Synthesis of Donor–Acceptor Copolymers Derived from Diketopyrrolopyrrole and Fluorene via Eco-Friendly Direct Arylation: Nonlinear Optical Properties, Transient Absorption Spectroscopy, and Theoretical Modeling

Abstract

A series of PFDPP copolymers based on fluorene (F) and diketopyrrolopyrrole (DPP) monomers were synthesized via direct arylation polycondensation using Fagnou conditions which involved palladium acetate as catalyst (a gradual catalyst addition of three different percentages were used), potassium carbonate as the base, and neodecanoic acid in N, N-dimethylacetamide. This synthesis provides a low cost compared with traditional methods of transition-metal-catalyzed polymerization. Among the different amounts of catalyst used in the present work, 12% was optimal because it gave the highest reaction yield (81.5%) and one of the highest molecular weights (Mn = 13.8 KDa). Copolymers’ chemical structures, molecular weight distributions, and optical and thermal properties were analyzed. The linear optical properties of PFDPP copolymers resulted very similarly independently to the catalyst amounts used in the synthesis of the PFDPP copolymers: two absorptions bands distinctive of donor–acceptor copolymers, Stokes shifts of 41 nm, a good quantum yield of fluorescence around 47%, and an optical bandgap of 1.7 eV were determined. Electronic nonlinearities were observed in these copolymers with a relatively high two-photon absorption cross-section of 621 GM at 950 nm. The dynamics of excited states and aggregation effects were studied in solutions, nanoparticles, and films of PFDPP. Theoretical calculations modeled the ground-state structures of the (PFDPP)_n copolymers with n = 1 to 4 units, determining the charge distribution by the electrostatic potential and modeling the absorption spectra determining the orbital transitions responsible for the experimentally observed leading bands. Experimental and theoretical structure–properties analysis of these donor–acceptor copolymers allowed finding their best synthesis conditions to use them in optoelectronic applications.

1. Introduction

Over the past decades, donor–acceptor polymers (D–A) comprising alternative arrays of the donor (an electron-rich moiety) and the acceptor (an electron-deficient moiety) units have emerged as alternative materials for a diverse range of applications such as photodetectors [1] and organic field-effect transistors [2]. These semiconducting polymers stand out due to their tunable absorption spectrum, light emission with high quantum yield, solubility in organic solvents, low bandgap, and electronic properties [3]. Owing to these characteristics, D–A polymers can be easily processed to form thin, lightweight, flexible, and uniform films [4]. In addition, water-soluble nanoparticles have been successfully fabricated from semiconductor polymers through a nanoprecipitation approach with DSPE-PEG-azide as the encapsulation matrix [5].

To obtain these kinds of macromolecules, traditional methods of transition-metal-catalyzed polymerization have been employed, such as Stille [6], Suzuki [7], Negishi [8], and Kumada [9]. However, with these synthesis routes, the long-awaited method to generate semiconductor polymers at low cost has not yet been reached since these polymers are part of the total cost of the production of optoelectronic devices such as organic solar cells [10].

Given the considerations mentioned above, a synthesis method without expensive reagents, toxic organometallic byproducts, and fewer synthesis steps is needed. Recent advances in polymer catalysis research have extended to palladium-catalyzed coupling reactions. In this regard, the eco-friendly direct arylation polycondensation (DArP), consisting of a metal-promoted C-H activation and an aryl halide to generate C-C bonds, has taken the advantage over the other methodologies [11]. Additionally, with this chemical reaction, a reduction in manufacturing cost of 35% can be achieved [12]. Among all the moieties used in the design of chromophores and plastic semiconductors, the diketopyrrolopyrrole (DPP) unit is widespread because it has high charge carrier mobilities and a semi-crystalline nature [13]. The DPP monomers can be used as electron-deficient (acceptor) type building blocks in D–A semiconductor polymers. This electron-deficient character of the DPP building block, ascribed to its fused lactam rings, induces outstanding light-harvesting properties, and push–pull-type conjugated polymers containing DPP can deliver excellent photovoltaic performance in organic solar cells [14,15]. Likewise, research on fluorene-based polymers has had continuous growth because of its diverse applications such as electroluminescence devices [16], wavelength converters [17], optical imaging, and phototherapy [18]. The development of gas sensors based on conjugated DPP polymers, which are sensitive to NO₂ and NH₃, has also been reported [19,20]. Bitellurophene-based DPP polymer (PDPPBTe) is another sensing material used to detect Br₂ [21].

On the other hand, organic materials such as D–A semiconductor polymers that possess π electronic delocalization have also become attractive due to their large nonlinear susceptibilities, which lead to useful optical effects such as intense nonlinear absorption, namely two-photon absorption (TPA) [22]. Efficient TPA in combination with some other molecular functionalities such as high quantum yield of fluorescence or energy transfer has been of great interest for its promising applications in cell imaging and two-photon photodynamic therapy [23]. DPP and fluorene moieties, with their inherent electron-deficient and electron-rich characters, have been studied as TPA materials through the correlation with their molecular symmetry, electronic character, conjugation length π, and geometry [24], but the combination of DPP and fluorene in D–A copolymer has not been fully explored [25]. Thus, using these two moieties to construct D–A semiconductor copolymers comprising enhanced optical nonlinearities deserves further investigation.

In this contribution, polycondensations via eco-friendly direct arylation of copolymers derived from DPP were carried out under Fagnou conditions (Scheme 1), varying the amount of palladium acetate (catalyst). To assess the photophysical properties and the structure–properties relationship of this group of polymers achievable using the title synthesis method, the so-obtained copolymers were analyzed by ¹H NMR, TGA-DSC, GPC, FT-IR, and UV-vis. Lifetime fluorescence and quantum yield were also measured. Furthermore, aggregation properties were studied in organic nanoparticles and solid films. Nonlinear optical properties and excited states’ dynamics were investigated using the Z-scan technique and transient absorption spectroscopy.

2. Experimental Section

2.1. Materials and Equipment

All reagents and chemical compounds were bought from Aldrich Chemical Co. and used in commercial quality. 3,6-Bis(5-bromo-2-thienyl)-2,5-bis(2-hexyldecyl)-2,5-dihydro-pyrrolo [3,4-c]pyrrole-1,4-dione 98%, 2,2′-(9,9-dioctyl-9H-fluorene-2,7-diyl) bisthiophene 97%, potassium carbonate 99%, palladium acetate (Pd(OAc)₂) 99.9%, neo-decanoic acid 99%, N,N-dimethylacetamide (DMA) 99.8%, potassium pivalate (PivOK) and hexadecyltrimethyl-ammonium bromide (CTAB). All reactions were performed under a nitrogen atmosphere.

FT-IR spectra (3500–500 cm⁻¹) were obtained on a Perkin Elmer Spectrum 400 spectrophotometer (using ATR reflectance mode). Polymeric products were characterized in CDCl₃ by Proton Nuclear Magnetic Resonance (¹H NMR) using a Bruker Avance III HD 500 MHz spectrometer. TGA analysis was performed with an SDT Q600 V8.2 Build 100 equipment, mode: Standard TGA-DSC with sapphire as a calibration material and an alumina crucible. Nitrogen gas was used at 100 mL/min, equilibrated at 50 °C. The molecular weights distributions of the copolymers were carried out by gel permeation chromatography (GPC) on a chromatograph Alliance 2695, using polystyrene standards for calibration and two PL gel lineal columns; measurements were performed at 30 °C in THF, and an injection volume of 25 µL with the Universal V4.2E program.

The ultraviolet–visible (UV–vis) absorption spectra were measured on a spectrophotometer Genesys 10 s UV-vis with a spectral window of 190 to 1100 nm (1 cm path length). A standard Z-scan setup measured the TPA cross-sections with a femtosecond Ti: sapphire regenerative amplifier laser (Libra, Coherent Inc., Santa Clara, CA, USA) delivered pulses of 80 fs at 800 nm (1 kHz repetition rate) that were used to pump optical parametric amplifier (TOPAs, from Light Conversion, Vilnius, Lithuania) to generate pulsed radiation at 950 nm. Transient absorption experiments were performed using the Harpia spectrometer from Light Conversion. To get the NPs size distribution, the dynamic light scattering (DLS) technique with the Nano Zetasizer Malvern system (Model ZEN 3600) and scanning electron microscopy (model JSM-7800F from Jeol) were used.

2.2. Synthesis and Characterization

Synthesis of PFDPP Polymers (General Procedure)

3,6-bis(5-Bromo-2-thienyl)-2,5-bis(2-hexyldecyl)-2,5-dihydro-pyrrolo [3,4-c]pyrrole-1,4-dione, 2,2′-(9,9-dioctyl-9H-fluorene-2,7-diyl) bisthiophene, potassium carbonate, and neodecanoic acid were added to a ball flask. The oxygen was removed from the solution by passing a flow of nitrogen for 10 min, after which palladium acetate was added in three fractions every 2 h from a 0.1 M solution in N,N-dimethylacetamide in anhydrous conditions. An additional volume of DMA was injected to complete a 5 mL reaction volume. The flask was closed, and the reaction mixture was stirred for 24 h in a 70 °C oil bath. The reaction was cooled at room temperature, and the polymer was precipitated in methanol. The polymer was washed on a Soxhlet apparatus with methanol, acetone, hexane, and chloroform. The maximum reaction yield was obtained using a 12% catalyst.

Table 1. Summary of the parameters used in the synthesis for PFDPP copolymers.

Polymer	DPP (g, mmol)	Fluorene (g, mmol)	Neodecanic Acid (mg, mmol)	Pd(OAc)2 a (% mol, mmol × 10−2)	K2CO3 (mg, mol)	Time (h)	Yield (% b, % c)
PFDPP-1	0.20, 0.22	0.12, 0.22	0.01, 0.06	8, 0.017	0.09, 0.65	24	15, 64.2
PFDPP-2	0.20, 0.22	0.12, 0.22	0.01, 0.06	12, 0.026	0.09, 0.65	24	0.4, 81.5
PFDPP-3	0.25, 0.27	0.15, 0.27	0.01, 0.06	16, 0.034	0.09, 0.65	24	0.1, 41.0

^a Solution 0.01 M, ^b Hexane fraction, ^c Chloroform fraction.

summarizes the so-obtained PFDPP copolymers.

2.3. Elaborating Polymer Nanoparticles by the Reprecipitation Method

Nanoparticles of PFDPP-1 were obtained using the reprecipitation methodology as follows: 2 mg of this polymer was dissolved in 2 mL of THF, then 0.16 mL was injected quickly and under sonication in 8 mL of an aqueous solution containing CTAB (0.8 mM) as surfactant.

2.4. Elaborating Organic Thin Films

Glass substrates were cleaned using soapy water, ethanol, and acetone in an ultrasonic apparatus. PFDPP-1 polymer solutions of 1 mg/mL were prepared using chlorobenzene, chloroform, and THF as solvents. To form two types of films, the mixtures were then deposited by the spin-coating method at 1600 rpm, heat-treated at 80 °C for 20 min in an oven, and allowed to cool down to room temperature.

3. Results and Discussions

3.1. FT-IR

To identify the functional groups that each PFDPP copolymer contains, Fourier transform infrared (FT-IR) spectroscopy was employed. The obtained spectra are shown in Figure 1, and all-principal bands are described in Table S1 in the supplementary information. It can be appreciated that the polymers have similar absorption peaks, which could be due to the use of DPP as an acceptor unit in the main chain, and the difference in catalyst loads may result in changes in the intensity of the partial absorption peak.

In particular, we observed the following vibration modes (see Figure 1): =C-H st at 3066 cm⁻¹, C-H₃ ν_as at 2959 cm⁻¹, C=O st at 1658 cm⁻¹ is a typical stretching mode of alkylating lactam groups [26], C=C ring st of aromatic backbone at 1507 cm⁻¹, C-N st in the DPP ring at 1302 cm⁻¹ [27], δ(C-H) in the thiophene plane at 1068 cm⁻¹, C-H out of the plane of the thiophene ring at 731 cm⁻¹, and deformations C-H out of the plane of the thiophene ring at 625 cm⁻¹ [28].

3.2. ¹H NMR

¹H NMR spectroscopy was used to confirm the structures of all the polymers. Figure 2 shows the ¹H NMR spectra (CDCl₃, 500 MHz) of the PFDPP-1 polymer, the spectra corresponding to PFDPP-2,3 are presented in the supplementary information (Figures S1 and S2). Spectra were plotted in the Mestre Nova program. The broad signals that appear at 0.80 ppm (-CH₃), 1.20 ppm, 1.18 pm, 1.51 ppm, and 1.9 ppm (-CH₂ (α)-fluorene, -CH-DPP) correspond to the aliphatic chains of the D–A polymer, the signal at 4.05 ppm are the hydrogens of the chains adjacent to the DPP’s nitrogen atoms. The signals from 7.24 to 7.61 ppm are the hydrogens of the thiophene and fluorene rings [29], and at 8.85 ppm are the hydrogens of DPP thiophene [30]. Making a critical analysis of the aromatic and aliphatic region, we can see a correlation between the hydrogens of the monomeric units of the polymer.

3.3. Differential Scanning Calorimetry and Thermogravimetric Analysis (DSC-TGA)

TGA determined the thermal stability of the copolymers, and the results are shown in Figure 3. According to the TGA analysis, all polymers retained 60% residual weights when the temperature increased to 400 °C. It has been reported that some copolymers based on fluorene and diketopyrrolopyrrole have decomposition temperatures (T_d) of 402 °C [31].

DSC was used to obtain the glass transition (T_g), melting (T_m), and crystallization (T_c) temperatures of the PFDPP polymers by the heating and cooling methods, the results are shown in Figure 4 and

Table 2. Summary of thermal properties of PFDPP polymers obtained by DSC and TGA.

Polymer	Tg (°C)	Tm (°C)	ΔHm (J/g)	TC (°C)	ΔHC (J/g)	Td (°C)	Wc (%)
PFDPP-1	65	146	1.27	142	1.26	443	60.8
PFDPP-2	64	159	1.59	152	2.67	437	69.8
PFDPP-3	63	157	0.89	151	1.04	425	65.7

T_m and T_c are melting, and crystallization temperatures, respectively, and W_c is weight loss.

. The Tg values are around 60 °C for PFDPP polymers. The first polymer (PFDPP-1) exhibits a slightly higher glass transition temperature value than the other polymers, which means it has a greater crystalline character. The melting temperature is very similar (see Table 2) for PFDPP-2 and PFDPP-3, which is explained by the similitude of their molecular weights. The grade of crystallinity for each polymer was calculated using the reported enthalpy of fusion (ΔH_m) [32], and the results were 1 for PFDPP-1 and PFDPP-3 and 2 for PFDPP-2. In the DSC heating traces, no peaks were found in the temperature range between 50 and 140 °C, and we can infer that the morphology of the PFDPP films is stable in this range of temperatures [33]. Using DArP with microwave heating at 120 °C for 1 h, 5% mol of catalyst Pd(OAc)₂ and PivOK 1.5 equiv., a similar DPP-F polymer derived from DPP and fluorene was obtained with 5% weight loss temperature (T_d5%) of 378 °C, and it does not present clear melting and crystallization peaks, an exothermic peak around 101–103 °C upon cooling, was probably due to a glass transition [34]. From all this information, we concluded that all of these PFDPP polymers are thermally suitable, for instance, for applications in organic solar cells since they can withstand thermal treatments.

3.4. Gel Permeation Chromatography (GPC)

Using the GPC technique calibrated relative to polystyrene standards, the average molecular weights in number and weight (Mn and Mw) and the polydispersity index (PDI) of the polymers synthesized in the present work were determined and are shown in

Table 3. Distribution of molecular weights determined by GPC.

Polymer	Mn (g/mol)	Mw (g/mol)	PDI
PFDPP-1	11,347 ± 3.8%	34,812 ± 0.5%	3.07 ± 3.3%
PFDPP-2	13,848 ± 0.7%	43,803 ± 1.4%	3.16 ± 2.2%
PFDPP-3	13,875 ± 0.7%	43,011 ± 0.3%	3.10 ± 0.4%

, and the supplementary information (Figures S3–S5). The molecular weights had the smallest value, Mw = 34,812 g/mol for PFDPP-1 with the lowest catalyst load, and the highest value, Mw = 43,803 g/mol for PFDPP-2, which used a 12% catalyst load. The PDIs were 3.1 in all PFDPP polymers. These results are higher than the Suzuki synthesis for similar derivatives of DPP and fluorene, which are 20,000 g/mol [35]. Compared with other DPP-based polymers obtained by DArP, they showed Mn values of 7.9–12.6 kg mol⁻¹ and the polydispersity (Mw/Mn) of 2.2–2.8 as determined by GPC. A similar DPP-F polymer obtained by microwave-assisted DArP achieved 12.6 kg/mol, and PDI of 2.6 [32], and a PTFDPP polymer synthesized using Suzuki polycondensation showed an Mn value of 9.138 Kg/mol, Mw = 17,841 Kg/mol and PDI = 1.95 [36]; our PFDPP polymers obtained by DArP using conventional heating present slightly higher values than the before-mentioned [37]. Using DArP with a more complex catalyst than palladium acetate, such as trans-bis(acetate)bis[o-(dio-otolylphosphino)benzyl]dipalladium(II) and phosphine ligands such as tris(2-methoxyphenyl), a molecular weight Mw = 44 Kg/mol was found [38].

3.5. Photophysical Properties: Absorption and Emission Spectra

The absorption and emission spectra of the PFDPP polymers in chloroform solutions are shown in Figure 5a, while the main features are summarized in

Table 4. Optical parameters of PFDPP polymers.

Polymer	Absorbance λmax (nm)	Emission λmax (nm)	Stokes Shift (nm)	Φ (%)	Lifetime Fluorescence (ns)	Ꜫ (M−1cm−1)	${\mathit{E}}_{\mathit{g}}^{\mathit{o}\mathit{p}\mathit{t}} \left(\mathbf{eV}\right)$	${\mathit{E}}_{\mathit{g}}^{\mathit{e}\mathit{l}\mathit{e}}$ (eV)
PFDPP-1	412, 655	696	41	43.8	0.78	486,786	1.74	2.32
PFDPP-2	408, 655	696	41	47.4	0.77	635,623	1.76	2.40
PFDPP-3	407, 655	696	41	47.9	0.70	736,762	1.77	2.52

${\mathit{E}}_{\mathit{g}}^{\mathit{o}\mathit{p}\mathit{t}}$ and ${\mathit{E}}_{\mathit{g}}^{\mathit{e}\mathit{l}\mathit{e}}$ are the optical and electrochemical bandgaps, respectively, and Ꜫ is the molar absorption coefficient.

. The three samples showed the characteristic absorption bands observed for donor–acceptor polymers. The first absorption band is located at short wavelengths of 370−420 nm, and the second strong absorption band is situated between 620 and 680 nm, which is due to a π − π* transition and internal charge transfer effect [39]. In addition, PFDPP-1 and PFDPP-2 show a shoulder at 740 nm. This is further investigated through our theoretical calculations presented below. We noted that the molar absorptivity coefficient increased as the catalyst load increased in the synthesis processes, varying nearly monotonically from 4.8 × 10⁵ M⁻¹ cm⁻¹ in PFDPP-1 to 7.3 × 10⁵ M⁻¹ cm⁻¹ in PFDPP-3 according to the increase in molecular weight (from 11.3 KDa to 13.8 KDa). Despite the notorious differences in molar absorptivity, the three copolymers showed the same optical bandgaps (Eg) calculated using the Tauc method [40], as shown in Table 4. The linear absorption coefficient (α) was obtained from PFDPP-1 in solid film resulting in a value of 2.41 × 10⁵ cm⁻¹. The relative intensities of the 0–0 and 0–1 absorption bands (A_0–0/A_0–1 > 1) indicate that these polymers tend to have intrachain J-aggregation in solution. The emission spectra for all the polymers are identical exhibiting vibronic features replicating the absorption spectra for the transitions 0–0 and 0–1, presenting a Stokes shift of 41 nm. The maximum wavelength of emission λ_max obtained in the present work is 655 nm, greater than the 500 nm reported in analogous polymer synthesis by Suzuki [32] and equal to the value of 655 nm reported by direct arylation with complex catalysts and phosphines [33].

For the PFDPP-1 polymer, the absorption and emission spectra were obtained in different solvents (chloroform, chlorobenzene, tetrahydrofuran, and toluene), as shown in Figure 5b. We found that the ground state is insensitive to the solvent effect because the solvent polarity does not lead to significant shifts in the absorption bands. The excited states are barely sensitive to the solvent, as can be seen in the small changes in the emission intensity bands (around 696 and 765 nm), although no hypso- or bathochromic shifts were found.

The fluorescence quantum yields (Φ) of the PFDPP polymers were measured in chloroform at 10⁻⁵ M, observing small increments with the catalyst load. The Φ obtained were from 43.8 to 47.9% for PFDPP-1 and PFDPP-3, respectively. Fluorescence lifetimes (FLTs) for the PFDPP polymers were determined using the time-correlated single-photon counting (TCSPC) method. Interestingly, the fluorescence decay for PFDPP-3 resulted differently. Concerning PFDPP-1 and PFDPP-2, being that the FLT of PFDPP-3 is the fastest and nearly mono-exponential, with a time constant of 0.70 ns, in contrast, the PFDPP-1 and PFDPP-2 copolymers have a rapid decay with a time constant of 0.77–0.78 ns accompanied by a slow component of decay of 1.16 ns (see Table S2 in Supplementary Information) as it is presented in Figure 6.

To determine the distinct features introduced by the donor and acceptor moieties in the polymers studied here, normalized absorption and emission spectra of fluorene monomer (F), diketopyrrolopyrrole monomer (DPP), and PFDPP-1 polymer in chloroform were performed, as shown in Figure 7. The absorption maximum of F is located at 355 nm, DPP has two absorption maxima at 529 nm and 567 nm, and PFDPP-1 also presents two absorption bands at 412 nm and 655 nm. The absorption band of the acceptor DPP with its characteristics transitions 0–0 and 0–1 remain in the PFDPP-1 spectrum but with a displacement due to conjugation; by comparing the UV–vis spectrum of the polymer with its monomer F, a red-shifting of 300 nm is due to the π-conjugated polymer backbone [41]. Similar polymers, for example, the reported DPP-F polymer, exhibit a more extended wavelength absorption in the range of 500–700 nm because of the intramolecular charge transfer (CT) interactions [32]. In another PTFDPP polymer, the λ_max reported were 374 nm, 599 nm, and 640 nm [34]. Additionally, the emission maxima of F are located at 386 nm, and 408 nm, and DPP shows an emission spectrum with two maxima at 585 nm and 634 nm; the features of these monomer spectra also appear in PFDPP-1 but are red-shifting to 699 nm and 765 nm. The optical bandgaps of these PFDPP polymers, around 1.7–1.8 eV, are close to the reported 1.74 eV for DPP-F polymer [32] and 1.71 for PTFDPP [34].

3.6. Electrochemical Properties

The electrochemical study provides information on the position of the energy levels of HOMO and LUMO of semiconductor polymers. This property of the PFDPP-1 polymer was studied using cyclic voltammetry (CV) in drop cast film on ITO-glass in acetonitrile with tetrabutylammonium hexafluorophosphate as an electrolyte 0.1 M. Scan rate: 100 mVs⁻¹, T = 20 °C under nitrogen atmosphere and the potentials were calibrated with ferrocene/ferrocenium (Fc/Fc⁺). The corresponding CV diagrams are shown in Figure 8. PFDPP-1 exhibit quasi-reversible oxidative processes with onset potential E_onset(ox) at 0.82 eV. It can be appreciated that the reduction is irreversible. The onset potentials were used for calculating ionization potentials (E_HOMO and E_LUMO) values vs. Ag/Ag⁺ as the reference electrode. The next equations were used in the HOMO–LUMO calculus:

E_HOMO = −E_ox/onset (vs. Fc/Fc⁺) + 5.1 [eV]

E_LUMO = −E_red/onset (vs. Fc/Fc⁺) + 5.1 [eV]

$$E_g^{ele}=|{\mathrm{E}}_{\mathrm{HOMO}}-{\mathrm{E}}_{\mathrm{LUMO}}|$$

The values obtained for PFDPP-1 were: E_HOMO = −5.92 eV, E_LUMO = −3.6 eV and $E_g^{ele}$ = 2.32 eV. All $E_g^{ele}$ are shown in Table 4, and Figures S6 and S7 show the other voltamperograms. The electrochemical bandgaps were determined during processes of reduction (LUMO) or oxidation (HOMO) in a molecular species. In a photoexcitation, the exciton formed has a Coulombic interaction, which makes the optical bandgap in general smaller than the electrical one. Further, the effect of solvation on the electrochemical bandgap in almost all cases is invariably larger. This solvent contribution makes the difference.

The exciton binding energy (E_b) is another critical parameter in determining the use of some semiconducting polymers in devices. The E_b and power conversion energy (PCE) of the organic cells are inversely correlated; a polymer with a lower E_b can more efficiently produce electric current from the absorbed photons [36]. The E_b value of PFDPP-1, the difference between the electronic and optical bandgap energies, is 0.67 eV, smaller than the difference between the LUMO energy values of the copolymer and PC₇₁BM, 0.79 eV. These copolymers may efficiently dissociate excitons at the donor/acceptor interface.

3.7. Organic Nanoparticles and Films

The properties of the copolymers were also evaluated in the suspension of nanoparticles (NPs) and films. Nanoparticles of PFDPP-1 copolymer were prepared as mentioned in Section 2.3. The precipitation method here employed produced NPs of 45 nm on average, with somewhat significant dispersion measured by dynamic light scattering (DLS) shown in Figure 9a. Another analysis of the size distribution of these NPs was carried out from several micrographs obtained by field emission scanning electron microscopy (FESEM) using ImageJ software, getting an average diameter size of 62 nm, Figure 9b. The absorption spectra of PFDPP-1 in the form of NPs and from films deposited in a glass substrate using different solvents are shown in Figure 10. Compared with the corresponding absorption spectrum in solution, in the case of the film, a bathochromic shift occurs due to intramolecular interactions in the solid state. In the case of nanoparticles, a broadening of the absorption band can be seen. In NPs, the relative intensities A_0–0/A_0–1 < 1 are contrary to the solution where A_0–0/A0–1 > 1, which suggests that in solid-phase dominates the intrachain H-aggregation. For the films, the dominant aggregation type depends on the solvent used for film deposition in the spin-coating method. For instance, in the case of films deposited from THF solutions, the H-aggregation dominates. Still, the J-aggregation is the distinctive feature for films processed from chloroform and chlorobenzene solutions. The origin of this difference may be the physicochemical properties of THF, such as the greater dipole moment and dielectric constant and the low boiling point. The absorption coefficients obtained from films in different solvents are very similar, with values around 2.41 × 10⁵ cm⁻¹, so it is inferred that no representative changes in α are generated when the solvent varies.

3.8. Transient Absorption

To know the dynamics of the excited states in the title D–A polymers, ultrafast transient absorption (TA) spectroscopy was performed [42]. These studies were carried out in organic solutions, in suspensions of NPs, and in films of the PFDPP-1 copolymer. Figure 11a presents the TA signal for the THF solution of PFDPP-1 under excitation at 410 nm. This excitation corresponds to the generation of excited species from the donor part of PFDPP-1. Distinctive spectral regions are observed. The first region corresponds to a negative signal between approximately 520 and 770 nm assigned to photobleaching. A redshift of this band is observed at early times related to exciton diffusion to lower energy states. Notice that the band has a structure that reproduces the vibronic 0–0 and 0–1 structure observed in stationary absorption spectroscopy at approximately 615 and 660 nm, respectively. The TA spectrum also resolves the weak shoulder observed in the stationary absorption spectrum at 750 nm. The second region corresponds to a positive absorption band for wavelengths larger than 770 nm in which apparent relative peaks at 900 and 960 nm can be observed. The third region in the TA spectra corresponds to wavelengths shorten than 520 nm, which coincides with the tail of the first band of the stationary absorption producing the absorption of excited states instead of photobleaching. According to these observations, the wavelengths of 520 nm and 770 nm conform to two isosbestic points such that the wavelengths shorter than 520 nm are sufficiently energetic to produce transitions of the type S₂ → S_n. In comparison, wavelengths less energetic than 770 nm should be assigned to S₁ → S_n transitions. Both the negative and positive bands of TA are characterized practically with the same decay dynamics, thus suggesting that all the observed transitions originate from the first excited singlet state. Generally, the characteristic decay times for both photobleaching and absorption of excited states are approximately a few ps, as shown in Figure 11c, with similar singlet decay times reported for other low-bandgap polymers [43].

Figure 11b presents a comparison of the TA signals for PFDPP-1 copolymer in the solutions, in the form of NPs and solid films for the specific delay time τ of 1 ps, for which most of the thermal decay has occurred. The figure also depicts the stationary absorption spectra for the three cases in which the characteristic absorption bands for D–A polymers are observed, with a strong absorption band centered at around 650 and a weaker peak at 415 nm. The first observation is that when PFDPP-1 is processed into NPs, the relative amplitude of the 0–1 absorption transition becomes higher than the intensity of the transition 0–0 compared to the case of PFDPP-1 in solution. These spectral changes are not accompanied by significant spectral shifting upon aggregation of the polymer into NPs. Thus, the increase in the 0–1 absorption band might be correlated to the formation of H-type aggregates and possibly to a reduction in J-aggregation. It is also observed that the solid film shares with the NPs the same spectroscopic characteristics in both stationary and time-resolved regimens. Figure 11c shows that the temporal dynamics of bleaching (TA signal at 600 nm) and the absorption of excited states (TA signal at 900 nm) for films and NPs of PFDPP-1 are notoriously shortened compared with the polymer in solution; as expected, this indicates that in the solid-phase the intermolecular interaction and the conformation of the polymer reduce the lifetime of the excited state S₁.

3.9. Nonlinear Optical Properties

Z−scan technique with femtosecond open aperture at wavelength 950 nm was employed to measure two-photon absorption cross-sections (σ_TPA) of PFDPP polymers (Figure 12) [44]. σ_TPA was calculated by using the equation σ_TPA = ħωβ/CN_A, where ħω represents the photon energy, C is the solute molar concentration, and N_A is the Avogadro number. A 1 mM concentration of the copolymer in chloroform was used. The Z−scan setup was previously calibrated with the standard CS₂. The solid green line in the inset of Figure 12 represents the fit obtained and, using the previously mentioned equation, from which we can determine the cross-section. σ_TPA of 621 GM was measured; Figure 12 shows typical Z−scan traces obtained with this polymer. We attribute the nonlinear behavior of these polymeric materials mainly to the increase in electronic conjugation due to the donor–acceptor behavior in the polymeric backbone and the increase in conjugation by the bicyclic dilactam of DPP when coupled with fluorene, as occurs in other D–A polymers [45]. Copolymers that include DPP fragments, such as the present work, have cross-section values of 205 and 209 GM [46]. Polyfluorene-based conjugated polymers with cross-section values of 305 and 891 GM showed to be applicable for two-photon live cell imaging [47].

3.10. Theoretical Calculations

All calculations were performed within the linear combination of Gaussian type orbitals density functional theory (LCGTO−DFT) framework within the generalized gradient approximation (GGA), using the B3LYP functional [48] and employing the deMon2k software [49,50]. All electron DZVP [51] basis sets in combination with the A2 auxiliary functions sets were used. The Coulomb potential was evaluated using the variational fitting procedure developed by Dunlap, Connolly, and Sabin [52]. The search for the minimum structures was performed by building geometries in visualization software and optimizing them without symmetry constraints using delocalized internal coordinates [53]. This theoretical model has been shown to accurately calculate organic systems [36,54].

Figure 13 presents the optimized geometries of the (PFDPP)_n with n = 1, 2, and 4 units. We performed test calculations of the system with n = 11 units and found a similar trend; however, we did not fully optimize this system, so we have not reported it here. We observed that the systems present a slight out of planarity at the extremes of the chains and that as the systems increase in size, the curvature is more notable. This curvature can be explained by the dipole–dipole intramolecular interaction of hydrogen atoms adjacent to the sulfur atom, with distances ranging from 2.65 to 2.98 Angstroms. The calculated HOMO–LUMO gap for the models is 2.02 eV, 1.77 eV, and 1.70 eV, for n = 1, 2, and 4, respectively. We note that as we increase the number of units, the bandgap decreases, converging to around 1.7 eV, which agrees, although sub-estimates, the experimental electronic bandgap, which ranges from 2.32 to 2.52 eV (see Table 4). The relative difference between the two should be alleviated by considering the solvent effects, which we are investigating and will report in a forthcoming paper.

The charge distribution was calculated by plotting the molecular electrostatic potential (MEP), which is color-coded, showing electron-rich regions in red color and electron-deficient regions in yellow color, as shown in Figure 14. Fluorene concentrates the charge as an electron donor, and DPP loses charge being an electron acceptor.

We calculate the UV–vis absorption spectra for the PFDPP polymer using time-dependent density functional theory (TD-DFT) [55] with the B3LYP hybrid functional [48] and the GEN-A2* auxiliary basis set. The spectrum presents two major transitions around 350 nm and 502 nm, which are characteristics of D–A polymers, as shown in Figure 15. We found that the theoretical spectra are shifted to lower wavelengths. According to test calculations modeling, the system with more units shifts the bands to longer wavelengths. To better understand these absorption bands, we plotted the iso-surfaces of the orbitals that mainly contribute to these bands, as shown in Figure 16. The main band corresponds to transitions from the occupied orbitals HOMO, HOMO-1, HOMO-2, and HOMO-3 to the unoccupied LUMO, LUMO+1, and LUMO+2, as in

Table 5. Absorption transitions for each absorption band.

from	to	Energy Gap (eV)	λ (nm)
The absorption band at 350 nm		Oscillator Strength = 0.70
HOMO-2	LUMO	3.19	389
HOMO-1	LUMO + 1	3.45	359
HOMO	LUMO + 1	2.86	434
HOMO	LUMO + 2	3.39	366
The absorption band at 502 nm		Oscillator Strength = 2.47
HOMO-2	LUMO	3.19	389
HOMO-1	LUMO	2.59	479
HOMO	LUMO	2.00	620
HOMO	LUMO + 2	3.39	366

. As observed in the molecular orbitals, electronic density is transferred from the DPP to fluorene upon photoexcitation. This transfer is interesting as fluorene has an electronic charge concentration, as shown in the electrostatic potentials of Figure 14.

4. Conclusions

Our work shows that copolymers derived from F and DPP can be synthesized by using eco-friendly direct arylation polycondensation (Fagnou conditions). Out of the three catalyst charges used, the optimal is 12% since it obtained the highest reaction yield, and it is also slightly the one with the highest molecular weight. All polymers (in their chloroform fraction) were soluble in most of the organic solvents and had two absorption bands which are characteristics of D–A polymers. Their photophysical properties (absorption, emission, fluorescence half-lives, quantum efficiency, band gaps) are very similar when comparing the three copolymers. A broadening of the absorption spectrum is obtained by generating nanoparticles around 50 nm in suspension. The Z-scan technique analyzed the polymers’ nonlinear optical properties, showing these achieve a maximum of 621 GM. Theoretical calculations showed a trend towards bending as the number of units increases, an intramolecular charge transfer from fluorene towards DPP, and the orbitals responsible for the observed UV–vis main bands were elucidated. The thermal stability of these semiconducting polymers together with their optical (linear and nonlinear) and electrochemical properties suggest that they are suitable for optoelectronic devices such as organic solar cells and organic field-effect transistors and other potential applications of these organic materials, for instance, optical imaging, and photodynamic therapy.