT1 was revealed by fluorescence excitation spectroscopy, while in the solid-state, PT1 was deposited onto three anionic substrates: (i) drop-casted films onto glass were observed by means of fluorescence spectroscopy; (ii) spin-coated films onto plasma-activated glass were used to estimate the SFE by means of contact angle goniometry; and (iii) drop-casted films onto mica were used to observe the morphology by means of AFM.
To the best of our knowledge: (a) the use of the fluorescence excitation spectra to gain insight on J-like aggregation has been only reported by Deng et al. [10] and our previous work [11], for water-soluble, conjugated fluorophore polymers, (besides studies on small molecules [12]); (b) there are not reports on the correlation between solution and solid-state J-like aggregation enhancement of a CPT due to the polarity/H-bonding capacity of the media/processing solvent, and (c) there are not reports on the effect of J-like aggregation on the SFE of films made of CPEs.
