Cyclopentadithiophene derivatives: a step towards an understanding of thiophene copolymer excited state deactivation pathways

Abstract

We have carried out the excited state characterization of cyclopentadithiophene (CPDT) and its didodecyl-substituted derivative (C₁₂CPDT) to obtain detailed insights into their photophysical pathways. These are the cis analogues of α,α′-bithiophene (α2), which normally exists in the more stable s-trans geometry. We report absorption, fluorescence, phosphorescence, and triplet-singlet difference spectra, together with quantum yields (fluorescence, phosphorescence and singlet oxygen). Fast time-resolved fluorescence and transient absorption techniques (ps-TCSPC and fs-TA) were used to characterize the initial excited state dynamics of the CPDTs and α2. In general, the two CPDTs show similar spectral and photophysical properties to α2, although phosphorescence is absent in α2 (the trans conformer). From TDDFT calculations it is shown that the large Stokes shift (SS) observed for CPDT and C₁₂CPDT (planar in S₀ and S₁) is due to a bond length change in the C–C bond connecting the two thiophene units, which decreases from 146 to 137 pm. With α2 this change is accompanied by a change in the dihedral angle (25.5° in S₀ to 0.2° in S₁).

---

Introduction

Following the ground-breaking work of Tang and Vanslyke,^1,2 the interest in the field of organic materials for organic light-emitting diodes (OLEDs) has continued to rise for more than 30 years.³ One of the ongoing and relevant issues receiving particular attention is how to overcome the problem that 3/4 of the electrically generated energy is dissipated as heat via the formation of triplets.³ This is essentially due to the fact that with the great majority of the systems derived from the most relevant families of conjugated organic polymers (polythiophenes, polyphenylenevinylenes, polyfluorenes) phosphorescence is generally absent. The search for phosphorescent polymers and oligomers, and, in general, small organic molecules with short natural phosphorescent lifetimes, is, therefore, of utmost relevance.

The α,α′-bithiophene moiety is a key constituent of many conjugated organic polymers and copolymers, and its conformational behavior, both in the ground and excited state, has received considerable attention, mainly from a theoretical perspective.^4–13 The s-trans–gauche structure is generally considered to be the dominant conformation, having a ground state geometry with an ensemble of conformers (with small differences in the barrier widths between them), but with the more stable ones having reported torsion angles for the anti-like conformation varying from 133,¹⁴ to 135,¹⁵ 142.2 and 147,¹² 146,^16,17 150¹¹ and 160⁷ degrees. The s-cis planar conformer is less stable, although it has been suggested from theory and terahertz spectroscopy to coexist in the gas phase or in cyclohexane solution in 46% concentration,⁶ with the cis to trans ratio being temperature dependent.^13,15

The decay mechanisms, including ultra-fast dynamics, involved in other similar systems (oligomers and polymers of the most representative structures, for example terpyridines, p-phenylene vinynele and fluorene) have been the subject of recent studies.^18–20

In CPDT, α2 is forced to adopt the cis conformer. This is investigated experimentally, together with its didodecyl-substituted derivative (C₁₂CPDT), with the aim of rationalizing their properties in the excited state and obtaining insights into the physical chemistry of the system. Although a previous nanosecond laser flash photolysis study has been published on CPDT,²¹ and includes some spectral and photophysical data, detailed understanding of its excited state properties, and in particular the change between the ground state (GS) and excited state (ES) geometries, demands a comprehensive photophysical study, including the behavior of its alkylated derivative, C₁₂CPDT.

Experimental section

CPDT and C₁₂CPDT were prepared following literature procedures.^22–24 In particular CPDT was synthesized using a modified synthetic route first described by Asawapirom et al.²² summarized in Scheme 1.

Scheme 1 Synthetic route, here followed, to CPDT (C5) and C₁₂CPDT (C6).

4H-Cyclopenta[1,2-b:5,4-b′]dithiophene (CPDT, C5)

4H-Cyclopenta[1,2-b:5,4-b′]dithiophene-4-carboxylic acid (7.68 g, 34.61 mmol) and copper (1.65 g, 25.89 mmol) were dissolved in freshly distilled quinoline (50 mL). The reaction mixture was heated to 245 °C for 45 minutes. After cooling to room temperature, the suspension was poured into a mixture of ice and hydrochloric acid. After adding diethyl ether, the mixture was filtered off and the filtrate washed with 10% hydrochloric acid, water, and sodium carbonate solution. The organic phase was dried over magnesium sulfate and the solvent was removed under reduced pressure. 4H-Cyclopenta[1,2-b:5,4-b′]dithiophene was purified by silica gel column chromatography (eluent: hexane) to afford the greyish product (4.29 g, 69.6%).

¹H NMR (400 MHz, C₂D₂Cl₄) δ [ppm] = 7.14 (d, J = 4.9 Hz, 2H, Ar-H), 7.05 (d, J = 4.9 Hz, 2H, Ar-H), 3.48 (s, 2H, Ar2-CH2). ¹³C NMR (101 MHz, C₂D₂Cl₄) δ [ppm] = 150.1 (Ar-R), 138.8 (Ar-R), 125.0 (Ar-H), 123.4 (Ar-H), 32.2 (Ar2-CH2). GC-MS: m/z [M⁺] = 178.0.

4,4-Didodecyl-4H-cyclopenta[1,2-b:5,4-b′]dithiophene (C₁₂CPDT, C6)

4H-Cyclopenta[1,2-b:5,4-b′]dithiophene (0.85 g, 4.75 mmol), potassium hydroxide (1.07 g, 18.99 mmol), and potassium iodide (0.03 g, 0.19 mmol) were suspended in dimethyl sulfoxide (30 mL). 1-Bromododecane (2.37 g, 9.53 mmol) was added and the reaction was stirred at room temperature for 14 hours. After adding water, the reaction mixture was poured into brine. The mixture was extracted with diisopropyl ether and the combined organic phase was dried over magnesium sulphate and the solvent was removed under reduced pressure. The desired product was obtained as a yellowish oil (1.21 g, 49.1%) after silica gel column chromatography (eluent: hexane).

¹H NMR (600 MHz, CDCl₃) δ [ppm] = 7.14 (d, J = 4.9 Hz, 2H, Ar-H), 6.93 (d, J = 4.9 Hz, 2H, Ar-H), 1.95–1.74 (m, 4H, C–(CH₂)₂), 1.43 (dt, J = 14.7, 7.2 Hz, 4H, C–(CH₂–CH₂)₂), 1.38–1.02 (m, 32H, CH₂), 0.98–0.92 (m, 4H, CH₂–CH₃), 0.90–0.87 (m, 6H, CH₃). ¹³C NMR (151 MHz, CDCl₃) δ [ppm] = 158.8 (Ar-R), 137.0 (Ar-R), 124.8 (Ar-H), 122.0 (Ar-H), 53.3 (C–(CH₂)₂), 37.7 (CH₂), 32.8 (CH₂), 31.8 (CH₂), 29.5 (CH₂), 29.4 (CH₂), 29.3 (CH₂), 29.2 (CH₂), 28.7 (CH₂), 28.01 (CH₂), 24.4 (CH₂), 22.5 (CH₂), 13.9 (CH₃). GC-MS: m/z [M⁺] = 515.4.

All the solvents used were of spectroscopic grade and no further purification was needed.

Photophysical characterization

Absorption and fluorescence spectra were recorded on a Cary 5000 UV-Vis-NIR instrument, equipped with an integrating sphere for solid state diffuse reflectance measurements, and a Horiba-Jobin-Ivon Fluorolog 3-22 fluorimeter, respectively. Phosphorescence measurements were made in methylcyclohexane glasses at 77 K and used the Horiba-Jobin-Ivon Fluorolog 3-22 spectrometer in phosphorescence mode with a pulsed excitation source. All of the fluorescence and phosphorescence spectra were corrected for the wavelength response of the system.

Fluorescence quantum yields in solution were measured by the comparative method using α2 (α,α′-bithiophene) in dioxane solution (ϕ_F = 0.017)²⁵ as a standard, while the phosphorescence quantum yields were determined using benzophenone (ϕ_Ph = 0.84)²⁶ in ethanol solution as a standard. The fluorescence quantum yield for CPDT in the solid state was obtained by the absolute method using a Hamamatsu absolute PL quantum yield spectrometer Quantaurus C11347 (integrating sphere).

Fluorescence decays were measured using a home-built picosecond time correlated single photon counting, TCSPC, apparatus (3 ps time resolution) described elsewhere.²⁷ The fluorescence decays and the instrumental response function (IRF) were collected using a time scale of 1024 (or 4096) channels, until 5 × 10³ counts at maximum were reached. Deconvolution of the fluorescence decay curves was performed using the modulating function method, as implemented by G. Striker in the SAND program, and previously reported in the literature.²⁸

The experimental setup used to obtain the triplet–triplet absorption spectra has been described elsewhere.³¹

The experimental setup for ultrafast spectroscopic and kinetic measurements consists of a broadband (350–1600 nm) HELIOS pump–probe femtosecond transient absorption spectrometer from Ultrafast Systems, equipped with an amplified femtosecond Spectra-Physics Solstice-100F laser (800 nm central wavelength displaying a pulse width of 128 fs at 1 kHz repetition rate) coupled with a Spectra-Physics TOPAS Prime F optical parametric amplifier (195–22 000 nm) for pulse pump generation. Probe light in the visible range was generated by passing a small portion of the 800 nm light from the Solstice-100F laser through a computerized optical delay (with a time window up to 8 ns) and focusing in a vertical translating CaF₂ crystal to generate a white-light continuum in the 340–750 nm range. All measurements were obtained in a 2 mm quartz cuvette with optical densities lower than 0.3 at the pump excitation wavelength. The instrumental response function of the system was assumed to be equal to that of the pump–probe cross correlation determined from the measurement of the instantaneous stimulated Raman signal from the pure solvent (in a 2 mm cuvette). Typical values for the IRF of the system were found to be better than 250 fs. To avoid photodegradation the solutions were stirred during the experiments or in movement using a motorized translating sample holder. Transient absorption data were analyzed using the Surface Xplorer PRO program from Ultrafast Systems and the global analysis of the data (from which the lifetimes of the observed transients were obtained) was performed using Glotaran software.²⁹

TDDFT calculations

All calculations were of DFT type and carried out using GAMESS-US³⁰ version R3. A range corrected CAMB3LYP³¹ functional, with 65% HF exact exchange at long range and 19% at short range, was used in the calculations of both ground- and excited-states. TDDFT calculations, with similar functionals, were used to probe the excited-state potential energy surface (PES).³² The solvent was included using the polarizable continuum model with the solvation model density to add corrections for cavitation, dispersion, and solvent structure. In TDDFT calculation of Franck–Condon (FC) excitations the dielectric constant of the solvent was split into a “bulk” component and a fast component, which is essentially the square of the refractive index. In “adiabatic” conditions only the static dielectric constant is used. Hessians were computed by numerical differentiation of analytically computed first derivatives. A 6-31G** basis set was used in either DFT or TDDFT calculations. An “ROHF” type wavefunction was used in the triplet DFT geometry optimizations to prevent spin contamination. CAMB3LYP slightly overestimates excitations with a 0.98 scale factor applied to the reported values. Excitations were corrected for Zero Point Vibrational Energy by, essentially, the frequency of the relaxing vibrational mode ±0.15 eV.

Results and discussion

The structures and acronyms of the compounds investigated, together with that of the parent (trans) α2, are given in Scheme 2.

Scheme 2 Structures of unsubstituted (CPDT) and didodecyl-substituted cyclopentadithiophene (C₁₂CPDT), together with α,α′-bithiophene (α2) in its trans conformation.

The electronic spectra of α2 have previously been described in various solvents and temperatures, together with those of its naphthalene and cyano derivatives.^25,33–35 These consist of broad absorption and emission bands. However, in the subsequent oligothiophenes (with n = 3–7, etc.) although the absorption remains broad and structureless, the emission clearly becomes vibronically resolved.²⁵ The absorption and emission spectra of CPDT and C₁₂CPDT in methylcyclohexane solution are presented in Fig. 1. Similar spectroscopic features were, in general, observed for these compounds with, however, two distinct differences. The first is the red-shift in both the absorption and emission spectra of C₁₂CPDT relative to CPDT; the second is that the spectra of CPDT are vibronically resolved, which is particularly noticeable in the absorption (Fig. 1), and fluorescence excitation spectra (Fig. S1 in ESI†). It is worth noting that the excitation spectra for both compounds match the absorption spectra, clearly confirming the purity of the compounds (Fig. S1, ESI†). The vibronic structure of the absorption spectra of CPDT should be further compared with that of α2. Observation of the absorption spectra of CPDT and of α2^25,35,36 shows that the former is vibronically resolved whereas the latter is devoid of vibronic resolution. In the case of α2 this was attributed to the existence of an ensemble of conformers at room temperature that, as the temperature was lowered, lead to a bigger average planarity of the existing conformers.²⁵ In the case of CPDT the planar geometry induced by the CH₂ bridge connecting the two thiophene units leads to a rigid structure (in the ground S₀ state) and therefore to a vibronic structure similar to that observed by α2 at low temperature.²⁵ In the case of C₁₂CPDT a similar pattern (to CPDT) occurs although with a lower vibronic resolution. It is also worth noting that, as will be discussed in light of the TDDTF calculations, CPDT and C₁₂CPDT present equal planar structures in both the ground and first singlet excited states and therefore the differences in the two (S₀ and S₁) are found in their different bond distances/lengths.

Fig. 1 Normalized absorption and fluorescence emission spectra for (A) CPDT and C₁₂CPDT in methylcyclohexane solution and (B) for CPDT in the solid state (powder) at 293 K.

Fig. 1B presents the absorption (obtained by diffuse reflectance) and fluorescence emission spectra for CPDT as a powder in the solid state. A broader absorption was found in the solid state compared with the solution spectra. This broadening and the blue-shift in the absorption maxima (∼248 nm) may suggest the presence of H-aggregates. However, interestingly the fluorescence emission spectra, although red shifted, maintain the spectroscopic features observed in solution, while for CPDT similar fluorescence quantum yields were found in solution (0.010) and in the solid state (0.016), Table 1.

Table 1 Spectroscopic (absorption, fluorescence and phosphorescence emission and triplet absorption maxima) and photophysical data (including fluorescence, ϕ_F, and phosphorescence, ϕ_Ph, quantum yields and lifetimes, τ_F, τ_Ph and τ_T) for the samples investigated in methylcyclohexane (MCH) solution

Compd	λ max (abs) (nm) 293 K	λ max (fluo) (nm) 293 K	λ max (phosph) (nm) 77 K	λ max (T1 → Tn) (nm) 293 K	ϕ F 293 K	τ F (ps) 293 K	ϕ Ph 77K	τ Ph (ms) 77 K	τ T (μs) 293 K	ϕ Δ 293 K
a In the case of C12CPDT the fluorescence decays were obtained from a HPLC purified methanol solution. b Data are from ref. 34; photophysical data in benzene (ϕF and τF) and spectral data (wavelength maxima) in dioxane; NE = non-existent.
CPDT	308	355	610	370	0.01	47	0.004	5.5	26	0.27
C12CPDT^a	316	363	610	370	0.05	166^a	0.004	5.9	64	0.46
α2^b	303	362	NE	385	0.026	46	NE	NE	27	0.96

---

For C₁₂CPDT, it was not possible to obtain absorption or fluorescence spectra in the solid state due to the limited quantity available of this viscous oil-like derivative at 293 K. Table 1 summarizes the spectral and photophysical properties of CPDT and C₁₂CPDT, together with those of the oligomer α2.

The fluorescence decays for CPDT and C₁₂CPDT are presented in Fig. S2 (ESI†). Analysis of the decays of both CPDT and C₁₂CPDT shows that they are, essentially, single exponential.

Information from the fluorescence decays is complemented by analysis of femtosecond transient absorption (fs-TA) data. Femtosecond time-resolved transient difference absorption spectra (fs-TA) for CPDT and C₁₂CPDT were obtained with excitation at 310 nm and collection in the 340–750 nm range (Fig. 2 and Fig. S3 in ESI†). The fs-TA spectra for CPDT and C₁₂CPDT present a positive broad transient absorption band in the 410–700 nm region, with maxima at 487 and 497 nm for CPDT and C₁₂CPDT, respectively, that are attributed to the excited singlet state absorption (ESA), and are accompanied by a negative feature at shorter wavelengths, which, from the similarity with the fluorescence emission bands, is attributed to stimulated emission (SE), see Fig. 1. As the delay time increases the ESA band disappears and concomitantly a new band arises with a maximum at 385 nm (Fig. 3 and Fig. S3 in ESI†), which does not decay completely up to the longest delay time measured. The spectroscopic features of this latter band resemble in shape and maxima the transient triplet–triplet absorption spectra obtained by laser flash photolysis at 266 nm of degassed solutions of the two cyclopentadithiophene derivatives (Fig. S4, ESI†) and is, consequently, attributed to the triplet state absorption.

Fig. 2 (A) Time-resolved transient absorption data for CPDT obtained with λ_exc = 310 nm (pump energy 1 μJ) in aerated dioxane solution at room temperature; together with (B) the representative kinetic traces with fits from the global analysis of the transient absorption data. For a better judgment of the quality of the fits the residuals are also presented.

Fig. 3 Phosphorescence emission spectra (A) and decays (B) for CPDT and C₁₂CPDT in methylcyclohexane solution at 77 K.

In ns-TA data, similar triplet–triplet absorption spectra and maxima were found for CPDT and C₁₂CPDT, with triplet lifetimes (τ_T = 26–64 μs), see Fig. S4 (ESI†) and Table 1. A previous report has shown a similar maximum and lifetime following photoexcitation of CPDT and was attributed to the triplet state since efficient sensitization of the β-carotene triplet state through triplet energy transfer was observed.²¹

The global fit analysis of all the kinetics in the 340–710 nm range (using a sequential model) showed that for CPDT and C₁₂CPDT the experimental traces over the whole spectra-temporal range are well fitted with the sum of three exponentials (see Table 2 together with Fig. 3 and Fig. S3 in ESI†). Initial fast decay transients of 0.98 ps, CPDT, and 0.22 ps, C₁₂CPDT, were seen that are in good agreement with the inertial solvation dynamics times previously reported for Coumarin C153 in dioxane (∼0.35–0.92 ps);³⁷ intermediate decay times of 29 ps for CPDT, and 192 ps for C₁₂CPDT were observed, that in agreement with the fluorescence lifetimes obtained by TCSPC (Table 1 and Fig. S2, ESI†) should be attributed to the singlet excited state deactivation lifetime; finally a long-lived transient (associated with the triplet state) was seen for both compounds. Since the kinetic traces could not be fitted within the monitoring time-window (see Fig. 3 and Fig. S3, ESI†), this time constant was fixed in the analysis to the triplet lifetime obtained by ns-TA (Table 1). The distinct rise of this latter feature (see Fig. 3 and Fig. S3, ESI†), together with the observed negative amplitude values observed in the decay associated spectra (DAS) of the 29 ps (CPDT) and 192 ps (C₁₂CPDT) time constants in the wavelength region where the triplet–triplet absorption occurs (Fig. S5, ESI†), supports the idea that the triplet state is being formed at the expense of the singlet excited state through intersystem crossing.

Table 2 Results of the global fit analysis (lifetimes, τ_i) obtained from the ultrafast time-resolved transient absorption for the cyclopentadithiophene derivatives in dioxane solution (λ_exc = 310 nm) at 293 K

Compd	τ 1 (ps)	τ 2 (ps)	τ 3 (μs)
a This transient lifetime was fixed in the global analysis to the triplet lifetime obtained by ns-TA.
CPDT	0.98	29	26
C12CPDT	0.22	192	64

---

For comparison, the fs-TA spectra for α2 were also obtained in aerated n-hexadecane solutions (see Fig. S6, ESI†), a solvent in which the solvation dynamics time should be absent or cannot compete with the expected excited state decay time constants. Similar spectroscopic features were found for α2 to those observed for the two cyclopentadithiophene derivatives: ESA band centered at 495 nm together with the SE band and the T₁–T_n absorption band at shorter wavelengths. The fs-TA kinetic data for α2 in the 360–680 nm range were well fitted with the sum of three exponentials using a sequential model, Fig. S6 (ESI†). Again, the long-lived transient was fixed in the analysis to the triplet lifetime (27 μs, see Table 1), while the intermediary transient time, 42 ps, is in good agreement with the fluorescence lifetime, 46 ps, and thus is attributed to the oligothiophene quinoidal-type excited state decay. For the α,α′-oligothiophene trimer, α3, the shorter lifetime component observed in the fs-TA data has been attributed to the occurrence of fast conformational relaxation from the less planar Franck–Condon excited state to the relaxed quinoidal-like excited state conformation.³⁸ Detailed analysis of the nature of this fast decay component will be the subject of a forthcoming work.

The triplet state of the cyclopentadithiophene derivatives was further characterized through the determination of the phosphorescence properties. For the cyclopentadithiophene derivatives phosphorescence was observed in frozen methylcyclohexane solution, 77 K, with broad and vibronically structured bands (Fig. 3). This contrasts with the lack of phosphorescence from α2. Indeed, with the oligothiophene series, αns, only the simplest α1 and its corresponding cyano derivative, CNα1, display phosphorescence.^25,33 However, the trimer with the central thiophene replaced by a pyrrole ring also presents phosphorescence.³⁹ Similar phosphorescence quantum yields (ϕ_Ph = 0.004) and lifetimes (τ_Ph = 5.5–5.9 ms) were found for the two derivatives, see Table 1. The phosphorescence spectrum of CPDT is in good agreement with that previously reported^4,21 and subsequently reproduced with theoretical calculations.⁴ The vibronic structure of the phosphorescence spectra has been attributed to the rigid structure of the triplet state,²¹ in line with what was observed for the oligothiophene singlet states, in particular for α2.²⁵ However, it is worth noticing in Fig. 3 the loss of vibronic structure in the case of the C₁₂CPDT, which may indicate a partial departure from planarity in T₁.

Singlet oxygen formation quantum yields (ϕ_Δ) were also obtained following photolysis of aerated methylcyclohexane solutions of the cyclopentadithiophene derivatives. The ϕ_Δ values were determined by plotting the initial phosphorescence intensity of singlet oxygen (at 1270 nm) as a function of the laser dose and comparing the slope with that obtained with biphenyl in cyclohexane as standard (ϕ_Δ = 0.73), see Fig. S7 (ESI†). From Table 1 it can be seen that although the singlet oxygen formation yield is higher for the dodecyl-substituted cyclopentaditiophene (ϕ_Δ = 0.47) than for the unsubstituted CPDT (ϕ_Δ = 0.27), these values are significantly lower than that reported for α2 in benzene solution (ϕ_Δ = 0.96).²⁵ Considering that for the unsubstituted α,α′-oligothiophenes the ϕ_Δ values are a good estimate of the triplet formation quantum yields [displaying unitary singlet oxygen sensitization efficiency yields, S_Δ = (ϕ_Δ/ϕ_T) ≈ 1],²⁵ in the present work these can be considered as lower limits for the ϕ_T values. Thus, from the overall set of photophysical data presented in Table 1 the radiationless channels dominate the deactivation of the S₁ in CPDT and C₁₂CPDT.

We have studied the spectral and photophysical characteristics of the fused ring CPDT as a model for the cis form of α,α′-bithiophene. In addition, we have studied the C₁₂ derivative, where the alkyl chains increase the solubility of the corresponding copolymers containing these or similar structural units. However, we have also had the aim of investigating the potential changes relative to the parent two structures: (i) CPDT that has a rigid structure in the ground (crystal)⁴⁰ and excited states and (ii) α2, that is basically rigid (quinoidal-type structure) in S₁ (and T₁) but that has an increased degree of freedom relative to the C–C central bond connecting the two thiophene units in the ground state. Following this, the detailed spectral and photophysical data of CPDT and C₁₂CPDT are summarized in Scheme 3 in a Jablonski–Perrin diagram and can be further compared with that of α2. In this discussion, we will also consider results from TDFT calculations. The predicted transitions and oscillator strengths obtained from TDDFT calculations are given in Table 3.

Scheme 3 Jablonski–Perrin like diagram showing the excited state deactivation quantum yields (with the fluorescence, ϕ_F, and singlet oxygen sensitization quantum yields, ϕ_Δ, obtained at 293 K while the phosphorescence quantum yields were obtained in frozen methylcyclohexane solution at 77 K) and associated lifetimes (fluorescence, τ_F, and phosphorescence lifetimes, τ_Ph), together with the energies of the lowest singlet and triplet excited states for CPDT and C₁₂CPDT.

Table 3 TDDFT Spectroscopic properties including wavelength maxima, λ (nm), of absorption, emission (fluorescence and phosphorescence) and oscillator strengths (f). For phosphorescence the calculations were of DFT type. All values calculated for DMSO as solvent

Absorption				Emission
				Fluorescence			Phosphorescence
Compd	α2	CPDT	C1CPDT	α2	CPDT	C1CPDT	CPDT	C1CPDT
λ (nm)	300	311	318	358	359	365	657	673
f	0.509	0.443	0.390

---

Comparison of the experimental data in Table 1 with the calculated (from TDDFT and DFT) data in Table 3 shows that the absorption wavelength maxima (S₀–S₁) can be considered basically identical, with a difference of 2–3 nm. Although the predicted triplet emission is overestimated by 50 nm, this translates to just 0.15 eV, well within the expected precision of the CAMB3LYP functional.

Table 4 presents the calculated (TDDFT) bond angles and bond distances for α2, CPDT and C₁CPDT. For a schematic view of the structures see Fig. S8 (ESI†).

Table 4 Bond distances and dihedral angles obtained from TDDFT calculations for α2, CPDT and C₁CPDT (mimetic of C₁₂CPDT). See Fig. S8 (ESI) for atom numbering

S0		S1
Atoms	Angle/bond distance	Atoms	Angle/bond distance
α2
1-3-6-7	25.5°	1-3-6-7	0.2°
3-6	146 pm	3-6	137 pm
2-4/3-5	136 pm	2-4/3-5	142 pm
4-5	142 pm	4-5	139 pm
CPDT
5-3-6-7	0.0°	5-3-6-7	0.4°
3-6	145 pm	3-6	137 pm
2-4/3-5	136 pm	2-4/3-5	142 pm
4-5	142 pm	4-5	139 pm
C1CPDT
5-3-6-7	0.0°	5-3-6-7	0.3°
3-6	145 pm	3-6	137 pm
2-4/3-5	136 pm	2-4/3-5	142 pm
4-5	142 pm	4-5	139 pm

---

From Fig. 1 and the data in Table 1, it can be seen that the spectral characteristics of CPDT (and C₁₂CPDT) closely resemble those of the α-oligothiophene, α2. The absorption wavelength maximum for α2²⁵ in dioxane is 303 nm and the emission is at 362 nm (Δλ = 49 nm), which can be compared with 308 nm and 355 nm (Δλ = 47 nm) for the absorption and emission maxima, respectively, of CPDT in Table 1. This parallel between α2 and CPDT can be extended to the photophysical properties (ϕ_F and τ_F), and, in particular, the significant Stokes shift (Fig. 1), which indicates that even in the rigid CPDT, with a methylene bridge linking the two thiophene units, different geometries exist in the ground and singlet excited states.

Indeed, as with α2, in CPDT the geometry in S₁ is likely to be associated with a planar quinoidal-like structure. The same is not true, however, for the geometry of the ground state, where it is evident from Table 4 that both CPDT and C₁₂CPDT (here replaced by the dimethyl derivative, C₁CPDT) have similar planar geometries in both S₀ and S₁, whereas for α2 this only happens in S₁ since in S₀ a torsional angle of ca. 25.5° is predicted for the trans α,α′-bithiophene, α2. For the ground states, with CPDT a planar structure is found in the crystal state,⁴⁰ while for α2 from both experiment and different levels of calculations dihedral angles from 20° to 47° have been reported (see Introduction section).

As a consequence, for CPDT and C₁₂CPDT the marked differences between the GS (S₀) and ES (S₁) states must be found not in the change in their dihedral angles (identical planar structures in both states), but in the change in their bond distances/lengths. This is confirmed in Table 4, where the bond angles and lengths of CPDT and C₁CPDT are identical in the GS, but there is a decrease in the length of the central C₃–C₆ bond (connecting the two thiophene units) from 146–145 pm (S₀) to 137 pm (S₁) with all three compounds. This indicates a clear change in geometry for all three compounds upon excitation, which explains the huge Stokes shift observed. In previous studies, the large Stokes shifts observed for N-carbazole end-capped oligo-thiophene-phenylenes and planar polyenes, were attributed to bond length reorganization between the ground- and excited-states.^41,42 In the present case the decrease in the C4–C5 bond lengths (142 pm in S₀ and 139 pm in S₁) and the increase in C2-4/C3-5 (136 pm in S₀ and 142 pm in S₁) also imply a more delocalized nature of the π-electrons in S₁ in contrast with a more localized one in S₀, which is compatible with the suggested quinoidal-like structure in S₁. This can be visualized using the density plots of the HOMO and LUMO orbitals for CPDT and C₁CPDT in Fig. 4.

Fig. 4 HOMO and LUMO plots for CPDT and C₁CPDT.

The complete behavior of CPDT and C₁₂CPDT can now be fully understood with the help of the fs-TA data. The tri-exponential analysis of the transient profiles indicates an identical number of species. The longest lifetime in CPDT is clearly associated with the decay of the triplet state, and is in agreement with previous results for this compound.²¹ Although there are slight differences for the two compounds, they are in the same time range. The intermediate transient lifetimes 29 ps and 192 ps are, from comparison with the fluorescence lifetimes (Tables 1 and 2), associated with the decay of the relaxed S₁ state, whereas the initial fast decays can be associated with the solvation dynamics.

It is worth noting the significant increase in the fluorescence lifetime observed going from CPDT, 47 ps, to the C₁₂CPDT derivative, 166 ps. This is mirrored by the increase in the ϕ_F values, CPDT, ϕ_F = 0.01, C₁₂CPDT, ϕ_F = 0.05, and the concomitant decrease in the non-radiative excited state decay processes (k_NR = 21.1 ns⁻¹, CPDT, vs. k_NR = 5.7 ns⁻¹ for C₁₂CPDT). This can be explained by the overall increase in structural rigidity of the CPDT chromophore induced by the introduction of the alkyl substituents that reduces the excited state radiationless energy losses due to decreased interactions with the solvent molecules. This is also mirrored in the initial fast decay transient value in Table 2, which is reduced from 0.98 ps to 0.22 ps from CPDT to C₁₂CPDT.

In summary, the excited state dynamics of the compounds investigated can be described as follows: upon photoexcitation the initially formed excited state (Franck–Condon state) undergoes fast solvent reorientation to a relaxed excited singlet state, which leads to the formation of the long-lived triplet excited state, and its subsequent decay to the ground state, as the main deactivation pathway in these compounds.

Conclusions

A comprehensive photophysical investigation was carried out of the cis conformer of α,α′-bithiophene (α2) in solution using the fused ring derivative CPDT, and its alkylderivative, C₁₂CPDT. A change from 145 pm (S₀) to 137 pm (S₁) in the central C–C bond connecting the two thiophene units is responsible for the observed pronounced Stokes shift. In the case of α2 an additional departure from planarity in S₀ (with a dihedral angle 25.5°) also accompanies this Stokes shift. The spectral and photophysical properties of the cis (CPDT) conformer parallel those of the trans conformer of α2; in particular nonradiative decay is the main channel responsible for the deactivation of the S₁ state. The overall results also show that, except for differences in the phosphorescence behavior, the cis (CPDT) keeps the spectral and photophysical characteristics of the trans α2.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge funding by FEDER (Fundo Europeu de Desenvolvimento Regional) through COMPETE (Programa Operacional Factores de Competitividade). The Coimbra Chemistry Centre is supported by the Fundação para a Ciência e a Tecnologia (FCT), Portuguese Agency for Scientific Research, through the programme UID/QUI/UI0313/2013. CQE is also supported by FCT through the project UID/QUI/00100/2013. This work was performed under the project “SunStorage – Harvesting and storage of solar energy”, with reference POCI-01-0145-FEDER-016387, funded by the European Regional Development Fund (ERDF), through COMPETE 2020 – Operational Programme for Competitiveness and Internationalisation (OPCI), and by national funds, through FCT. The FCT is also acknowledged for a post-doctoral grant to J. Pina (ref. SFRH/BPD/108469/2015). The research leading to these results has received funding from Laserlab-Europe (grant agreement no. 284464, EC's Seventh Framework Programme). We also acknowledge MSc Catarina Pinto (UC) for the help with the purification of C₁₂CPDT by HPLC-DAD.

References

1. C. W. Tang and S. A. Vanslyke, Appl. Phys. Lett., 1987, 51, 913–915.
2. C. W. Tang, S. A. Vanslyke and C. H. Chen, J. Appl. Phys., 1989, 65, 3610–3616.
3. J. S. Seixas de Melo, H. D. Burrows and J. Pina, in Photochemistry, ed. E. Fasani and A. Albini, The Royal Society of Chemistry, 2016, vol. 43, pp. 83–102 10.1039/9781782622772-00083.
4. M. Andrzejak, D. W. Szczepanik and L. Orzel, Phys. Chem. Chem. Phys., 2015, 17, 5328–5337.
5. G. Cinacchi, J. Phys. Chem. A, 2010, 114, 8114–8118.
6. A. M. Fedor, D. G. Allis and T. M. Korter, Vib. Spectrosc., 2009, 49, 124–132.
7. W. J. D. Beenken, Chem. Phys., 2008, 349, 250–255.
8. M. C. Ruiz Delgado, F. J. Ramírez, V. Hernández, J. Casado, F. Enríquez and J. T. López Navarrete, J. Mol. Struct., 2005, 744–747, 393–401.
9. G. Raos, A. Famulari, S. V. Meille, M. C. Gallazzi and G. Allegra, J. Phys. Chem. A, 2004, 108, 691–698.
10. G. Raos, A. Famulari and V. Marcon, Chem. Phys. Lett., 2003, 379, 364–372.
11. M. Belletête, N. Di Césare, M. Leclerc and G. Durocher, Chem. Phys. Lett., 1996, 250, 31–39.
12. E. Orti, P. M. Viruela, J. Sanchezmarin and F. Tomas, J. Phys. Chem., 1995, 99, 4955–4963.
13. J. E. Chadwick and B. E. Kohler, J. Phys. Chem., 1994, 98, 3631–3637.
14. Y. Kimura, Y. Katano, S. Tanaka, T. Yoshinari, H. Itoh, Y. Kuriyama and S. Nagasaka, Synth. Met., 2010, 160, 1131–1135.
15. B. Gombojav, T. Yoshinari, H. Itoh, S. Nagasaka, Y. Kuriyama and K. Koyama, J. Phys. Soc. Jpn., 2004, 73, 3166–3170.
16. J. L. Houben, R. Cimiraglia, A. Carpita and M. Ciofalo, J. Mol. Liq., 1994, 61, 189–207.
17. V. Hernandez and J. T. L. Navarrete, J. Chem. Phys., 1994, 101, 1369–1377.
18. R. Siebert, D. Akimov, M. Schmitt, A. Winter, U. S. Schubert, B. Dietzek and J. Popp, ChemPhysChem, 2009, 10, 910–919.
19. R. E. Di Paolo, J. Seixas de Melo, J. Pina, H. D. Burrows, J. Morgado and A. L. Maçanita, ChemPhysChem, 2007, 8, 2657–2664.
20. J. Pina, J. S. Seixas de Melo, N. Koenen and U. Scherf, J. Phys. Chem. B, 2013, 117, 7370–7380.
21. M. Fujitsuka, T. Sato, F. Sezaki, K. Tanaka, A. Watanabe and O. Ito, J. Chem. Soc., Faraday Trans., 1998, 94, 3331–3337.
22. U. Asawapirom and U. Scherf, Macromol. Rapid Commun., 2001, 22, 746–749.
23. R. C. Coffin, J. Peet, J. Rogers and G. C. Bazan, Nat. Chem., 2009, 1, 657–661.
24. S. Ko, R. Mondal, C. Risko, J. K. Lee, S. Hong, M. D. McGehee, J.-L. Brédas and Z. Bao, Macromolecules, 2010, 43, 6685–6698.
25. R. S. Becker, J. Seixas de Melo, A. L. Maçanita and F. Elisei, J. Phys. Chem., 1996, 100, 18683–18695.
26. M. Montalti, A. Credi, L. Prodi and M. Gandolfi, Handbook of Photochemistry, CRC Presss and Taylor & Francis, Boca Raton, 3rd edn, 2006.
27. J. Pina, J. Seixas de Melo, H. D. Burrows, A. L. Macanita, F. Galbrecht, T. Bunnagel and U. Scherf, Macromolecules, 2009, 42, 1710–1719.
28. G. Striker, V. Subramaniam, C. A. M. Seidel and A. Volkmer, J. Phys. Chem. B, 1999, 103, 8612–8617.
29. J. J. Snellenburg, S. Laptenok, R. Seger, K. M. Mullen and I. H. M. van Stokkum, J. Stat. Softw., 2012, 49, 22.
30. M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347–1363.
31. T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51–57.
32. J. Pina, D. Sarmento, M. Accoto, P. L. Gentili, L. Vaccaro, A. Galvão and J. S. Seixas de Melo, J. Phys. Chem. B, 2017, 121, 2308–2318.
33. J. Pina, H. D. Burrows, R. S. Becker, F. B. Dias, A. L. Maçanita and J. Seixas de Melo, J. Phys. Chem. B, 2006, 110, 6499–6505.
34. J. Seixas de Melo, L. M. Silva and M. Kuroda, J. Chem. Phys., 2001, 115, 5625–5636.
35. R. S. Becker, J. Seixas de Melo, A. L. Maçanita and F. Elisei, Pure Appl. Chem., 1995, 67, 9.
36. J. Seixas de Melo, H. D. Burrows, M. Svensson, M. R. Andersson and A. P. Monkman, J. Chem. Phys., 2003, 118, 1550–1556.
37. M. L. Horng, J. A. Gardecki, A. Papazyan and M. Maroncelli, J. Phys. Chem., 1995, 99, 17311–17337.
38. J. P. Yang, W. Paa and S. Rentsch, Chem. Phys. Lett., 2000, 320, 665–672.
39. J. Seixas de Melo, F. Elisei and R. S. Becker, J. Chem. Phys., 2002, 117, 4428–4435.
40. P. B. Koster, F. van Bolhuis and G. J. Visser, Acta Crystallogr., Sect. B: Struct. Sci., 1970, 26, 1932–1939.
41. A. Hlel, A. Mabrouk, M. Chemek, I. Ben Khalifa and K. Alimi, Computational Condensed Matter, 2015, 3, 30–40.
42. F. Blomgren and S. Larsson, Theor. Chem. Acc., 2003, 110, 165–169.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7qm00440k

---