Covalent photo-crosslinking of diketopyrrolopyrrole based polymeric layers for cutting-edge near-IR absorption and DSSC performance

Abstract

We have developed a novel solvent-free approach for the synthesis and photocross-linking cycloaddition of diketopyrrolopyrrole (DPP)-based semiconducting polymers, tailored with carbazole side chains incorporating linear octene (DPP-OCT-CBZ) and cyclic vinyl benzyl (DPP-VC-CBZ) spacers. Utilizing UV light (365 nm), these materials undergo photopolymerization cyclo-addition, resulting in robust films characterized by extended UV absorption up to the near-infrared region (1100 nm) and enhanced thermal stability up to 400 °C. Fourier Transform Infrared Spectroscopy (FTIR) confirmed the formation of cross-linked structures while ¹H-NMR and GPC confirm structures of polymers, while morphological uniformity and mechanical rigidity were verified through Atomic Force Microscopy (AFM). Importantly, these materials demonstrate exceptional barrier properties against moisture, evidenced by improved water contact angles. When integrated as barrier layers in DSSCs with N719 dye, DPP-OCT-CBZ-C particularly exhibited a remarkable power conversion efficiency (PCE) of 18.2% under illumination of 800 lux and 7.28% under 1 sun. This efficiency is attributed to the superior alignment of the linear octene chains with the DSSC layers, effectively suppressing back electron transfer. Our findings highlight the potential of DPP-based polymers in advancing the performance and durability of organic electronic devices, offering a promising avenue for the design of new materials for energy conversion and storage applications.

---

1. Introduction

Organic semiconductors, notably π-conjugated polymers, stand out for their pivotal role in advancing flexible, eco-friendly technology. The urgent need for clean, sustainable energy solutions has propelled significant research into photovoltaic (PV) technologies, with solar power at the forefront due to its availability and environmental compatibility.¹ Organic conjugated polymers, especially those featuring π-conjugation, have emerged as attractive materials for next-generation solar cells because of their light weight, mechanical flexibility, and potential for cost-effective production via solution processing. Among these, diketopyrrolopyrrole (DPP)-based polymers stand out for their strong near-infrared (NIR) absorption, high charge-carrier mobility, and chemical stability-properties that are especially advantageous in dye-sensitized solar cells (DSSCs). Additionally, by acting as a barrier layer, DPP-based polymers help shield critical interfaces from oxygen and moisture, thus safeguarding long-term device performance.

However, the structural and thermal stabilities of organic polymers remain major concerns in their commercial deployment. Environmental factors, such as temperature fluctuations, oxygen permeation, and moisture ingress, can compromise device lifespan and efficiency. To address these challenges, covalent photo-crosslinking has gained attention as an effective strategy for fortifying polymeric networks. Upon UV exposure, crosslinkable sites on the polymer backbone form robust, three-dimensional networks that substantially enhance mechanical strength, thermal tolerance, and chemical resilience. This process simultaneously preserves, or even improves, the polymer's optoelectronic properties by reducing defect pathways and improving film uniformity-factors that collectively lead to higher power conversion efficiencies and operational stability in PV devices.² This advancement opens the door to crafting organic electronics with distinct and desirable properties.³

In the realm of organic electronics, diketopyrrolopyrrole (DPP)-based materials stand out for their exceptional optical and electronic properties.⁴ While both small molecules and polymeric forms of DPP⁵ have been investigated, polymeric DPP exhibits distinct advantages due to its higher molecular weight and extended π-conjugation, which confer superior processability, mechanical strength, broad absorption, and high carrier mobilities. Crosslinking these polymers⁶ further enhances their environmental stability and durability by forming robust networks within the polymer matrix, thereby improving structural integrity and prolonging the operational lifetime of photovoltaic cells.⁷ Moreover, incorporating carbazole units into the DPP backbone significantly enhances light absorption and charge transport, while conjugation-breaking alkylations, such as octene and vinyl benzyl groups,optimize crosslinking without sacrificing optical or electronic performance. These modifications collectively boost charge transfer, short-circuit current density (J_sc), and power conversion efficiency (PCE) in dye-sensitized solar cells (DSSCs), underscoring the crucial role of DPP-based polymers as effective barrier layers against moisture and morphological degradation in photovoltaic applications.^8,9 Moreover, phenomena such as J-aggregates (ordered molecular stacking that enhances stronger absorption) and intramolecular charge transfer (ICT) bands (electron transitions between donor and acceptor units within the same molecule) are pivotal in defining the optical and electronic characteristics of DPP-based polymers.

Photocrosslinking in DPP-based polymers significantly enhances their stability and functionality under UV exposure, which is crucial for solar cell applications. When irradiated, these polymers form covalent bonds between chains, creating a network that resists chemical, thermal, and mechanical stress. This is especially beneficial for dye-sensitized solar cells (DSSCs), where improved barrier properties against oxygen and moisture can substantially prolong device lifespan and boost efficiency. Zhang C. et al.¹⁰ synthesized and characterized photo-crosslinkable polymers, demonstrating that UV-induced crosslinking improves both stability and efficiency. Furthermore, the strategic placement of crosslinkable groups in DPP polymers can preserve optimal optical properties for light harvesting. Ye et al.¹¹ showed that such crosslinked copolymers exhibit enhanced mechanical performance and photovoltaic efficiency, underscoring the advantages of a crosslinking approach. The resulting dense, impermeable networks protect sensitive solar cell layers from environmental degradation and bolster moisture resistance, as highlighted by Noh et al.¹² Overall, these findings showcase the potential of photo-crosslinked DPP-based polymers to advance the performance and durability of organic photovoltaic systems. Lastly, C–H activation at the C-5 position on the thiophene moiety of a 3,6-di(thiophen-2′-yl) diketopyrrolopyrrole derivative occurs regioselectively, yielding trans-configured cross-coupled products. This synthetic strategy further illustrates how precise molecular design can enhance the functional attributes of DPP-based materials.

In this work, we introduce linear (DPP-OCT-CBZ) and cyclic (DPP-VC-CBZ) DPP-based polymers engineered for photo-crosslinking. These materials exhibit near-infrared absorption and form robust polymer networks with enhanced thermal stability, moisture resistance, and morphological integrity—properties vital for the durability and efficiency of organic solar cells. By integrating these crosslinked networks as a protective barrier in dye-sensitized solar cells (DSSCs), we demonstrate how photo-crosslinking can effectively reduce charge recombination and shield sensitive layers from external degradation with an increased PCE of 18.2% (DPP-OCT-CBZ-C) from 16.3% (DPP-OCT-CBZ) upon crosslinking. Such improvements underscore the importance of carefully tuning side chains and crosslinkable groups to achieve both strong optical absorption and superior environmental stability in organic semiconductors. Furthermore, the straightforward C–H coupling route used to synthesize these DPP-based polymers highlights their scalability and potential for broader commercial adoption, thereby expanding the possibilities for high-performance, durable organic photovoltaics.

2. Experimental

2.1 Resources and general information

3,6-Di(thiophene-2-yl) pyrrolo [3,4-c] pyrrole-1,4-(2H, 3H) dione (DPP-H) was prepared by following the literature;^13,14 vinylbenzylchloride/bromo-7-octene (>90.0% (GC), Sigma) and 3,6-dibromo-carbazole (>98.0% (GC), TCI) were purchased from Sigma and TCI respectively. Anhydrous N,N-dimethylformamide (DMF) and dimethylacetamide (DMAc) were purchased from Sigma-Aldrich. Catalysts Pd (OAc)₂ and pivalic acid were acquired from Sigma-Aldrich, whereas the base potassium carbonate was purchased from TCI chemicals and sodium hydroxide (Anhydrous, Finar) was used. The purification of all the chemicals and solvents follows standardized practices.

2.2 General management and characterization

On an ECS 400 MHz (JEOL) NMR spectrometer in CDCl₃, ¹H NMR spectra were recorded. A CHNS analyser and NICOLAS-6700, USA, FTIR spectrometer both were used to perform elemental analyses and record FTIR spectra in the 400–4000 cm⁻¹ scan range. HRMS data were calculated using a QTOF instrument by Waters (Xevo G3 QTof). Using Waters 1515 gel permeation chromatography (GPC) analysis, polystyrene was used as the standard and THF as the eluent to evaluate the weight and number of average molecular weights and the polydispersity index (PDI) of the terpolymers. 1, 2-Dichloromethane (DCM) was used as the solvent and 0.1 M tetrabutylammonium hexafluorophosphate in 30 mL acetonitrile was used as the electrolyte for cyclic voltammetry studies on a KEITHLY instrument. Supporting electrolyte (Bu₄NPF₆) was purchased from TCI Chemicals. With a scan rate of 100 mV s⁻¹, a three-electrode cell was used, consisting of a working electrode made of a glassy carbon electrode (coated with a thin film of the polymers), a counter electrode made of platinum mesh, and a reference electrode made of Ag/AgCl (0.01 M). Under a nitrogen atmosphere, thermogravimetric analysis (TGA) was carried out using a mass of 5 mg and a heating rate of 10 °C min⁻¹. Both solutions and films of polymers in chloroform solvent were subjected to UV-vis spectroscopic measurements on a PerkinElmer Lambda 750 and an Agilent (200–800 nm) UV-vis spectrophotometer. An Agilent's Cary Eclipse fluorescence spectrophotometer was used to perform the photoluminescence (PL) spectrum measurements (model: G9800A). The topography of thin films made from terpolymers was evaluated using an Atomic Force Microscope (AFM) via XE-100, PARK-SOUTH KOREA. The film of polymers was created in chloroform solvent with a 5 mg mL⁻¹ solution spin coated on an ITO substrate on a hotplate at 200 °C inside of a glove box.

2.3 Device fabrication and characterization

FTO/TiO₂/N719/Polymers/Electrolyte/Pt is the device's structural breakdown. Commercially available FTO with a surface resistivity of 7 Ω sq⁻¹ was first ultrasonically treated for 15 × 3 minutes with soap water, distilled water, isopropanol, and acetone and then dried in an oven. The TiO₂ semiconductor was then screen printed on FTO, and this was followed by one hour of 450 °C sintering. The prepared photoanode was then extensively cleaned with ethanol to eliminate any remaining unanchored dye particles from the sensitized photoanode. Then, manufactured polymers (DPP-OCT-CBZ-C, DPP-OCT-CBZ, DPP-VC-CBZ, and DPP-VC-CBZ-C) at a concentration of 1 mg mL⁻¹ in chlorobenzene solution were spin-coated on top of the active layer. Commercial Pt was employed as a counter electrode.¹⁵ To finish off the device's structure, the iodine-based electrolyte was kept in place between the counter electrode and photoanode with binder clips. The fabricated devices with active area 0.25 cm² were characterized in an electrochemical workstation (BIOLOGIC) with a Solar 800 Lux indoor 9 W LED bulb (F6500, 825 lm, 0.050 A, 91.7 lm per W, Philips, power factor > 0.9) providing illumination at the aforementioned intensity and a solar simulator from SAN-EI electronics, JAPAN, at a power density of 100 mW cm⁻² (see Scheme 1).

Scheme 1 Dye sensitized solar cell fabrication process with crosslinked polymers as barrier layers (FTO/TiO₂/N719/Polymers/Electrolyte/Pt).

3. Synthesis

3.1 DPP-VC/DPP-OCT

The following were added to a dry 50 mL round-bottom flask: DMSO (6 mL), 4-vinylbenzylchloride/bromo-7-octene (0.9 mmol, 129/154 μL), DPP-H (0.6 mmol, 200 mg), and NaOH solution (2.5 mmol, 100 mg diluted in 1 mL of water). After four hours at 50 degree celsius, the mixture was allowed to cool to ambient temperature. Brine (10 mL) was used to quench the reaction mixture, and chloroform (3 × 15 mL) was used for extraction. After mixing, washing with brine and water, drying over anhydrous MgSO₄, filtering, and concentration under low pressure, the organic layers were mixed. By using column chromatography to elute the crude product with CH₂Cl₂/petroleum ether (4/6 v/v), the crude product was purified and the title compounds were obtained as red solids (Scheme 2).

Scheme 2 Synthesis procedure of DPP-VC/OCT monomers.

3.1.1 DPP-VC: yield 84%. ¹H NMR (400 MHz, CDCl₃, δ) 7.81 (d, 2H), 7.18 (d, 6H), 7.53 (d,4H), 6.63 (dd, 2H), 5.61 (dd, 2H), 5.84 (m, 2H), 5.18 (m, 2H).

HRMS ESI-TOF m/z calculated for C₃₀ H₄₀ N₂ O₂ S₄ [M + H]: 525.62; found: 525.63.

3.1.2 DPP-OCT: yield 86%. ¹H NMR (400 MHz, CDCl₃, δ) 7.79 (d, 2H), 7.18 (d, 2H), 5.02–5.82 (t, 6H), 4.18 (t, 4H), 2.18 (m, 4H), 1.63 (m, 4H), 1.29 (m, 8H).

HRMS ESI-TOF m/z calculated for C₃₂ H₂₄ N₂ O₂ S₄ [M + H]: 533.12; found: 533.13.

3.2 DPP-VC-CBZ

In a 25 mL one-necked round bottom flask that has been previously charged and outfitted with a condenser and magnetic pellet, 200 mg DPP-VC (1 eq., 0.375 mmol) and 182.81 mg of 3,6-dibromo-carbazole (1.5 eq., 0.562 mmol) with 4 mL of DMAc were added along with 38 mg of pivalic acid, 155 mg of potassium carbonate, and 5% of palladium acetate as 1 eq., 3 eq., and 0.1 eq., respectively. N₂ gas is then used to purge the reaction mixture for two hours, rendering it inert. The reaction was run at 100 °C for 24 hours (see Scheme 3). The reaction mixture was cooled after 24 hours had passed. After completion, the reaction mixture was brought to room temperature. After that, the flask is filled with 25 mL of chloroform and filtered. The filtrate is then concentrated to 4–5 mL using a rotary evaporator and added dropwise into the methanol with vigorous stirring. The material gets precipitated and is collected by filtration. The residue with filter paper was subjected to Soxhlet extraction by methanol, petroleum ether and chloroform respectively. The desired product gets dissolved in chloroform and collected using a rotary evaporator. There are blue solids with 75% yield.

Scheme 3 Schematic representation of the synthesis of both the crosslinked polymers and their monomers and mechanism. (a) 8-bromo octene/4-vinyl benzyl chloride, NaOH, anhydrous DMF, 5 h, 50 °C and (b) 3,6-dibromo carbazole, potassium carbonate, pivalic acid, palladium acetate, dry DMAc, 24 h, 110 °C.

M _w = 72 300 g mol⁻¹, M_n = 31 400 g mol⁻¹, PDI = 2.3.

¹H NMR (400 MHz, CDCl₃, δ) 9.12 (d, 1H), 8.27 (d, 1H), 8.04 (d, 1H), 7.98 (d, 2H), 7.77 (d, 2H), 7.63 (d,1H), 7.5 (dd, 1H), 7.18 (dd, 1H), 7.59 (dd, 1H), 4.98 (t, 4H), 5.84 (m, 2H), 5.18 (m, 1H).

Elemental analysis (calcd): C, 76.00; H, 4.80; N, 5.18; O, 4.41; S, 8.83. Found: C, 76.00; H, 4.10; N, 5.4; S, 8.617.

3.3 DPP-OCT-CBZ

In a 25 mL one-necked round bottom flask that has been previously charged and outfitted with a condenser and magnetic pellet, 200 mg DPP-OCT (1 eq., 0.38 mmol) and 185.25 mg 3,6-dibromocarbazole (1.5 eq., 0.57 mmol) with 4 mL of DMAc were added along with 38.04 mg of pivalic acid, 157 mg of potassium carbonate, and 5% of palladium acetate as 1 eq., 3 eq., and 0.1 eq., respectively. The purification procedure is the same as that mentioned above.

The desired product gets dissolved in chloroform and collected using a rotary evaporator. There are violet films with 85% yield (see Scheme 3).

M _w = 110 000 g mol⁻¹, M_n = 81 000 g mol⁻¹, PDI = 1.35.

¹H NMR (400 MHz, CDCl3, δ) 9.84 (d, 1H), 8.24 (d, 1H), 8.05 (d, 1H), 7.99 (d, 2H), 7.51 (d, 2H), 7.63 (d,1H), 7.5 (dd, 1H), 5.02–5.82 (m, 6H) 4.18 (t, 4H), 1.84 (m, 4H), 1.21 (m, 16H).

Elemental analysis (calcd): C, 74.00; H, 6.10; N, 5.18; O, 4.90; S, 8.82. Found: C, 74.00; H, 6.10; N, 5.4; S, 6.617.

3.4 Crosslinking (DPP-VC-CBZ-C and DPP-OCT-CBZ-C)

Glass Petri dishes are drop cast with solutions prepared by dissolving DPP-VC-CBZ/DPP-OCT-CBZ in anhydrous chloroform (Scheme 4) at a concentration of 5 mg mL⁻¹ and swirled for 24 hours at 40 °C. Then they were exposed to UV (365 nm) at 176 W m⁻² for an hour before being withdrawn from the photoreactor. The resultant components were either scraped off the glass slide using a scalpel or blade, or used directly for characterisation.

Scheme 4 Photocrosslinking of polymers under UV light (365 nm).

4. Results and discussion

The DPP-based polymers DPP-OCT-CBZ and DPP-VC-CBZ, along with their crosslinked counterparts (Scheme 3), have been extensively characterized to confirm their molecular structures and evaluate their potential in photovoltaic applications. Notably, these polymers exhibit promising polydispersity indices (PDIs) of 1.35 for DPP-OCT-CBZ and 2.32 for DPP-VC-CBZ, indicating a controlled molecular weight distribution which is crucial for consistent performance in device applications. These PDIs suggest that DPP-OCT-CBZ forms more uniform polymer chains compared to DPP-VC-CBZ, potentially leading to more reproducible photovoltaic properties.

The structural integrity of these polymers has been rigorously verified through various analytical techniques. NMR (Nuclear Magnetic Resonance) and CHNS (Carbon, Hydrogen, Nitrogen, Sulfur) analysis have been employed to confirm the backbone structure of the polymers, ensuring accurate synthesis according to the designed molecular architecture. Additionally, the molecular weights of the monomers, DPP-VC and DPP-OCT, were precisely determined using mass spectrometry, further corroborating the successful synthesis of the intended monomeric units.

To understand the film formation in DPP-OCT-CBZ-C, FTIR (Fourier Transform Infrared Spectroscopy) was utilized to study the specific interactions and bonding within the polymer, especially focusing on the linear octene chains. This analysis confirmed the presence of expected functional groups and highlighted the structural changes occurring during the crosslinking process, which contribute to the crosslinked film properties.¹⁶ Moreover, all these molecules demonstrated good solubility in organic solvents, a characteristic that facilitates their application in organic electronics and ensures ease of processing in manufacturing environments. The combined analytical approaches provide a comprehensive understanding of the material properties, confirming the successful synthesis and functional potential of these innovative DPP-based polymers for advanced photovoltaic applications. Photo-crosslinking of DPP polymers enhances the DSSC architecture by improving interfacial stability, suppressing recombination, and increasing environmental and morphological robustness-all of which contribute to higher efficiency and longer device lifespan. The photo-crosslinked DPP polymers (DPP-OCT-CBZ-C and DPP-VC-CBZ-C) are used as an interfacial barrier layer within the DSSC stack. This layer is positioned between the photoanode (typically TiO₂/dye) and the electrolyte, where it plays a key role in controlling interfacial interaction.

4.1 Structural properties

4.1.1 Gel permeation chromatography & nuclear magnetic resonance analysis. The ¹H-NMR spectra (Fig. 1) provided peak characteristics of the two DPP-based (diketopyrrolopyrrole) polymers with varying side chains: one with vinyl chloride (DPP-VC-CBZ) and the other with 8-octene (DPP-OCT-CBZ), both containing a carbazole (CBZ) unit. DPP is known for its strong electron-accepting ability, while carbazole acts as an electron donor, forming a donor–acceptor architecture that enhances the polymer's optoelectronic properties. The downfield peaks (7–9 ppm) represent aromatic protons on the carbazole and DPP core, where electron-withdrawing effects from the DPP core cause deshielding, shifting these protons downfield. Peaks in the mid-region (around 5–6 ppm) likely correspond to vinyl protons in DPP-VC-CBZ or protons near the double bond in the DPP-OCT-CBZ side chain. The up-field signals (1–3 ppm) are attributed to aliphatic protons in the octene or vinyl chloride chains, reflecting the shielding effect in these alkyl groups. The splitting patterns observed provide insights into the polymer backbone structure and substitution on the side chains, confirming successful functionalization with vinyl chloride and octene side chains. These spectral features align with the characteristic NMR signatures observed for DPP and carbazole-containing conjugated polymers, which are commonly studied for their semiconducting properties.¹⁷

Fig. 1 ¹H NMR spectra of synthesized copolymers.

The DPP-based polymers, DPP-OCT-CBZ and DPP-VC-CBZ, exhibit distinct molecular weight profiles and polydispersity indices (PDIs), which influence their material properties (Table 1). DPP-OCT-CBZ has a higher molecular weight (M_w = 110 000 g mol⁻¹) and a narrower PDI of 1.35, indicating a more uniform polymer chain length, beneficial for applications requiring consistent film formation and stability. In contrast, DPP-VC-CBZ has a lower molecular weight (M_w = 71 300 g mol⁻¹) and a broader PDI of 2.3, reflecting a more varied chain length distribution. This broader PDI may impact its mechanical properties and electronic performance differently, potentially offering more flexibility but less uniformity in applications compared to DPP-OCT-CBZ. These variations make each polymer suited to different functional requirements in optoelectronic and materials science applications.¹⁸

Table 1 GPC of DPP based copolymers

Polymers	M w (g mol^−1)	M n (g mol^−1)	PDI
DPP-OCT-CBZ	110 000	81 000	1.35
DPP-VC-CBZ	71 300	31 400	2.3

---

4.1.2 Fourier transform infrared spectroscopy analysis. The FTIR spectra in Fig. 2 provide clear evidence of the presence of DPP, CBZ, octene, and vinyl benzyl moieties in the synthesized polymers. The diketopyrrolopyrrole (DPP) units are characterized by a strong peak at around 1730 cm⁻¹, corresponding to the C=O stretching vibrations.¹⁹ Carbazole (CBZ) units are identified by the broad N–H stretching peak at around 3300 cm⁻¹ and the aromatic C=C stretching vibrations in the range of 1600–1500 cm⁻¹. For octene, the aliphatic C–H stretching vibrations are observed at 2920 cm⁻¹ and 2850 cm⁻¹. Similarly, vinyl benzyl groups also show C–H stretching vibrations at around 2920 cm⁻¹ and 2850 cm⁻¹, with additional aromatic C=C stretching vibrations in the 1600–1500 cm⁻¹ range. These characteristic peaks confirm the presence of the polymer's corresponding functional group structures. The crosslinked polymers, DPP-VC-CBZ-C and DPP-OCT-CBZ-C, exhibit similar FTIR peaks to their non-crosslinked counterparts, indicating that the primary structural components are retained post-crosslinking. The N–H stretching peak at around 3300 cm⁻¹ and the C–H stretching peaks at 2920 cm⁻¹ and 2850 cm⁻¹ remain present in the crosslinked samples. Aromatic C=C stretching vibrations at around 1600–1500 cm⁻¹ and the C=O stretching peak at 1730 cm⁻¹ are also observed, demonstrating the preservation of the aromatic and DPP units. Crosslinking may introduce additional peaks or variations in intensity, particularly in the fingerprint region (1500–500 cm⁻¹), indicating the formation of new bonding interactions.¹⁷ These observations confirm that crosslinking does not disrupt the essential structural features of the polymers while potentially enhancing their stability and performance for applications in DSSCs.

Fig. 2 FTIR spectra of copolymers and their crosslinked polymers from 500 cm⁻¹ to 4000 cm⁻¹.

4.2 Electrochemical properties

The cyclic voltammetry (CV) analysis for the synthesized DPP-VC-CBZ and DPP-OCT-CBZ polymers, along with their crosslinked counterparts (DPP-VC-CBZ-C and DPP-OCT-CBZ-C), provides important insights into their electrochemical properties with error and potential performance in photovoltaic applications (Fig. 3 and Table 2). The CV curves, as shown in the provided figure, reveal distinct electrochemical behaviours influenced by the molecular structure and crosslinking. The crosslinking of DPP-VC-CBZ and DPP-OCT-CBZ to form DPP-VC-CBZ-C and DPP-OCT-CBZ-C results in a slight shift in both the HOMO and LUMO levels, with the HOMO level becoming less negative and the LUMO level slightly more negative. This results in a reduced band gap from 1.4 eV (DPP-VC-CBZ) to 1.28 eV (DPP-VC-CBZ-C) and 1.19 eV (DPP-OCT-CBZ) to 1.04 eV (DPP-OCT-CBZ-C) tabulated in Table 2.²⁰ The reduction in the band gap indicates improved charge carrier separation and transport, which can enhance the photovoltaic performance by increasing V_oc and J_sc.

Fig. 3 CV of copolymers and their crosslinked counterparts in a potential window of −1.5 V to 1.5 V at a scan rate of 100 mV s⁻¹.

Table 2 CV of DPP based copolymers and their crosslinked counterparts

Polymers	E onset oxidation (eV)	E HOMO (eV)	E onset reduction (eV)	E LUMO (eV)	E g (eV)
DPP-OCT-CBZ	0.51 ± 0.012	−5.31 ± 0.012	−0.68 ± 0.008	−4.12 ± 0.008	1.19 ± 0.014
DPP-OCT-CBZ-C	0.40 ± 0.011	−5.2 ± 0.011	−0.64 ± 0.010	−4.16 ± 0.010	1.04 ± 0.015
DPP-VC-CBZ	0.71 ± 0.012	−5.51 ± 0.012	−0.69 ± 0.012	−4.11 ± 0.012	1.4 ± 0.017
DPP-VC-CBZ-C	0.62 ± 0.014	−5.42 ± 0.014	−0.66 ± 0.011	−4.14 ± 0.011	1.28 ± 0.018

---

DPP-OCT-CBZ and its crosslinked form (DPP-OCT-CBZ-C) demonstrate superior electrochemical properties compared to DPP-VC-CBZ and DPP-VC-CBZ-C, making them more suitable for advanced photovoltaic applications. The higher HOMO levels of DPP-OCT-CBZ (−5.31 eV) and DPP-OCT-CBZ-C (−5.2 eV) compared to DPP-VC-CBZ (−5.51 eV) and DPP-VC-CBZ-C (−5.42 eV) indicate enhanced hole transport properties, which are crucial for efficient charge separation. Additionally, the narrower band gaps of DPP-OCT-CBZ (1.19 eV) and DPP-OCT-CBZ-C (1.04 eV) allow for broader light absorption, leading to increased exciton generation and higher short-circuit currents (J_sc). The improved alignment of HOMO and LUMO levels in DPP-OCT-CBZ polymers ensures more efficient charge transfer and reduced recombination losses. Crosslinking further enhances the thermal and chemical stability of these polymers, with DPP-OCT-CBZ-C exhibiting significant improvements in light absorption and charge transport due to its reduced band gap. Consequently, DPP-OCT-CBZ polymers offer better photovoltaic performance, potentially achieving higher power conversion efficiencies (PCEs) and improved device stability compared to DPP-VC-CBZ polymers.²¹

Compared to previously reported DPP-based polymers used in DSSCs, the shifts in HOMO/LUMO levels observed here underscore a more precise tuning of bandgap via crosslinking. Notably, reducing the bandgap from 1.19 eV to 1.04 eV (DPP-OCT-CBZ to DPP-OCT-CBZ-C) suggests stronger intramolecular charge transfer, enabling enhanced charge generation and reduced recombination. This improvement in electrochemical stability upon crosslinking has not been extensively reported in prior DPP systems, indicating a new route for designing low-bandgap polymeric barriers with superior charge-transport characteristics. These results expand on established approaches by showing that side-chain engineering and UV-induced crosslinking can further lower the HOMO–LUMO gap while improving device-relevant electrochemical robustness.

4.3 Optical properties

The UV-vis absorption and fluorescence spectra (Fig. 4) provide important insights into the optical and photophysical properties (Table 3) of the synthesized DPP-VC-CBZ and DPP-OCT-CBZ polymers and their crosslinked counterparts (DPP-VC-CBZ-C and DPP-OCT-CBZ-C). Fig. 4(a) depicts that DPP-OCT-CBZ shows a broad absorption peak at around 600–700 nm, extending into the near-infrared region up to 1100 nm (both solid and solution) and has maximum absorption at 556 nm. This broad absorption is beneficial for capturing a wider spectrum of sunlight. DPP-VC-CBZ exhibits a narrower absorption peak cantered at around 500–600 nm and shows λ_max at 552 nm, with less absorption in the near-infrared region in the solution UV spectra.²²

Fig. 4 Normalized UV-vis absorption spectra of DPP-OCT-CBZ and DPP-VC-CBZ (a) in chloroform solution and (b) and (c) in thin films of both copolymers and their crosslinked counterparts in the range of 400–1100 nm, (d) photoluminescence spectra of DPP-PCT-CBZ and DPP-VC-CBZ in the range of 500–750 nm.

Table 3 Absorption properties of the respective polymers

Name of polymers	λ max solution (nm)	λ max solid (nm)	λ Onset,sol ^n (nm)	E g = 1240/λOnset (eV)
DPP-OCT-CBZ	556	560, 612, 850, 938	800	1.55
DPP-OCT-CBZ-C	—	561, 613, 851, 939	815	1.52
DPP-VC-CBZ	552, 709, 973	552,710, 976	600	2.06
DPP-VC-CBZ-C	—	557, 600, 830, 980	681	1.82

---

From Fig. 4(b) both the crosslinked (DPP-OCT-CBZ-C) and non-crosslinked (DPP-OCT-CBZ) polymers show almost similar absorption profiles with increased intensity, indicating that crosslinking significantly affects intensity of the absorption characteristics. The crosslinked DPP-VC-CBZ-C polymer shows a slightly red-shifted absorption peak compared to DPP-VC-CBZ in Fig. 3(c), indicating improved conjugation and potentially better charge transport properties. Notably, the crosslinked polymer films show a higher intensity absorption with a slightly narrower band gap of 1.52 eV and 1.82 eV compared to the non-crosslinked form at 1.55 eV and 2.04 eV in DPP-OCT-CBZ and DPP-VC-CBZ respectively. This reduction in the band gap, coupled with enhanced absorption at longer wavelengths, suggests that crosslinking effectively broadens the light-harvesting capabilities and improves the energy alignment in order to convert solar energy more effectively.²³ Lower energy absorption peaks (400–700 nm) correspond to intramolecular charge transfer bands (vide supra) where a higher energy band corresponds to π–π* transitions (800–1100 nm).

The absorption of DPP-based polymers and their crosslinked counterparts in the near-infrared range of 800–1100 nm and beyond allows these materials to effectively harness a wider spectrum of solar radiation, particularly NIR light, which is often underutilized by conventional photovoltaic materials. Having such extensive photon harvesting capabilities is essential to raising solar cells' quantum efficiency and overall power conversion efficiency. Moreover, the maintenance of absorption properties post-crosslinking not only suggests that the crosslinking process preserves the optical performance but also implies added mechanical and thermal stability, making these polymers robust candidates for durable, high-efficiency photovoltaic applications.²⁴ A redshift of approximately 200 nm was observed in the film (Fig. 4(b) and (c)) for DPP-OCT-CBZ compared to DPP-VC-CBZ, which is attributed to the formation of J-aggregates resulting from π–π stacking in the solid state, potentially leading to supramolecular assembly. In the gel state of DPP-OCT-CBZ-C, the absorption maximum exhibited a more pronounced redshift and a broader band, also indicative of π–π stacking and J-type aggregate formation within the supramolecular assembly. Additionally, small shoulder peaks at around 500 nm were noted in both the film and gel states of DPP-OCT-CBZ-C, attributed to the vibronic patterns of the aggregates. The tailing observed in the gel state is linked to scattering effects resulting from the gelation of the compound.^25,26

As shown in Fig. 4(d), the fluorescence spectra provide further insights into the photophysical properties of DPP-OCT-CBZ compared to DPP-VC-CBZ, highlighting their emission characteristics which are critical for evaluating their potential in photovoltaic applications. DPP-OCT-CBZ exhibits fluorescence peaks at 550 nm and 571 nm and a maximum at 594 nm, aligning with its broader UV-vis absorption spectrum. This broader emission profile suggests a higher ability to capture and convert a wider range of solar radiation into usable electronic excitations. The presence of multiple peaks also indicates a complex environment for photon emission, possibly due to varying molecular conformations or interactions within the polymer matrix that can affect the photovoltaic performance. DPP-VC-CBZ, on the other hand, shows fluorescence peaks at 576 nm and a maximum at 604 nm. The relatively narrower emission spectrum of DPP-VC-CBZ limits its light-harvesting capability to a smaller portion of the solar spectrum compared to DPP-OCT-CBZ. The narrower spectrum can result in reduced efficiency in solar energy applications where capturing a wide range of light wavelengths is beneficial for enhancing power conversion efficiency. The differences in the fluorescence spectra between DPP-OCT-CBZ and DPP-VC-CBZ reflect their distinct electronic structures and light absorption capabilities. DPP-OCT-CBZ, with its broader absorption and emission spectra, is better suited for photovoltaic devices aimed at maximizing solar energy capture and conversion, making it a more promising material for advanced solar cell technologies.²⁷

The optical and photophysical analyses indicate that DPP-OCT-CBZ, with its broader absorption spectrum, is better suited for capturing sunlight compared to DPP-VC-CBZ. The crosslinked versions of these polymers retain their beneficial properties while obtaining enhanced stability, indicating that they are potential high-performance candidates for DSSCs and other organic photovoltaic applications.^28,29 The DPP backbone inherently absorbs in the visible to near-infrared region. The polymer layer thus contributes to light harvesting, supplementing the dye in the DSSC and potentially enhancing photocurrent generation. Compared to the narrower absorption profiles commonly reported for DPP derivatives, the broad near-infrared absorption (up to ∼1100 nm) in both DPP-OCT-CBZ and DPP-OCT-CBZ-C signifies a key advance for low-bandgap polymer applications in DSSCs. Prior studies typically focus on visible-range absorption; here, the extended red-to-NIR absorption-retained even after crosslinking-enables more efficient solar photon harvesting. The redshift and increased intensity in crosslinked variants highlight a novel advantage of UV-induced crosslinking, which can stabilize the polymer conformation in a state conducive to strong π–π stacking and J-aggregate formation. This level of spectral broadening and absorption intensity enhancement goes beyond what has been achieved with simpler passivation layers, confirming the high potential of DPP-based materials in advanced DSSCs.

4.4 Thermal properties

The thermogravimetric analysis (TGA) curves presented in the Fig. 5(a) image provide insightful data regarding the thermal stability of DPP-based polymers, specifically DPP-OCT-CBZ, DPP-OCT-CBZ-C (crosslinked), DPP-VC-CBZ, and DPP-VC-CBZ-C (crosslinked). DPP-OCT-CBZ and DPP-VC-CBZ exhibit similar thermal behaviours with initial degradation starting slightly above 300 °C, which indicates good thermal stability suitable for processing and application in environments that may experience elevated temperatures. The complete decomposition occurs close to 500 °C, showing that these materials are capable of withstanding high temperatures before significant degradation. The crosslinked versions DPP-OCT-CBZ-C and DPP-VC-CBZ-C demonstrate enhanced thermal stability compared to their non-crosslinked counterparts as their degradation starts later than for the pure DPP. Notably, the onset of degradation for both crosslinked polymers is delayed by approximately 50 °C, beginning degradation at around 350 °C and maintaining a higher residual weight percentage at elevated temperatures up to 600 °C. This indicates that crosslinking significantly enhances the thermal stability of the polymers, likely due to the formation of a more rigid and thermally resistant network structure. The crosslinking clearly affects the thermal degradation process by increasing the thermal stability. The higher thermal stability in crosslinked polymers could be attributed to the increased crosslink density, which restricts molecular mobility and delays the decomposition process. Moreover, crosslinked polymers tend to form more char as a result of stronger and more numerous bonds within the material, which contributes to their improved stability at high temperature. The TGA data suggest that crosslinking is an effective strategy to enhance the thermal robustness of DPP-based polymers, making them more suitable for applications requiring high thermal resistance, such as in certain electronics and photovoltaic materials where long-term stability against thermal degradation is critical.³⁰ Relative to standard DPP polymers or alternative organic interlayers, the 50 °C higher onset of degradation for crosslinked DPP-OCT-CBZ-C and DPP-VC-CBZ-C surpasses the thermal stability commonly reported in the literature. This enhanced resilience is particularly meaningful for DSSC barrier layers that may experience elevated temperatures during prolonged operation. Demonstrating a robust crosslinked network that maintains structural integrity above 350 °C opens new possibilities for stable long-term device performance. In contrast to conventional TiO₂ blocking layers requiring high-temperature sintering, our mild-process crosslinking approach yields equally high (or better) thermal tolerance, highlighting a significant step forward in polymeric barrier design.

Fig. 5 (a) TGA curve of copolymers and their crosslinked counter polymers obtained between RT and 600 °C at a heating rate of 10 °C min⁻¹; (b) Horowitz and Metzger plots of activation energy for synthesized materials.

Thermogravimetric analysis (TGA) data can give useful information on the thermal breakdown kinetics of materials. The activation energy (E_a) of breakdown processes can be estimated using proper TGA data processing techniques. This work uses the Horowitz and Metzger approach (see Fig. 5(b)), which provides consistent activation energy estimates for thermal degradation of doped and dedoped polymers under nitrogen (N₂) conditions. This approach includes charting the double logarithm of the reciprocal of the reactant's weight fraction against temperature. In a first-order decomposition process, the reference temperature (T_s) equals the experimental temperature plus a constant (T = T_s + θ). At this reference point, the ratio of the residual mass (w) to the starting mass (w₀) is roughly 0.268, where w₀ is the original mass and w is the mass at T. An equation is obtained from this approximation:

ln(ln(w₀/w)) = (E_aθ/RT_s²)

The activation energy (E_a) of the polymers was determined using the relation E_a = slope × R × T_s², where R is the universal gas constant (8.314 J mol⁻¹ K⁻¹) and T_s is the characteristic degradation temperature. The slopes were obtained from the linear regions of the ln(w/w₀) versus θ= (T − T_s) plots. Based on this analysis, the calculated activation energies were 35.46 kJ mol⁻¹ for DPP-OCT-CBZ, 26.08 kJ mol⁻¹ for DPP-OCT-CBZ-C, 28.65 kJ mol⁻¹ for DPP-VC-CBZ, and 29.77 kJ mol⁻¹ for DPP-VC-CBZ-C. The variation in activation energies suggests that crosslinking slightly enhances the thermal stability of the DPP-based polymers by increasing the energy barrier for degradation. This analysis highlights the influence of structural modifications on the thermal degradation kinetics of these materials.

4.5 Morphological properties

The contact angle measurements presented in Fig. 6 provided for DPP-VC-CBZ, DPP-OCT-CBZ, and their crosslinked versions DPP-VC-CBZ-C and DPP-OCT-CBZ-C reveal significant insights into the effects of crosslinking and the specific molecular architecture of these polymers on their surface properties. DPP-VC-CBZ shows a contact angle of 64.03°, indicating moderate hydrophobicity. Upon crosslinking, DPP-VC-CBZ-C exhibits an increased contact angle of 79°. This increase can be attributed to the formation of a more rigid and structured network through crosslinking, which typically reduces the surface energy by limiting the mobility of polymer chains and decreasing the ability of polar groups to interact with water. DPP-VC-CBZ shows a contact angle of 64.03°, indicating moderate hydrophobicity. DPP-OCT-CBZ presents a slightly higher contact angle of 66.01° compared to DPP-VC-CBZ. The longer octene side chains in DPP-OCT-CBZ may contribute to this increase by introducing more hydrophobic character to the surface. However, it's the crosslinked form, DPP-OCT-CBZ-C, that shows a substantial increase in hydrophobicity with a contact angle of 82°. This notable increase is primarily due to the formation of the crosslinked network. This increase can be attributed to the formation of a more rigid and structured network through crosslinking, which typically reduces the surface energy by limiting the mobility of polymer chains and decreasing the ability of polar groups to interact with water.³¹ The dense, crosslinked network formed upon UV irradiation acts as a protective shield against moisture and oxygen. This is particularly important for DSSCs, which are otherwise prone to degradation due to environmental exposure.

Fig. 6 Water contact angle images of respective polymers DPP-OCT-CBZ and DPP-VC-CBZ spin coated over ITO (left site); images and photo-crosslinking effect images of DPP-OCT-CBZ-C and DPP-VC-CBZ-C (right site).

The AFM images (Fig. 7) of DPP-OCT-CBZ and its crosslinked version, DPP-OCT-CBZ-C, along with DPP-VC-CBZ, reveal notable differences in their surface structures. DPP-OCT-CBZ shows a relatively smooth surface with minor irregularities, which are typical for linear chains that can form semi-solid networks with a degree of surface roughness.³² Upon crosslinking, DPP-OCT-CBZ-C exhibits a more uniform surface, indicating that crosslinking helps in smoothing out the surface irregularities, likely by filling gaps at the molecular level and creating a denser network. In contrast, the DPP-VC-CBZ images display a significantly rougher surface with visible particulate inclusions and striations, suggesting less uniformity and more pronounced surface features, which may affect its light absorption and charge transport properties in photovoltaic applications. Also, DPP-OCT-CBZ shows a roughness of 2.7 nm, which decreases to 2.2 nm upon crosslinking in DPP-OCT-CBZ-C. Similarly, DPP-VC-CBZ starts with a roughness of 4.2 nm, which reduces to 3.2 nm in its crosslinked version, DPP-VC-CBZ-C, as given in Table 4. Such a reduction in roughness can lead to more effective light absorption and enhanced charge transport by minimizing surface defects that can trap charges or scatter light, thus potentially increasing the efficacy and stability of DSSCs.¹³

Fig. 7 AFM images of polymers DPP-OCT-CBZ and DPP-VC-CBZ spin coated over ITO (left site); topography images and photo-crosslinking effect images of DPP-OCT-CBZ-C and DPP-VC-CBZ-C (right site).

Table 4 AFM RMS values of studied materials

Polymers	RMS
DPP-OCT-CBZ	2.7 nm
DPP-OCT-CBZ-C	2.2 nm
DPP-VC-CBZ	4.2 nm
DPP-VC-CBZ-C	3.2 nm

---

In particular, the DPP-VC-CBZ film (RMS 4.2 nm) exhibits noticeable particulate inclusions and striations in its AFM image, which contribute to its higher roughness. Upon crosslinking (DPP-VC-CBZ-C, RMS 3.2 nm), these irregular features are reduced, yielding a smoother film. This improved morphology is expected to reduce light-scattering sites and minimize charge-trapping surface irregularities, thereby benefiting the overall photonic and electronic performance. Crosslinking stabilizes the polymer film morphology over time, reducing phase separation or crystallization that can occur under heat or light stress. This leads to greater thermal and long-term stability of the device.

While many polymeric interlayers reduce surface roughness to some extent, the degree of RMS smoothing observed here (from 2.7 nm down to 2.2 nm for DPP-OCT-CBZ and from 4.2 nm down to 3.2 nm for DPP-VC-CBZ) is notable and further substantiates the advantage of crosslinking in achieving uniform coverage. This level of morphological control, combined with the hydrophobicity increase (e.g., contact angles increasing to 82° for DPP-OCT-CBZ-C), surpasses that of previously reported DPP films and ensures better moisture resistance. Such a smooth, water-resistant surface has rarely been reported for DPP-derived DSSC barrier layers, indicating a distinct pathway to reducing charge-trapping defects and extending device lifetime under real-world operating conditions.

4.6 Photovoltaic performance

The photovoltaic performance data for DPP-based polymers, both non-crosslinked and crosslinked, reveal significant insights into how crosslinking influences the dye-sensitized solar cell (DSSC) performances with active area 0.25 cm² (see Fig. 8(a) and (b) and Table 5). At 1 sun, DPP-OCT-CBZ-C (crosslinked linear) shows a higher J_sc value (14.5 mA cm⁻² from 11.6 mA cm⁻²) and slightly higher V_oc (0.75 V from 0.73 V) and FF (67% from 65%) compared to the non-crosslinked DPP-OCT-CBZ. This translates to a PCE of 7.28%, a significant increase over 5.50% for the non-crosslinked DPP-OCT-CBZ; at the same time DPP-OCT-CBZ-C maintains a higher V_oc (0.73 V vs. 0.62 V) and FF (52% vs. 45%), resulting in an indoor PCE of 18.2%, compared to 16.3% for non-crosslinked polymers. By contrast, DPP-VC-CBZ-C (crosslinked cyclic) exhibits a more modest performance gain over DPP-VC-CBZ. Under 1 sun, the crosslinked cyclic polymer has nearly the same J_sc (10.98 mA cm⁻² from 10.11 mA cm⁻²) and V_oc (0.72 V vs. 0.71 V) as the non-crosslinked cell, with a small change in the FF (60% vs. 62%). The net result is only a slight increase in efficiency (4.74% vs. 4.38%, ∼8% improvement). At 800 lux, crosslinked DPP-VC-CBZ yields a moderate downtick in J_sc (3.12 mA cm⁻²vs. 2.64 mA cm⁻²) but an uptick in V_oc (0.57 V vs. 0.55 V), while the FF actually decreases slightly (62% vs. 60%). Consequently, the PCE increases from about ∼13.2% to ∼14.8% – a marginal gain relative to the improvement seen for the linear polymer. The crosslinked network of DPP-OCT-CBZ clearly enhances performance, especially boosting the photocurrent and stabilizing the photovoltage. This improvement can be attributed to several factors: (i) higher dye loading/retention: the crosslinked polymer likely fixes more dye on the surface as the synthesized crosslinked polymers absorb near IR so it can absorb more solar spectrum (preventing desorption in the liquid electrolyte) and may allow a denser packing, thus absorbing more light and generating higher J_sc. (ii) Reduced recombination: the polymer layers can act as a physical barrier to electrolyte species, suppressing electron recombination at the TiO₂ interface, which raises V_oc and yields a more ideal diode behavior (higher FF). Indeed, the crosslinked DPP-OCT-CBZ device shows a ∼20 mV V_oc gain and slightly improved FF relative to its non-crosslinked counterpart, consistent with suppressed back-electron transfer. These results indicate that simply crosslinking the bulky cyclic side-chain dye provides limited benefits; the non-crosslinked and crosslinked DPP-VC-CBZ devices perform similarly, with only minor differences in charge transport characteristics. The slight drop in the FF upon crosslinking (especially under 1 sun) suggests that the crosslinked VC dye layer may introduce a bit of resistance or irregularity which is possibly due to a less uniform polymer network or reduced pore infiltration by electrolyte, partially offsetting its recombination suppression benefits. In summary, crosslinking is far more effective for the linear-chain polymer than for the cyclic-chain polymer: it substantially boosts J_sc, V_oc, and FF for DPP-OCT-CBZ (hence PCE), whereas for DPP-VC-CBZ the changes are smaller and the FF can even decline slightly. At 800 lux, the reduced light intensity typically lowers the overall photocurrent (J_sc) across all devices, but the crosslinked film's better surface passivation and improved film morphology further suppress recombination, thus boosting V_oc and FF. The consistent increase in the FF and efficiency upon crosslinking across both polymer types highlights the efficacy of crosslinking in enhancing photovoltaic performance by optimizing electronic properties for more effective solar energy conversion.¹⁴ The increase in V_oc after crosslinking is likely due to a more significant reduction in charge recombination. The stable crosslinked structure effectively separates the electrons in the semiconductor from the oxidized dye and electrolyte, thus enhancing V_oc.

Fig. 8 J–V curves of respective polymers as barrier layers where Pt acts as a counter electrode (a) under 800 lux and (b) under 1 sun. (c) The EIS of the respective devices FTO/TiO₂/N719/pPolymers/electrolyte/Pt; (d) efficiency with error bars for DSSCs with barrier layers.

Table 5 Photovoltaic device properties of studied materials under 800 lux (LED) and a 1 sun for DSSC (aaverage values are over three cells of each type)

Materials	Power W m^−2	800 lux DSSC				1 sun DSSC				Resistance
		J sc (mA cm^−2)	V oc (V)	FF (%)	Eff (%)	J sc (mA cm^−2)	V oc (V)	FF (%)	Eff (%)	R s (Ω)	R ct (Ω)
DPP-OCT-CBZ	67.4	3.39 ± 0.065	0.67 ± 0.005	45 ± 0.82	16.3 ± 0.49	11.6 ± 0.57	0.73 ± 0.008	65 ± 0.82	5.50 ± 0.34	48	123
DPP-OCT-CBZ-C	64.2	3.08 ± 0.069	0.73 ± 0.008	52 ± 0.81	18.2 ± 0.66	14.5 ± 0.49	0.75 ± 0.008	67 ± 0.82	7.28 ± 0.37	46	104
DPP-VC-CBZ	71.3	3.12 ± 0.19	0.7 ± 0.005	45 ± 0.82	13.2 ± 0.99	10.11 ± 0.54	0.71 ± 0.008	62 ± 0.81	4.38 ± 0.21	42	178
DPP-VC-CBZ-C	69.1	2.64 ± 0.24	0.72 ± 0.009	55 ± 0.82	14.7 ± 1.51	10.98 ± 0.15	0.72 ± 0.008	60 ± 0.82	4.74 ± 0.062	43	154

---

Under 1 sun, the non-crosslinked DPP-OCT-CBZ already shows ∼25% higher efficiency than DPP-VC-CBZ (5.50% vs. 4.38%), stemming from both a higher J_sc (+1.5 mA cm⁻²) and slightly higher FF and V_oc. This suggests that the intrinsic properties of the linear side chain favors higher photocurrent and fill factor. After crosslinking, the performance gap between linear and cyclic polymers further widens – DPP-OCT-CBZ-C achieves nearly double the PCE of DPP-VC-CBZ-C under 1 sun (7.28% vs. 4.74%) and about 22% improvement under indoor light (∼18.2% vs. ∼14.8%). The crosslinked linear polymer's J_sc is 25% higher than that of the crosslinked cyclic polymer barriers in both cases, indicating much lower recombination. These differences highlight how side-chain structures influence the effectiveness of crosslinking: the linear chain not only provides a better baseline performance but also responds to crosslinking with larger gains, whereas the cyclic chain yields lower initial performance and smaller improvement. The linear octenyl chain is flexible and can lie relatively flat along the TiO₂ anchored dye surface, likely enabling dyes to pack closely together prior to crosslinking. This may lead to some π–π stacking or aggregation in the non-crosslinked state (potentially harming V_ocvia aggregation-induced recombination), but once crosslinked, the linear chains can form an interlinked polymer that covers the surface more uniformly, minimizing dye aggregation and maximizing surface coverage. The result is a high dye density with strong light absorption (boosting J_sc) and a continuous organic network that effectively passivates the TiO₂ surface (boosting V_oc/FF by impeding recombination). In contrast, the bulky VC side chain likely forces greater spacing between dye molecules due to steric hindrance. This could reduce dye loading (limiting J_sc) and also reduce deleterious dye aggregation initially. The more upright or spaced-out VC dyes leave more exposed TiO₂, explaining the lower V_oc and FF for DPP-VC-CBZ vs. DPP-OCT-CBZ. When crosslinked, the cyclic substituents – due to their rigidity – might form only a loosely connected network with gaps. Thus, DPP-VC-CBZ-C results in only a small increase in J_sc and V_oc over its non-crosslinked form. In fact, any beneficial reduction in recombination for the cyclic dye upon crosslinking appears to be minor, and it may be offset by a slight decrease in electrode wettability or charge mobility (hence the small FF drop). In summary, the linear side chain is more conducive to high-performance DSSCs, especially when combined with post-adsorption polymerization, whereas the cyclic side chain, while offering structural rigidity, results in lower initial performance and a diminished crosslinking payoff.³³

In the EIS analysis (see Fig. 8(c) and Table 5), DPP-OCT-CBZ-C demonstrates the lowest R_s (46 Ohms) and R_ct, signifying superior conductivity and charge transfer, which directly contributes to its higher efficiency, 18.2% short circuit current and respectable open circuit voltage. This improvement is attributed to the UV-crosslinking process that reduces charge recombination. DPP-OCT-CBZ, with R_s at ∼43 Ohms and higher R_ct than its crosslinked counterpart, shows slightly lower efficiency 16.3%, but its larger semicircle in the Nyquist plot indicates more charge transfer resistance. On the other hand, the DPP-VC series shows relatively higher R_s and R_ct, with DPP-VC-CBZ-C having R_s around 50 Ohms and a moderate semicircle in the Nyquist plot. This corresponds to lower PCE 13.2% and reduced performance, as crosslinking via UV does not reduce resistance as effectively here. DPP-VC-CBZ, with an R_s of ∼42 Ohms, shows the highest R_ct and lower overall performance, correlating with its lower J_sc and reduced efficiency of 14.7% but higher than its crosslinked counterpart. The structural influence of the long-chain octene in DPP-OCT-CBZ appears to offer better charge mobility than the vinyl benzyl side chain in DPP-VC-CBZ, with UV crosslinking further enhancing performance, especially in DPP-OCT-CBZ-C.³⁴

The bar graph illustrates in Fig. 8(d) the power conversion efficiencies (PCEs) of DSSCs incorporating different DPP-based polymer barrier layers under 1 sun (AM1.5G) and 800 lux (LED) illumination, clearly showing the impact of polymer structures and crosslinking. Among all, DPP-OCT-CBZ-C (crosslinked linear side chain) exhibits the highest efficiencies—approximately 7.3% under 1 sun and ∼18.2% under 800 lux—attributed to its dense, uniform morphology and effective recombination suppression. In contrast, the non-crosslinked DPP-OCT-CBZ shows slightly lower efficiencies (5.5% and ∼16.8%, respectively), highlighting the advantage of UV-induced crosslinking. DPP-VC-CBZ-C (crosslinked cyclic side chain) offers moderate performance (∼4.7% and ∼14.7%), while DPP-VC-CBZ shows the lowest performance (∼4.3% and ∼13.2%), suggesting that linear side chains and crosslinking together yield the most effective barrier layers for enhancing both indoor and outdoor DSSC performance.

In summary, the superior performance of the crosslinked DPP-OCT-CBZ polymer can be attributed to more efficient charge extraction and higher recombination resistance arising from its linear side-chain structure and the crosslinked network it forms. The linear octenyl side chain enables a high dye load upon crosslinking, creates a tightly packed, polymerized layer that enhances light absorption, lowers recombination and provides a protective coating on the TiO₂ anchored dye. This leads to higher PCE, especially evident under low-light conditions. The cyclic vinyl side chain, in contrast, results in a less effective crosslinked barrier, yielding lower currents and more recombination (lower V_oc/FF). The trends observed – better J_sc, slightly increased V_oc, and improved FF with crosslinking (particularly for the linear polymer barrier)-align with the mechanistic understanding that better charge recombination and surface passivation improve DSSC performance. These findings demonstrate that molecular engineering of polymer side chains, combined with post-adsorption crosslinking, can markedly influence charge transport dynamics and efficiency in DSSCs. Optimizing both the DSSC's structure and its supramolecular organization on the electrode is key to maximizing performance, whether under full sun or dim indoor light. The crosslinked linear DPP-OCT-CBZ-C, in particular, exemplifies how a tailored structure can achieve excellent efficiency across diverse lighting conditions – offering >7% under 1 sun and ∼19% under an 800 lux LED – by facilitating high photocurrents and mitigating recombination losses (Fig. 9). Such design principles are valuable for developing DSSCs for both outdoor and indoor photovoltaic application.³⁵ Comparison of the photovoltaic performance of our crosslinked DPP-based barrier layers with selected literature on DSSCs featuring alternative barrier layers under indoor and outdoor conditions is shown in Table 6.

Fig. 9 Mechanism of a DSSC with a photocrosslinked barrier layer showing it suppressing recombination of charges and ensuring stability.

Table 6 Comparison of the photovoltaic performance of our crosslinked DPP-based barrier layers with reported DSSCs featuring alternative barrier layers under indoor and outdoor conditions

Barrier layers	J sc (mA cm^−2)	V oc (V)	FF	Efficiency (%)	Application	Ref.
ZnO	11.7	0.62	52	4.51	1 Sun DSSC	36
ZrO2	14.03	0.77	62	6.72	1 Sun DSSC	37
Al2O3	12.46	0.76	68	6.50	1 Sun DSSC	38
In2O3	7.61	0.83	55	3.49	1 Sun SS-DSSC	39
TiCl4	0.084	0.703	82	15.26	1000 lux	40
DPP-OCT-CBZ-C	3.08	0.73	52	18.2	800 lux LED	This work
DPP-OCT-CBZ-C	14.5	0.75	67	7.28	1 Sun DSSC	This work

---

4.7 Conclusion

The comprehensive analysis of DPP-based polymers, including DPP-OCT-CBZ and DPP-VC-CBZ, along with their crosslinked variants, has revealed several key enhancements in their properties crucial for photovoltaic applications. The cyclic voltammetry (CV) analysis demonstrated improvements in electrochemical stability upon crosslinking. UV-vis spectroscopy highlighted broadened absorption in the near-infrared range, essential for efficient solar energy capture, with notable stability improvements evidenced in the TGA analysis, where thermal degradation thresholds were raised significantly. Contact angle measurements indicated increased hydrophobicity post-crosslinking, with DPP-OCT-CBZ-C showing an increase from 66.01° to 82°, suggesting improved moisture resistance. AFM images revealed a smoother surface morphology in crosslinked variants, enhancing the optical clarity and uniformity critical for solar cell efficiency. Finally, the photovoltaic performance data showcased that crosslinking led to higher efficiencies, with DPP-OCT-CBZ-C achieving an impressive 18.2% under 800 lux and 7.28% under 1 sun efficiency due to improved charge extraction and reduced recombination. These collective improvements across various physical and electrochemical dimensions underscore the effectiveness of crosslinking in optimizing DPP-based polymers for enhanced solar cell performance.

Data availability

Data will be available upon request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support from the Science and Engineering Research Board (SERB), Govt. of India, through the Core Research Grant (CRG) (File no. CRG/2018/002880) and the Department of Chemicals and Petrochemicals (DCPC), Govt. of India, for the Centre of Excellence [file no. 25014/2/2015-PC-II (FTS:8418)] in DSSC research is gratefully acknowledged.

References

1. A. G. Al-Azzawi, S. B. Aziz, E. M. Dannoun, A. Iraqi, M. M. Nofal, A. R. Murad and A. M. Hussein, A mini review on the development of conjugated polymers: Steps towards the commercialization of organic solar cells, Polymers, 2022, 15(1), 164.
2. Y. He, Novel N-type π-Conjugated Polymers for All-Polymer Solar Cells, Doctoral Dissertation, Université de Bordeaux, University of Waterloo, Canada, 2017.
3. P. K. Yadav, S. Kumar, and P. Maiti, Conjugated polymers for solar cell applications, in Conjugated Polymers for Next-Generation Applications, Woodhead Publishing, 2022, pp. 367–401.
4. X. Song, N. Gasparini, M. M. Nahid, S. H. K. Paleti, C. Li, W. Li and D. Baran, Efficient DPP donor and nonfullerene acceptor organic solar cells with high photon-to-current ratio and low energetic loss, Adv. Funct. Mater., 2019, 29(34), 1902441.
5. Y. Li, P. Sonar, L. Murphy and W. Hong, High mobility diketopyrrolopyrrole (DPP)-based organic semiconductor materials for organic thin film transistors and photovoltaics, Energy Environ. Sci., 2013, 6(6), 1684–1710.
6. J. Lee, H. J. Park, J. M. Joo and D. H. Hwang, Synthesis and characterization of DPP-based conjugated polymers via Dehydrogenative direct Alkenylation Polycondensation, Macromol. Res., 2019, 27(2), 115–118.
7. C. Li, A. Zhang, Z. Wang, F. Liu, Y. Zhou, T. P. Russell and W. Li, All polymer solar cells with diketopyrrolopyrrole-polymers as electron donor and a naphthalenediimide-polymer as electron acceptor, RSC Adv., 2016, 6(42), 35677–35683.
8. D. Zhang, D. Liang, L. Gu, J. Li and H. Zhang, DPP-based polymers with linear/branch side chain for organic field-effect transistors, Front. Chem., 2022, 10, 1008807.
9. S. Dahlstrom, S. Wilken, Y. Zhang, C. Ahlang, S. Barlow, M. Nyman and R. Österbacka, Cross-linking of doped organic semiconductor interlayers for organic solar cells: potential and challenges, ACS Appl. Energy Mater., 2021, 4(12), 14458–14466.
10. C. Zhang and S. Holdcroft, Photocrosslinking of low band-gap conjugated polymers using alkyl chloride sidechains: Toward high-efficiency, thermally stable polymer solar cells, J. Mater. Res., 2018, 33(13), 1879–1890.
11. L. Ye, Y. Yang, C. Liu, X. Duan, S. Wang, W. Li and Y. Sun, Directly Cross-Linked Conjugated Polymer Donor Enables Efficient Polymer Solar Cells with Extraordinary Mechanical Robustness, Small, 2023, 19(41), 2303226.
12. Y. J. Noh, Y. J. Choi, J. H. Jeong, S. S. Kim, K. U. Jeong and S. I. Na, Photo-cross-linked perylene diimide derivative materials as efficient electron transporting layers in inverted polymer solar cells, Nanoscale, 2017, 9(45), 17731–17736.
13. H. L. Hsu, Y. C. Chao, Y. H. Liao, C. L. Chung, Y. J. Peng, C. P. Chen and R. J. Jeng, Embedding a diketopyrrolopyrrole-based cross-linking interfacial layer enhances the performance of organic photovoltaics, ACS Appl. Mater. Interfaces, 2018, 10(10), 8885–8892.
14. P. Panigrahi, M. K. Mallick, S. Mohanty and A. K. Palai, Influence of donor and acceptor moieties on photophysical properties of N-naphthylcarbazole based conjugated terpolymers, Opt. Mater., 2023, 138, 113692.
15. A. G. Al Lafi, A. Alzier and A. W. Allaf, Two-dimensional FTIR spectroscopic analysis of crystallization in cross-linked poly (ether ether ketone), Int. J. Plast. Technol., 2020, 24, 1–8.
16. A. Nyayachavadi, G. T. Mason, M. Nazir Tahir, M. U. Ocheje and S. Rondeau-Gagné, Covalent cross-linking of diketopyrrolopyrrole-based organogels with polydiacetylenes, Langmuir, 2018, 34(40), 12126–12136.
17. S. Ghosh, S. Das, A. Saeki, V. K. Praveen, S. Seki and A. Ajayaghosh, A hybrid organogel of a low band gap diketopyrrolopyrrole with PC71BM: Phase separated morphology and enhanced photoconductivity, ChemNanoMat, 2018, 4(8), 831–836.
18. D. Stegerer, M. Pracht, F. Günther, H. Sun, K. Preis, M. Zerson and M. Sommer, Organogels from diketopyrrolopyrrole copolymer ionene/polythiophene blends exhibit ground-state single electron transfer in the solid state, Macromolecules, 2022, 55(12), 4979–4994.
19. K. Sun, X. Tang, Y. Ran, R. He, W. Shen and M. Li, π-Bridge modification of thiazole-bridged DPP polymers for high performance near-IR OSCs, Phys. Chem. Chem. Phys., 2018, 20(3), 1664–1672.
20. M. N. Shah, Y. Yang, M. Sohail, S. R. Pathipati and X. Pan, Low-Bandgap DPP-Based Quinoxaline with Extended EQE and Low Energy Loss for Efficient Polymer Solar Cells, J. Electron. Mater., 2021, 50(8), 4488–4496.
21. Z. Liu, Y. Hu, P. Li, J. Wen, J. He and X. Gao, Enhancement of the thermoelectric performance of DPP based polymers by introducing one 3, 4-ethylenedioxythiophene electron-rich building block, J. Mater. Chem. C, 2020, 8(31), 10859–10867.
22. G. Oklem, X. Song, L. Toppare, D. Baran and G. Gunbas, A new NIR absorbing DPP-based polymer for thick organic solar cells, J. Mater. Chem. C, 2018, 6(12), 2957–2961.
23. A. Nyayachavadi, A. K. Sur, P. Kulatunga, Y. Wang, T. C. Gomes, M. Mooney and S. Rondeau-Gagné, Covalent Topochemical Photo-Cross-Linking of Diketopyrrolopyrrole-Based Polymers with Polydiacetylenes toward Tunable Mechanical and Electronic Properties, Chem. Mater., 2023, 35(22), 9682–9691.
24. G. Qian, J. Qi and Z. Y. Wang, Synthesis and study of low-bandgap polymers containing the diazapentalene and diketopyrrolopyrrole chromophores for potential use in solar cells and near-infrared photodetectors, J. Mater. Chem., 2012, 22(25), 12867–12873.
25. Y. T. Shen, C. H. Li, K. C. Chang, S. Y. Chin, H. A. Lin, Y. M. Liu and S. S. Sun, Synthesis, optical, and mesomorphic properties of self-assembled organogels featuring phenylethynyl framework with elaborated long-chain pyridine-2, 6-dicarboxamides, Langmuir, 2009, 25(15), 8714–8722.
26. X. Q. Li, X. Zhang, S. Ghosh and F. Würthner, Highly fluorescent lyotropic mesophases and organogels based on J-aggregates of core-twisted perylene bisimide dyes, Chem.–Eur. J., 2008, 14(27), 8074–8078.
27. M. F. Falzon, A. P. Zoombelt, M. M. Wienk and R. A. Janssen, Diketopyrrolopyrrole-based acceptor polymers for photovoltaic application, Phys. Chem. Chem. Phys., 2011, 13(19), 8931–8939.
28. M. Más-Montoya and R. A. Janssen, The effect of H- and J*aggregation on the photophysical and photovoltaic properties of small thiophene–pyridine–DPP molecules for bulk-heterojunction solar cells, Adv. Funct. Mater., 2017, 27(16), 1605779.
29. J. W. Chung, B. K. An and S. Y. Park, A thermoreversible and proton-induced gel–sol phase transition with remarkable fluorescence variation, Chem. Mater., 2008, 20(21), 6750–6755.
30. F. Bella, E. D. Ozzello, S. Bianco and R. Bongiovanni, Photo-polymerization of acrylic/methacrylic gel–polymer electrolyte membranes for dye-sensitized solar cells, Chem. Eng. J., 2013, 225, 873–879.
31. S. Ghosh, S. Shankar, D. S. Philips and A. Ajayaghosh, Diketopyrrolopyrrole-based functional supramolecular polymers: next-generation materials for optoelectronic applications, Mater. Today Chem., 2020, 16, 100242.
32. X. Yu, C. Li, C. Gao, L. Chen, X. Zhang, G. Zhang and D. Zhang, Enhancing the healing ability and charge transport thermal stability of a diketopyrrolopyrrole based conjugated polymer by incorporating coumarin groups in the side chains, J. Polym. Sci., 2022, 60(3), 517–524.
33. G. S. Thool, K. Narayanaswamy, A. Venkateswararao, S. Naqvi, V. Gupta, S. Chand and S. P. Singh, Highly directional 1D supramolecular assembly of new diketopyrrolopyrrole-based gel for organic solar cell applications, Langmuir, 2016, 32(17), 4346–4351.
34. P. Panigrahi, M. K. Mallick, S. Mohanty, S. K. Nayak and A. K. Palai, EDOT Based Terpolymers: Facile Synthesis via Direct C–H Arylation and Barrier Layer in Dye Sensitized Solar Cells, ChemistrySelect, 2020, 5(28), 8674–8678.
35. Y. J. Noh, Y. J. Choi, J. H. Jeong, S. S. Kim, K. U. Jeong and S. I. Na, Photo-cross-linked perylene diimide derivative materials as efficient electron transporting layers in inverted polymer solar cells, Nanoscale, 2017, 9(45), 17731–17736.
36. S. J. Roh, R. S. Mane, S. K. Min, W. J. Lee, C. D. Lokhande and S. H. Han, Achievement of 4.51% conversion efficiency using ZnO recombination barrier layer in TiO₂ based dye-sensitized solar cells, Appl. Phys. Lett., 2006, 89(25), 253512.
37. X. Yang, L. Zhao, K. Lv, B. Dong and S. Wang, Enhanced efficiency for dye-sensitized solar cells with ZrO₂ as a barrier layer on TiO₂ nanofibers, Appl. Surf. Sci., 2019, 469, 821–828.
38. C. Lin, F. Y. Tsai, M. H. Lee, C. H. Lee, T. C. Tien, L. P. Wang and S. Y. Tsai, Enhanced performance of dye-sensitized solar cells by an Al₂O₃ charge-recombination barrier formed by low-temperature atomic layer deposition, J. Mater. Chem., 2009, 19(19), 2999–3003.
39. T. P. Brennan, J. T. Tanskanen, K. E. Roelofs, J. W. To, W. H. Nguyen, J. R. Bakke and S. F. Bent, TiO₂ conduction band modulation with In₂O₃ recombination barrier layers in solid-state dye-sensitized solar cells, J. Phys. Chem. C, 2013, 117(46), 24138–24149.
40. I. P. Liu, W. H. Lin, C. M. Tseng-Shan and Y. L. Lee, Importance of compact blocking layers to the performance of dye-sensitized solar cells under ambient light conditions, ACS Appl. Mater. Interfaces, 2018, 10(45), 38900–38905.

---