Stable Open-Shell Conjugated Terpolymer with Extended NIR Absorption for Organic Photodetectors Detecting Beyond 1000 nm (2025)

2.1. Synthesis and Thermal Properties

The synthesis pathway of the random terpolymer PDPPTBBT is depicted in Figure 1a. PDPPTBBT was synthesized via Stille polymerization in a microwave oven using Pd(PPh₃)₄ as a catalyst when combining the monomers 3,6-bis(5-bromothiophen-2-yl)-2,5-bis(2-decyltetradecyl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (M1), 2,5-bis(trimethylstannyl)thiophene (M2), and 4,7-dibromobenzo[1,2-c:4,5-c′]bis([1,2,5]thiadiazole) (M3) in a 1:2:1 ratio. Polymer solubility was improved by the attachment of 2-decyltetradecyl alkyl chains to the DPP monomer; PDPPTBBT was well-dissolved in chloroform. The crude polymer product was purified via Soxhlet extraction and finally collected in chloroform to obtain the desired material in a yield of 41%. Gel permeation chromatography (GPC, 35 °C, tetrahydrofuran) revealed that the number-average molecular weight and dispersity of PDPPTBBT were 10.5 kDa and 3.13,kDa, respectively (Figure S1). Thermogravimetric measurements were performed to confirm the thermal stability. Decomposition commenced at 320 °C (Figure S2).

2.2. Photophysical and Electrochemcial Properties

Figure 1b shows the absorption spectra of PDPPTBBT, PDPPTBBT:Y6, and PDPPTBBT:P3HT thin films before and after annealing. PDPPTBBT and PDPPTBBT:P3HT films were annealed at 230 °C, whereas PDPPTBBT:Y6 films were annealed at 180 °C. As a typical D–A conjugated polymer, PDPPTBBT exhibits a shorter-wavelength band originating from π–π* transitions and a longer-wavelength band attributed to intramolecular charge transfer between monomers. The main absorption peak of the PDPPTBBT film is located at approximately 780 nm. Notably, the shoulder peak around 1200 nm intensifies after annealing, which suggests that interchain aggregations stabilize the diradical species generated in the open-shell polymer, thereby enhancing absorption near 1200 nm. A similar increase in the shoulder peak intensity is also observed from PDPPTBBT:Y6 and PDPPTBBT:P3HT films. In contrast, Figure 1c shows the absorption spectra of PDPPTBBT in toluene from 25 to 200 °C. As the solution temperature increases from room temperature to 200 °C, the absorption peak near 700 nm undergoes a blueshift and the shoulder peak at around 1200 nm gradually diminishes. This indicates that the polymer transitions from a disordered aggregate state at lower temperatures to a nonaggregated chain state at higher temperatures, reducing interchain interactions. Consequently, the stabilization of the diradical species in the open-shell polymer is weakened in solution at elevated temperatures, leading to decreased absorption near 1200 nm. This phenomenon will be addressed in detail below. Meanwhile, based on the thin-film absorption onsets in the thin films, the optical band gap (E_g^opt) of the PDPPTBBT-based films was estimated to be 0.75 eV.

To explore the electrochemical properties, cyclic voltammetry (CV) analysis was performed to estimate the frontier energy levels based on the onset potentials of oxidation (E_ox) and reduction (E_re) (Figure 1d). The E_ox/E_re of PDPPTBBT was 0.67/–0.76 V corresponding to highest occupied molecular orbital (HOMO)/ lowest unoccupied molecular orbital (LUMO) energy levels of–5.09/–3.66 eV, respectively. The electrochemical band gap (E_g^CV) of PDPPTBBT was 1.43 eV, differing from that of E_g^opt, due to the low dielectric constant and significant exciton-binding energy of the organic material. (33) The optical band gap is determined by the onset of photon absorption, while electrochemical measurements gauge intrinsic HOMO–LUMO gaps. Organic semiconductors often exhibit large exciton-binding energies; therefore, E_g^opt can be lower than E_g^CV. The optical and electrochemical properties are summarized in Table 1.

2.3. Diradical Characteristics

To understand the polymer electronic structure and properties, density functional theory (DFT) calculations were performed at the unrestricted B3LYP/6-31G(d,p) level for the DPPTBBT dimer. In the calculations, the hexadecyl alkyl side chains were replaced with methyl groups. In Figure 2a (top and middle), the α and β singly occupied molecular orbitals (SOMOs) of BBT are apparent, which confirms that BBT played a crucial role in diradical production. BBT allows two electrons with opposite spins to correlate in separate spaces. This correlation further reduces the covalent character of the singlet and thus lowers electron repulsion; an open shell ultimately forms. The singlet–triplet energy gap (ΔE_ST) for the DPPTBBT dimer was −22.1 kcal mol^–1. The transition between the closed- and open-shell states is facilitated by such a small transition energy. Additionally, the spin density distribution shown in Figure 2a (bottom) indicates that the BBT unit density is high across the entire unit, suggesting that the diradical form contributes to the ground state. Therefore, due to the two nonbonding electron pairs generated at the ends of BBT, the dimer loses its conjugation, causing thiophene and DPP to transform into quinoidal forms.

To investigate polymer spins and magnetic properties, zero-field-cooled (ZFC) and 10 kOe field-cooled (FC) magnetization measurements were conducted from 2 to 300 K using a superconducting quantum interference device. Amorphous solid polymer powders were analyzed; these resemble spin-coated thin films. (34) Figure 2c shows the mass magnetic susceptibility dependencies (χ_m values) of the powders. The ZFC-FC curves exhibit negligible splitting; no significant increase in susceptibility was observed even below 50 K. Between 45 and 80 K, the intriguing behavior suggests a potential role for oxygen in terms of magnetization.

Oxygen is easily adsorbed from air under ambient conditions, and the antiferromagnetic oxygen transition around 50 K is well-known. χ_m T is directly proportional to the temperature (Figure 2d). The magnetizations as functions of the magnetic field (M–H) curves of amorphous solid powders recorded at various temperatures (2, 3, 5, 10, 50, and 100 K) were temperature-independent (Figure 2e). Figure 2e also shows that the total magnetization was positive at a high magnetic field (+50 kOe) but negative at a low magnetic field (−50 kOe). Thus, the Pauli paramagnetic susceptibility (χ_Pauli) was 3.2 μemu g^–1 Oe^1–, and PDPPTBBT exhibited temperature-independent Pauli paramagnetic properties that differ from the typical Curie–Weiss behavior. (35,36) Consequently, it is inferred that the diradicals within PDPPTBBT align with the external magnetic field, resulting in a Pauli paramagnet-like susceptibility, which is different from other open-shell CPs (Table S1).

EPR spectroscopy was used to explore the ground state of PDPPTBBT. Initially, spin-coated thin films were evaluated before and after annealing. Figure 2f shows that EPR intensity was enhanced after annealing (red line) compared to that before annealing (pink line), but both lines exhibited the same g-factor of 2.0048, suggesting the presence of unpaired or loosely coupled electrons within the polymer thin films. The relative densities of those films increased after annealing due to enhanced interchain polymer aggregation, in line with the rise in absorption around 1200 nm mentioned above. Next, the EPR spectra were fitted using both Gaussian (blue dashed line) and Lorentzian (black dashed line) functions (Figures 2f and S6). Either function may adequately fit an EPR spectrum due to unpaired electron charges. The Gaussian function indicates that the charges are in denser states and thus relatively immobile; the Lorentzian function implies that the charges are relatively mobile. (37) The observed EPR signals conform to the Lorentzian (black dashed line), indicating the presence of delocalized diradicals. (38) To further understand the intrinsic magnetic properties of PDPPTBBT, low-temperature EPR measurements were performed from 10 to 293 K (Figures 2g and S7). The EPR signal intensity (I_EPR) is temperature (T)-dependent. In particular, as the temperature decreases, the I_EPR of the triplet state rises, (39) indicating a paramagnetic ground state. (35,36) All EPR signal g-factors were 2.0047, and the line widths were relatively broad (approximately 4–5 G). Figure 2h shows the relationship between I_EPR and the reciprocal temperature (1/T) fitted using the Bleaney–Bowers eq (eq 1); we then extracted the energy differences between singlet and triplet states (ΔE_ST values). The calculated ΔE_ST values were 14.3 × 10^–3, 18.3 × 10^–3, and 14.1 × 10^–3 kcal mol^–1 for PDPPTBBT, PDPPTBBT:P3HT, and PDPPTBBT:Y6, respectively; thus, the triplet ground state was more stable than the singlet ground state. The exchange coupling constants (eq 2) were 2.5, 3.2, and 2.5 cm^–1 for PDPPTBBT, PDPPTBBT:P3HT, and PDPPTBBT:Y6, respectively. These results are consistent with the weak intramolecular ferromagnetic coupling apparent between spins (Figure 2b). (35,36)

2.4. Film Morphology, Miscibility, and Molecular Stacking

Molecular stacking transitions within PDPPTBBT induced by thermal annealing and/or blending with P3HT or Y6 molecules were investigated by using a grazing incidence wide-angle X-ray scattering (GIWAXS) platform. Figures S8 and S9 show that PDPPTBBT films exhibited the (100) peaks of lamellar packing in both in-plane and out-of-plane directions, with a lamellar distance of ∼24.2 Å. After thermal annealing, PDPPTBBT, PDPPTBBT:P3HT, and PDPPTBBT:Y6 thin films displayed more distinct lamellar stacking peaks [(100), (200), (300), and (400)] in the out-of-plane direction (Figures 3a–c and S8–S11). This observation implies that all polymers displayed increased crystallinity and preferred lamellar-like molecular orientations. When PDPPTBBT was blended with P3HT and Y6, peaks of P3HT and Y6 were observed (Figure 3d,e). The 2D GIWAXS data for P3HT and Y6 thin films are shown in Figures S11–S14. The coherence lengths (L_c values) and d-spacings associated with various film fabrication conditions were calculated by using the out-of-plane (100) lamellar stacking peaks (Figure 3f,g). For PDPPTBBT thin films, L_c values increased, but the d-spacing decreased after thermal annealing due to enhanced molecular packing after thermal rearrangement. Such packing also increased diradical stability, which is associated with the above-mentioned increase in the EPR signal. The d-spacings of annealed PDPPTBBT:P3HT and PDPPTBBT:Y6 thin films were higher than those of PDPPTBBT thin films; this difference is attributable to certain steric effects of the introduced donor or acceptor molecules. However, the L_c values of the PDPPTBBT:P3HT thin films did not change; crystallinity was maintained. Interestingly, the PDPPTBBT:Y6 thin film showed a much clearer (010) peak in the out-of-plane direction, which is attributed to the face-on arrangement preference of Y6 molecules and their intermolecular interactions with PDPPTBBT polymers (Figures S11 and S14). Therefore, enhanced π–π stacking and face-on orientation in PDPPTBBT:Y6 blends lead to the optimal molecular conformation of the bulk heterojunction system for NIR OPDs.

Next, PDPPTBBT:P3HT and PDPPTBBT:Y6 thin films prepared under various annealing conditions were subjected to atomic force microscopy (AFM) (Figure S15). The root-mean-square roughness (R_q) values of PDPPTBBT:P3HT and PDPPTBBT:Y6 thin films were measured. The R_q values of PDPPTBBT:P3HT thin films increased (4.38 nm), but those of PDPPTBBT:Y6 films decreased (1.98 nm) after annealing. Comparison of the L_c values of annealed PDPPTBBT:P3HT and PDPPTBBT:Y6 thin films showed that the L_c values of PDPPTBBT:P3HT thin films were greater than those of the PDPPTBBT:Y6 thin film due to polymer aggregation. Conversely, the small molecule Y6 exhibited not only a decrease in L_c value but also a reduction in R_q due to the molecular rearrangement with PDPPTBBT. This difference could be attributed to the local aggregation of polymers and small molecules. However, the morphological properties of the blended films do not impair device performance, ultimately resulting in improved device performance after annealing.

2.5. Device Characteristics

OPDs were fabricated to demonstrate that PDPPTBBT detected NIR (Figure 4a,b). After consideration of the energy levels (Figure 4c), PDPPTBBT was blended with either Y6 or P3HT (acceptor or donor, respectively) when fabricating two types of bulk heterojunction OPDs with the device configuration ITO/ZnO/active layer/MoO₃/Au. In the two configurations, PDPPTBBT served as the donor or the acceptor, which affirmed the dual p-type and n-type functionalities of the polymer. The operating mechanisms under reverse bias of both PDPPTBBT:Y6- and PDPPTBBT:P3HT-based devices are illustrated in Figures 4c and S16a, respectively. Figure 1b confirms that the NIR absorption characteristics were maintained in both the PDPPTBBT:Y6 and PDPPTBBT:P3HT blends. After optimization of device fabrication, it was confirmed that NIR OPDs exhibited photoresponses up to a (long) wavelength of 1100 nm. The J–V curves of OPDs based on PDPPTBBT:Y6 and PDPPTBBT:P3HT in the dark and under light at 850 nm (150 μW cm^–2) and 1,050 nm (87 μW cm^–2) are shown in Figures 4d and S16b. Both devices exhibited low dark current densities (J_dark values) under reverse bias, and J_dark remained stable to −10 V. As the bias increased, the photocurrent density (J_ph) rose due to enhanced charge extraction. At −10 V, the J_dark of the OPDs based on PDPPTBBT:Y6 was 6.25 μA cm^–2, whereas that of the OPDs based on PDPPTBBT:P3HT was 82.5 μA cm^–2. On OPD illumination at 850 and 1050 nm, devices based on PDPPTBBT:Y6 exhibited J_ph values of 139 and 92.5 μA cm^–2, respectively, which confirmed high NIR sensitivity. OPDs based on PDPPTBBT:P3HT displayed J_ph values of 160 and 43.2 μA cm^–2 at 850 and 1050 nm, respectively. Compared with devices based on PDPPTBBT:Y6, those fabricated from PDPPTBBT:P3HT exhibited a greater current density increase at 850 nm but a fall in sensitivity at the (long) wavelength of 1050 nm, which is attributable to the relatively low absorption of PDPPTBBT:P3HT thin films in the long-wavelength region.

To quantitatively describe and compare the performances of the various NIR OPDs, the responsivity (R) and EQE values were calculated by using eqs 3–7). The EQE, R, and D* values of devices based on PDPPTBBT:P3HT and PDPPTBBT:Y6 were measured as a function of the wavelength (Figure 4e,f for PDPPTBBT:Y6 and Figure S16c,d for PDPPTBBT:P3HT). As the bias increased, J_dark remained relatively constant, but J_ph significantly increased in association with a higher EQE and D* at −10 V. In particular, OPDs based on PDPPTBBT:Y6 exhibited EQEs of 135 and 126% at 850 nm and 1,050 nm and the corresponding D* values of 6.6 × 10¹¹ and 7.5 × 10¹¹ Jones, respectively. OPDs based on PDPPTBBT:P3HT exhibited EQE values of 153 and 59% and D* values of 2.0 × 10¹¹ and 9.7 × 10¹⁰ Jones at 850 and 1050 nm, respectively. The results are summarized in Table 2. The novel open-shell conjugated polymer serves as either the acceptor or donor in OPDs, and it is noteworthy that it can act as one of the rare OPDs capable of detecting wavelengths beyond 1000 nm.

To investigate the linear dynamic range (LDR) of the OPDs, photocurrents were measured at −10 V over a wide range of incident intensities (Figure S17). Under these conditions, the OPDs based on PDPPTBBT:P3HT and PDPPTBBT:Y6 exhibited an LDR of 20 and 25 dB, respectively, indicating a more linear response in the Y6 blend. In addition, the dynamic performance was evaluated by measuring rise/fall times and the −3 dB cutoff frequency (f_–3db) at −10 V (Figure S18). The rise (t_r) and fall times (t_f) are defined as the time for the photocurrent to increase from 10 to 90% of its maximum value and from 90 to 10% of its maximum value, respectively. The OPDs based on PDPPTBBT:P3HT showed a t_r of 16 μs and a t_f of 27 μs, while those based on PDPPTBBT:Y6 exhibited a t_r of 14 μs and a t_f of 17 μs (Figure S18a). The f_–3db values were measured under 10 Hz-modulated LED illumination. The f_–3db value for the PDPPTBBT:P3HT devices was 125 kHz, whereas that of the PDPPTBBT:Y6 devices was 236 kHz, confirming faster response speeds in the PDPPTBBT:Y6 based on the OPDs (Figure S18b).

To further investigate the mechanisms involved, the charge mobilities of PDPPTBBT:P3HT and PDPPTBBT:Y6 thin films were measured using the space charge limited current (SCLC) method. (40,41) To determine the mobilities of holes and electrons in the two types of thin films, hole-only and electron-only devices were fabricated. The thin films were coated, as were the OPD devices discussed above. To create a hole-only device, PEDOT:PSS was substituted for ZnO; to prepare an electron-only device, BCP was used instead of MoO₃. As shown in Figure S19, when the slopes of the J^1/2–V curve plots for each device are compared, and the mobility ratios between holes (μ_h) and electrons (μ_e) are obtained. In PDPPTBBT:Y6 thin films, the electron mobility was approximately 2.61-fold greater than the hole mobility; in PDPPTBBT:P3HT thin films, the hole mobility was about 8.32-fold greater than the electron mobility. These differences in the mobility between holes and electrons signify unbalanced charge-transport properties within the devices. Faster photogenerated carriers that are formed during device operation are directed toward the electrodes; slower carriers become trapped within the films. It has been previously reported that artificial creation of charge carrier traps within blended films triggered photomultiplication; trapped carriers induced band bending and influx of oppositely charged carriers from an electrode. (42,43) These devices thus exhibit photomultiplication, enhancing the performance at sufficiently high voltages.

4.1. Materials

Anhydrous solvents, sodium metal, 2-methylbutan-2-ol, 2-thiophenecarbonitirile, diethyl succinate, potassium carbonate, 2-decyltetradecyl bromide, tetrakis(triphenylphosphine)palladium(0), potassium acetate, N-bromosuccinimide, and 2,5-bis(trimethylstannyl)thiophene were purchased from Sigma-Aldrich. 4,7-Dibromobenzo[1,2-c:4,5-c′]bis([1,2,5]thiadiazole) was obtained from J’s Science (South Korea). All chemicals were used without further purification.

4.2. Synthesis of PDPPTBBT

4.3. Characterization

4.4. Computational Methodology

All of the electronic structural calculations were performed with the Gaussian 09 software package. The molecular geometries of the electronic ground state and the first triplet state of the model dimer were optimized using DFT that incorporated the Becke three-parameter B3LYP functional and the 6-31G(d,p) basis set. The alkyl chains in the polymer backbone were substituted with methyl (−CH3) groups. It is known that the alkyl functional groups do not significantly affect either the ground-state geometry or the electronic structural properties.

4.5. Device Fabrication and Characterization

The OPD devices were fabricated as ITO/ZnO/active layer/MoO₃/Au preparations. ITO-coated glass substrates were cleaned by sonication in a detergent, deionized water, acetone, and isopropyl alcohol. The ZnO layers were spin-coated onto ITO-coated glass substrates at 3000 rpm and then thermally annealed at 110 °C for 10 min in air. The PDPPTBBT:P3HT (1:1 wt/wt)-based blends and PDPPTBBT:Y6 (1:2 wt/wt)-based blends were dissolved in chloroform to final concentrations of 20 and 30 mg mL^–1, respectively, and the solutions were stirred at room temperature overnight in a nitrogen-filled glovebox. The solutions were then spin-coated onto the ZnO films. Blends based on PDPPTBBT:P3HT were annealed at 230 °C for 30 min; blends based on PDPPTBBT:Y6 were annealed at 180 °C for 30 min. Next, MoO₃ and Au were thermally deposited to thicknesses of 5 and 20 nm, respectively.

The current–voltage characteristics and real-time photoswitching behaviors of the OPDs were measured in a vacuum chamber using a Keithley 4200-SCS semiconductor parametric analyzer.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.5c03911.

• GPC, TGA, and ¹H NMR data of the synthesized materials, as well as additional EPR data; 2D GIWAXS, AFM, SCLC, LDR, and response speed of the device; and energy diagram, J–V curve, EQE, responsivity, and detectivity of the PDPPTBBT:P3HT device (PDF)

• Corresponding Author

  • Joon Hak Oh - Schoolof Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 08826, Republic of Korea; https://orcid.org/0000-0003-0481-6069; Email: [emailprotected]

• Authors

  • Moon-Ki Jeong - Schoolof Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 08826, Republic of Korea; https://orcid.org/0009-0009-4199-9336

  • Sang Hyuk Lee - Schoolof Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 08826, Republic of Korea; https://orcid.org/0009-0009-4829-5120

  • Yousang Won - Schoolof Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 08826, Republic of Korea; https://orcid.org/0009-0003-7125-7858

  • Jaeyong Ahn - Schoolof Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 08826, Republic of Korea; Present Address: Department of Chemical Engineering, Stanford University, Stanford, California 94305, United States; https://orcid.org/0000-0002-0036-0300

  • Myeong In Kim - Departmentof Organic and Nano Engineering and Human-Tech Convergence Program, Hanyang University, Seoul 04763, Republic of Korea; https://orcid.org/0009-0008-5996-6308

• Author Contributions

  M.-K.J. and S.H.L. contributed equally to this work.

• Notes

  The authors declare no competing financial interest.
  

This work was supported by the National Research Foundation of Korea (NRF) grant (2023R1A2C3007715, RS-2024-00398065) through the NRF by the Ministry of Science and ICT (MSIT), Korea. The Institute of Engineering Research at Seoul National University provided research facilities for this work. This work was supported by Samsung Electronics Co., Ltd.

Download PDF