Versatile aza‐BODIPY‐based low‐bandgap conjugated small molecule for light harvesting and near‐infrared photodetection

Abstract

The versatile nature of organic conjugated materials renders their flawless integration into a diverse family of optoelectronic devices with light‐harvesting, photodetection, or light‐emitting capabilities. Classes of materials that offer the possibilities of two or more distinct optoelectronic functions are particularly attractive as they enable smart applications while providing the benefits of the ease of fabrication using low‐cost processes. Here, we develop a novel, multi‐purpose conjugated small molecule by combining boron‐azadipyrromethene (aza‐BODIPY) as electron acceptor with triphenylamine (TPA) as end‐capping donor units. The implemented donor–acceptor–donor (D–A–D) configuration, in the form of TPA‐azaBODIPY‐TPA, preserves ideal charge transfer characteristics with appropriate excitation energy levels, with the additional ability to be used as either a charge transporting interlayer or light‐sensing semiconducting layer in optoelectronic devices. To demonstrate its versatility, we first show that TPA‐azaBODIPY‐TPA can act as an excellent hole transport layer in methylammonium lead triiodide (MAPbI₃)‐based perovskite solar cells with measured power conversion efficiencies exceeding 17%, outperforming control solar cells with PEDOT:PSS by nearly 60%. Furthermore, the optical bandgap of 1.49 eV is shown to provide significant photodetection in the wavelength range of up to 800 nm where TPA‐azaBODIPY‐TPA functions as donor in near‐infrared organic photodetectors (OPDs) composed of fullerene derivatives. Overall, the established versatility of TPA‐azaBODIPY‐TPA, combined with its robust thermal stability as well as excellent solubility and processability, provides a new guide for developing highly efficient multi‐purpose electronic materials for the next‐generation of smart optoelectronic devices.

INTRODUCTION

Organic optoelectronic devices have attracted great attention over the past two decades for their exceptional characteristics and potential in wearable electronics. The versatility of this platform technology is set to transform the current way we use and interact with electronics in our everyday lives.^1–8 Among the most prominent organic and hybrid devices developed to date are organic photovoltaics (OPVs),⁹ perovskite solar cells (PSCs),¹⁰ organic photodetectors (OPDs),¹¹ organic light-emitting diodes (OLEDs),¹² and organic field-effect transistors (OFETs),¹³ as well a large family of chemical and biosensors.^8,14–16 The main features, which set these modern devices apart from their classical counterparts, are the mechanical flexibility, higher transparency, reduced weight, possibility of color tuning, scalability and solution-based deposition methods over large area on variety of rigid and flexible substrates. Although a great deal of efforts has been invested in improving the performance and stability of these soft electronic devices,^17,18 especially toward commercialization, the advancement in the field still relies, to a large extent, on developing novel low-cost organic semiconductors (OSCs) with superior optoelectronic characteristics and versatility.^19–21 In this context, their versatility or multifunctionality, that is, the material's ability to offer two or more distinct electronic functions, remains one of the most unique and attractive feature of organic electronic materials, expanding their possible applications even further while still benefiting from the ease of manufacture and low cost.

Historically, the concept of multifunctionality has emerged from three distinct groups of optoelectronic devices. The first one involves vertically stacked device architectures (often described as “tandem”) in which individual functionalities such as light emission or detection are incorporated separately using an intermediate connecting electrode. Dual mode OPV-OLED devices, which are capable of light harvesting and photon emission, are the perfect example of such systems, as theoretically they enable the possibility of stand-alone operation without the need of external power sources.^22–24 On this front, interactive displays for mobile handheld devices and smart window blinds are among the most exciting applications reported in recent times.²³ Furthermore, bifunctional devices with light emission (OLED) and sensing (OPD) abilities have also been demonstrated, with the prospect of manufacturing ultra-lightweight handheld emission panels incorporating a scanner.²⁵

The second family of multifunctional devices relies on a single architecture in which dual or multiple functions are achieved by changing the device operating conditions, typically by switching/changing the voltage polarity or the number of terminals. Among important examples are the bifunctional photoresponsive OFETs which act as photodetectors and light-stimulated synaptic transistors, with the gate voltage used as the switch.²⁶ Highly efficient organic electrochemical transistors (OECTs) implementing an additional function of resistive humidity sensors have also been reported.²⁷ Recently, our group has demonstrated multifunctionality using a single device architecture based on tetracene that combines three functions in one: OPV, OPD, and OLED.²⁸ Furthermore, we were also successful in extending the principle of multi-functional operation to new applications as an optical force sensor that benefits from unique bifunctionality (OLED and OPD modes) of rubrene-based organic diodes.²⁹

Such multifunctionality can also be tailored at the molecular scale design for new generation of OSCs. Here, typically a single material is engineered to fulfill two or more distinct functions in separated devices, each featuring a carefully adapted architecture. Conferring large versatility and adaptability to OSCs can be challenging yet rewarding as it greatly broadens possible applications while reducing processing methods and associated costs. Among notable examples, anthanthrone-based polymers have been demonstrated to be a promising versatile building block for the construction of semiconductors for multi-purpose applications including OFETs, OPVs, OLEDs, and chemical sensors.³⁰ In our recent work, we successfully demonstrated the use of a naphthalene flanked diketopyrrolopyrrole (DPP) dye as an active layer in three devices including OLEDs, single crystal OFETs and a thin film transistor.³¹ In addition, in our recent hybrid semiconducting material system, we layered a black phosphorous (BP) 2D material together with a low-bandgap donor–acceptor selenophene-flanked DPP-thienyl-vinylene-thienyl polymer (PDPPSe–TVT) in photonic devices. The active layer enhances both the optical absorption and ambipolar field-effect transistor characteristics in phototransistor devices. Enhancements in photodetection and carrier mobility devices using the same polymer demonstrated the multifunctionality aspects.³² Recently, multifunctional, stretchable silver nanowire-embedded chitosan biopolymer organic/hybrid films have been proposed for applications in organic electronics (OLEDs), heaters, and strain sensors.³³ The bifunctionality of heterocyclic spiro derivatives has also been achieved, with a single molecule acting efficiently as a photoactive material for OPVs and as emitter for OLEDs.³⁴

To further extend the above concept of multifunctionality, this work implements a novel synthetic strategy to modulate the optoelectronic properties of the boron-azadipyrromethene (aza-BODIPY)-based small molecule by attaching triphenylamine (TPA) donor units as the end-capping groups. Over the past years, 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) derivatives have been broadly reported as attractive materials in the fields of biochemistry and organic electronics, offering widespread applications and potential for multifunctionality.^35–38 Hence, owning to the structural versatility of BODIPY that enables fine-tuning of its photophysical and optoelectronic properties, we engineer a novel donor–acceptor–donor (D–A–D) configuration, in form of TPA-azaBODIPY-TPA, that offers a highly desirable absorption redshift, optimal charge transfer characteristics, and excitation energy levels, while maintaining high solubility and processability. Materials featuring dual functions of both electron accepting and donating are also reported as an effective strategy to improve the efficiency of organic optoelectronic devices.³⁹ In this work, we show that the newly developed TPA-azaBODIPY-TPA conjugated small molecule acts as an ultra-thin (6 nm) hole transport layer (HTL) in PSCs to achieve superior performance in comparison to control PSCs based on the conventional PEDOT:PSS HTL. Power conversion efficiencies (PCEs) of over 17% highlight the material's promising utilization in next-generation light-harvesting PSCs with robust thermal stability. Furthermore, owning to its low optical bandgap (1.49 eV), TPA-azaBODIPY-TPA is also implemented as a donor material together with [6,6]-phenyl-C71-butyric acid methyl ester (PC₇₁BM) acceptor as a 90-nm-thick blended active layer in near-infrared (NIR) OPDs. Optimized devices deliver robust light detection for photons in the wavelength range up to 800 nm, featuring a high responsivity of 77 mA W⁻¹, a large linear dynamic range (LDR) of 132 dB and fast transient responses below 100 μs. Overall, the multifunctionality of TPA-azaBODIPY-TPA provides a new guide for developing highly efficient multi-purpose conjugated OSCs for the next-generation optoelectronic devices.

RESULTS AND DISCUSSION Synthesis and properties of TPA-azaBODIPY-TPA

The starting material aza-BODIPY core 4 was synthesized over four-steps using a literature procedure.⁴⁰ The diaryl α,β-unsaturated ketone (chalcone) 1 was prepared by an aldol/dehydration reaction of the 4-bromobenzaldehyde and 4-methoxyacetophenone. The chalcone 1 product was isolated by filtration resulting in a quantitative yield (96%). The addition of nitromethane to chalcone 1 in an alkali ethanol solution gave 1,3-diaryl-4-nitrobutan-1-one 2, which was isolated as a crude amber oil. Condensation of 1,3-diaryl-4-nitrobutan-1-one 2, with ammonium acetate in refluxing butanol then yielded the azodipyrromethene 3, which was subsequently complexed with boron trifluoride to give aza-BODIPY 4 as a dark green solid in a modest yield over two steps (28%). The key step in the synthesis of TPA-azaBODIPY-TPA utilized a Suzuki cross coupling reaction between an electron-deficient (acceptor) aza-boron-dipyrromethene (aza-BODIPY) and an electron-rich (donor) TPA (Figure 1A). Aza-BODIPY 4 was then coupled with boronate ester 5 in a Suzuki reaction using Pd(PPh₃)₄ as catalyst to give TPA-azaBODIPY-TPA in moderate yield (55%). Compounds 2–4 were characterized via ¹H NMR spectroscopy (Figures S1–S3) and the obtained data matched that reported in the literature.⁴⁰ TPA was obtained using our previously published protocol.⁴¹ The chemical structure of TPA-azaBODIPY-TPA was characterized by ¹H NMR, ¹³C NMR, and FT-IR (Figures S4–S6), its density functional theory (DFT)-optimized structure as shown Figure 1B. Thermogravimetry analysis (TGA) shows a high decomposition temperature (T_d, 5% weight loss) of 412°C (Figure 1C). The differential scanning calorimetry (DSC) analysis indicates that TPA-azaBODIPY-TPA exhibits a crystalline isotropic melting transition at around 302.7°C due to the highly rigid conjugated backbone (Figure S7). This excellent thermal stability of TPA-azaBODIPY-TPA ensures that highly crystalline structures can be achieved through thermal annealing treatments often required to enable optimal performance.⁴² TPA-azaBODIPY-TPA exhibits an absorption peak at 712 nm (Figure 1D) attributed to the lower energy band in the visible region because of intramolecular charge transfer (ICT) transitions. The high energy bands at 284 and 350 nm are attributed to π–π* electronic transitions. Compared with aza-BODIPY, which peaks at 690 nm, a 22 nm redshift is observed for TPA-azaBODIPY-TPA, indicating an adequately low bandgap suitable for NIR absorption in a small molecule conjugated system, which is highly desirable for multi-purpose applications such as solar cells or photodetectors.

FIGURE 1. Synthesis and properties of TPA-azaBODIPY-TPA. (A) Detailed synthetic route for the precursors, aza-BODIPY and TPA-azaBODIPY-TPA. (B) Density functional theory (DFT)-optimized structure of TPA-azaBODIPY-TPA. (C) Thermogravimetric analysis (TGA) of TPA-azaBODIPY-TPA. Td stands for the decomposition temperature and corresponds to a weight loss of 5%. (D) The absorption spectra of the azaBODIPY and TPA-azaBODIPY-TPA solutions (5 μM) in acetonitrile (ACN). Both the spectra and the photograph highlight a small (22 nm) red shift after attaching TPA.

Versatility of TPA-azaBODIPY-TPA

The optical bandgap of TPA-azaBODIPY-TPA, extrapolated from the Tauc plot, was 1.49 eV (Figure 2A), in close accordance with the absorption edge (Figure 1D). Hence, the material can be used as an electron donor in NIR OPDs. The energy levels of TPA-azaBODIPY-TPA were measured to evaluate its compatibility with electron acceptors and interlayers. The highest occupied molecular orbital (HOMO) of TPA-azaBODIPY-TPA was deduced from the ultraviolet photoelectron spectroscopy (UPS) measurement and was found to be −5.1 eV (Figure 2B,C). Based on its HOMO value, TPA-azaBODIPY-TPA can be an ideal hole-transporting material (HTM) for PSC devices. To date, PEDOT:PSS is the most commonly used HTM in various organic electronic devices due to its high hole-transporting ability and conductivity but research efforts are being put into the development of novel more efficient materials.⁴³ The newly developed TPA-azaBODIPY-TPA's HOMO value matched the PEDOT:PSS work function of −5.0 eV very well, which makes TPA-azaBODIPY-TPA an ideal choice for low-cost, ultra-thin HTL in OPVs and PSCs.^44,45 Moreover, its HOMO value matches well with the work function of −4.8 eV of the frequently used indium tin oxide (ITO) electrodes.⁴⁴ Small energy-level offsets between materials are crucial for efficient charge transfer in both photodetectors and solar cells.^46–48

FIGURE 2. Determination of HOMO and LUMO levels of TPA-azaBODIPY-TPA. (A) Tauc plot of the ITO/TPA-azaBODIPY-TPA film depicting its optical bandgap (Eg). (B) Ultraviolet photoelectron spectroscopy (UPS) analysis of the TPA-azaBODIPY-TPA film. (C) Left: the secondary electron cut-off region of UPS spectrum from (B). Right: The valence band (VB) edge region of UPS spectrum from (B). EF is the Fermi energy. (D) Deduced highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels of TPA-azaBODIPY-TPA from a combination of the Tauc plot and UPS data. (E) Density functional theory (DFT)-computed HOMO and LUMO orbitals of TPA-azaBODIPY-TPA depicting their calculated electron densities along with the bandgap.

To further test the TPA-azaBODIPY-TPA suitability to act as either electron donor or HTL, we estimated its lowest unoccupied molecular orbital (LUMO) at −3.6 eV, using the Tauc plot and UPS data (Figure 2D). This value was further confirmed with a DFT-computed LUMO orbital that yielded a similar electron density of −3.48 eV (Figure 2E). In OPDs, LUMO level of an electron donor plays a significant role in enabling exciton transfer and separation at the interface with an electron acceptor. For this to happen, it is commonly thought that a minimum energy gap of 0.3 eV should exist between the LUMO levels of electron donor and acceptor.⁴⁹ Hence, with a LUMO level of −3.6 eV, TPA-azaBODIPY-TPA offers compatibility, as electron donor, with a large family of electron-accepting fullerene derivatives.^50,51 In parallel, a high LUMO level of TPA-azaBODIPY-TPA acting as a HTL can effectively block the electron leakage in PSCs,⁵² using materials with desirable deeper energy levels, such as MAPbI₃ reported here, featuring a conduction band minimum (CBM) of −3.9 eV.⁵³ As such, addressing intrinsically a large number of criteria for materials used in solar cells and photodetectors, TPA-azaBODIPY-TPA has the potential to become a versatile, multifunctional building block in optoelectronic devices.

TPA-azaBODIPY-TPA as HTL in PSCs

We first investigated this material's suitability in PSCs, where we used TPA-azaBODIPY-TPA as HTL to replace conventional PEDOT:PSS. To fabricate PSCs (Figure 3A), we used well-known and extensively studied halogenated methylammonium (MA) lead iodide—perovskite (MAPbI₃) as the light absorbing active material. MAPbI₃ is known for its excellent solution-processibility at low temperatures, high absorption in the visible range, and high photogenerated charge carrier mobility.⁵⁴ Fullerene C₆₀, used as electron-transporting material, completes the cell heterojunction. Bathocuproine (BCP) is used as a buffer layer, improving the ohmic contact with the top electrode silver (Ag) and reducing interfacial charge recombination.¹⁰ TPA-azaBODIPY-TPA is used as HTL, with its HOMO level being desirably situated in between the valence band maximum (VBM) of MAPbI₃, and the work function of ITO (Figure 3B). Additionally, since TPA-azaBODIPY-TPA features a high LUMO level, it can efficiently block the electron leakage in PSCs.⁵² Apart from optoelectrical functions, TPA-azaBODIPY-TPA also provides a smoother surface when compared to pure ITO (Figure 3C), with the root-mean-square (RMS) roughness reduced from 2.2 to 1.6 nm. This is ideal for depositing smooth and compact perovskite films. Overall, fabricated devices featured an inverted p–i–n architecture of ITO/TPA-azaBODIPY-TPA/MAPbI₃/C₆₀/BCP/Ag.

FIGURE 3. Performance of perovskite solar cells based on TPA-azaBODIPY-TPA. (A) Cross-sectional device structure of fabricated photovoltaic cells, ITO/TPA-azaBODIPY-TPA/MAPbI3/C60/BCP/Ag, highlighting the thicknesses of individual layers. A cross-sectional scanning electron microscope (SEM) image of the device architecture is also shown. (B) The energy level diagram of device components illustrating the photovoltaic effect: (1) photon absorption in MAPbI3 and exciton generation, (2) exciton dissociation (separation into free charges at room temperature) and diffusion within the active layer, (3) electron and hole charge carrier transfer at the interfaces, (4) charge transfer toward electrodes assisted by the presence of TPA-azaBODIPY-TPA and BCP. (C) Atomic force microscopy (AFM) image of a pure ITO (labeled as “A”), and a TPA-azaBODIPY-TPA film on ITO (denoted as “B”). (D) Comparison of the current density–voltage (J–V) characteristics of the fabricated solar cells with either TPA-azaBODIPY-TPA or PEDOT:PSS used as hole transport layer (HTL). Both forward and reverse scans are showed. (E) Evolution of the short-circuit current density (JSC) and the open-circuit voltage (VOC) of perovskite solar cells with the increasing thickness of TPA-azaBODIPY-TPA. Best thickness region is shown. (F) Evolution of the device fill factor (FF) and power conversion efficiency (PCE) to the increasing thickness of TPA-azaBODIPY-TPA. Best thickness is highlighted. (G) Histogram showing the PCE distribution for n = 20 optimized devices from the regions depicted in (E) and (F). (H) Average (n = 20) JSC, VOC, and FF of optimized perovskite solar cells. Error bars show the standard deviation. (I) External quantum efficiency (EQE) of an optimized device. Integrated JSC based on EQE is plotted for comparison. (J) Steady JSC and PCE evolution of the TPA-azaBODIPY-TPA-based photovoltaic cells held under 200 s of continuous light illumination (100 mW cm−2 AM 1.5G) at a forward bias of 0.84 V. The photograph shows TPA-azaBODIPY-TPA-based perovskite solar cells.

A series of studies on organic hole transport materials have indicated that the film morphology and energy level play a crucial role in determining the performance of PSCs.^44,52,55 We first directly compared the photovoltaic performance of devices with either a novel TPA-azaBODIPY-TPA molecule or an optimized PEDOT:PSS layer that offers a similar HOMO level of −5.1 eV. For the devices based on PEDOT:PSS with optimal thickness (Figure S8), a below average photovoltaic performance was obtained, with an average PCE of 10.6%, an open-circuit voltage (V_OC) of 1.00 V, a short-circuit current density (J_SC) of 18.5 mA cm⁻², and a fill factor (FF) of 57.2% (Table 1). Moreover, PEDOT:PSS-based devices showed a large difference (about 1.7%) in PCEs under different scan directions, suggesting a presence of J–V hysteresis effect (Figure 3D).⁵⁶ We attribute an overall poor performance of PEDOT:PSS-based devices to the low thermal stability of the HTL, as the relatively high temperature of the substrate during evaporation of the MAPbI₃ perovskite may lead to partial destruction of the underlying PEDOT:PSS layer.⁵⁷ In contrast to control devices, the TPA-azaBODIPY-TPA-based PSCs demonstrated a remarkable photovoltaic performance, reaching an average PCE of 17%, which is an improvement of 60%. The devices featured an average V_OC of 1.00 V, a J_SC of 22.0 mA cm⁻², a FF of 77.5% (best device: PCE:17.4%, V_OC: 1.01 V, J_SC: 21.7 mA cm⁻², and FF: 79.4%) with also a negligeable hysteresis effect. Furthermore, the results obtained here are similar to commonly reported highly-efficient MAPbI₃-based PSCs.⁵ While the V_OC remained unchanged, strong improvements in J_SC and FF are remarkable. A close look at the external quantum efficiency (EQE) spectra reveals a slight redshifted response upon introduction of TPA-azaBODIPY-TPA as HTL (Figure S9). The consistent high EQE values across the entire photocurrent spectrum indicate TPA-azaBODIPY-TPA additional role as an optical confinement layer (Figure S10). A detailed optical simulation is out of scope for this work but remains an important prospect for further investigations.

TABLE 1 Comparison of the photovoltaic parameters of perovskite solar cells (PSCs) based on PEDOT:PSS or TPA-azaBODIPY-TPA

HTL		VOC (V)	JSC (mA cm−2)	FF (%)	PCE (%)
PEDOT:PSS	Champion (RS)	1.00	18.9	59.9	11.3
	Champion (FS)	1.04	19.1	48.1	9.6
	Mean ± SDa	1.00 ± 0.01	18.5 ± 0.7	57.2 ± 2.8	10.6 ± 0.6
TPA-azaBODIPY-TPA	Champion (RS)	1.01	21.7	79.4	17.4
	Champion (FS)	1.01	21.5	80.1	17.3
	Mean ± SD	1.00 ± 0.01	22.0 ± 0.3	77.5 ± 1.1	17.0 ± 0.2

Abbreviations: FS, forward scan; RS, reverse scan.

^aSD, standard deviation calculated from RS J–V curves of at least 16 separate devices.

Considering its optical properties and their potential influence on the photovoltaic response, we also evaluated the role of the TPA-azaBODIPY-TPA thickness in the overall performance of PSCs. Various solution concentrations and spin-coating speeds were used to obtain TPA-azaBODIPY-TPA films with different thicknesses. As the TPA-azaBODIPY-TPA thickness increases from 2 nm to 6 nm, the V_OC and J_SC of the devices exhibit significant improvements (Figure 3E), which is probably due to the increase in the surface coverage of TPA-azaBODIPY-TPA film, thus leading to reduction of parasitic current loss and increase of shunt resistance. However, the FF decreases sharply with further increase of TPA-azaBODIPY-TPA thickness (Figure 3F), which can be ascribed to the enhancement of the intrinsic resistance or potential new routes of charge recombination arising from the TPA-azaBODIPY-TPA layer, thus decreasing the performance of devices. Overall, the optimized PCE was obtained at 6 nm for TPA-azaBODIPY-TPA. Furthermore, a series of 20 devices fabricated using the optimized thickness showed great uniformity in PCE (Figure 3G) and other photovoltaic parameters (Figure 3H). The device's EQE featured a large plateau of 80%–82% in the range from 430 to 750 nm (Figure 3I), with the J_SC integrated from the incident photon to current conversion efficiency (IPCE) spectrum (20.4 mA cm⁻²) being in close agreement with the values measured (Figure 3H). In addition, the stabilized power output efficiency under maximum power point (MPP) can also reach 17% (Figure 3J). Lastly, we examined the operational stability of the PEDOT:PSS and TPA-azaBODIPY-TPA-based PSCs. The operational stability tests were carried out at MPP in a N₂-filled glovebox (O₂ <10 ppm, H₂O <1.0 ppm) under one sun illumination. Consequently, 90.2% of the initial PCE was maintained after 1200 min of operation at MPP for the TPA-azaBODIPY-TPA-based PSC, which is remarkably enhanced when compared to the PEDOT:PSS-based PSC (Figure S11).

TPA-azaBODIPY-TPA as electron donor in NIR OPDs

As discussed above, with a low optical bandgap of 1.49 eV (Figure 2A), TPA-azaBODIPY-TPA also has a suitable LUMO that makes it a good candidate as an electron donor in NIR light absorber in OPDs. To test its photodetection response, we designed and fabricated standard OPD devices based on well-known and the most commonly used materials in the field of organic electronics (Figure 4A). The [6,6]-phenyl-C71-butyric acid methyl ester (PC₇₁BM) was chosen as electron acceptor. The literature values for the PC₇₁BM LUMO level ranges from −3.7 to at −4.2 eV. This should result in a suitable energy offset to the corresponding LUMO level of TPA-azaBODIPY-TPA (Figure 4B). Similarly to solar cells, for the exciton dissociation and charge separation to take place, an optimal offset has to be in the range of at least 0.3–0.4 eV.⁴⁹ A bulk heterojunction of TPA-azaBODIPY-TPA and PC₇₁BM blend thin-film forms the photodetector active layer for absorption and detection of photons. ITO acts as the bottom electrode whereas PEDOT:PSS was chosen as HTL to allow an efficient coupling with the HOMO level of TPA-azaBODIPY-TPA, which in turn promotes an efficient transport of holes toward the external circuit. Similarly, Barium (Ba) and Silver (Ag) were chosen to match the LUMO level of PC₇₁BM and improve electron transfer. The above electrodes and interlayers have been previously successfully used by our group for fabricating OPDs,^58,59 and OLEDs.⁶⁰

FIGURE 4. Optimization of organic photodetectors (OPDs) based on TPA-azaBODIPY-TPA. (A) Cross-sectional device structure of fabricated OPDs, ITO/PEDOT:PSS/TPA-azaBODIPY-TPA:PC71BM/Ba/Ag, highlighting the thicknesses of individual layers. (B) Energy level diagram of the TPA-azaBODIPY-TPA-based OPD highlighting the detection mode: (1) near-infrared light is absorbed by TPA-azaBODIPY-TPA, and an exciton (electron–hole pair) is created, (2) exciton diffusion toward TPA-azaBODIPY-TPA:PC71BM interface, (3) charge transfer and charge dissociation of the exciton, (4) free charge collection at electrodes assisted by a reverse bias shown in (5). (C) Photographic mosaic showing OPDs at various donor:acceptor ratios and thicknesses of the active layer. Medium and thick devices are shown for the 1:3 ratio. (D) Evolution of the J–V characteristics of the OPDs fabricated with a donor:acceptor ratio of 1:1, 1:2, and 1:3. Light intensity (730 nm) was set to 10 mW cm−2. (E) Average (n = 6) responsivity and dark current (JD) at −1 V measured as a function of the donor:acceptor ratio. The best ratio is highlighted. (F) Example responsivity spectra for three distinct OPDs operating at −1 V. A donor:acceptor ratio 1:3 yields the best performance. (G) Measured EQE spectra of the devices. The measurements are taken at −1 V. (H) Atomic force microscopy (AFM) images of the active layer spin-coated with 1:1 and 1:3 donor:acceptor ratios. (I) Evolution of J–V characteristics of the OPDs to the increasing active layer thickness. All devices have an optimized 1:3 donor:acceptor ratio and are illuminated by a 730 nm light source set at 10 mW cm−2. (J) Deduced average (n = 6) responsivity and JD at −1 V for TPA-azaBODIPY-TPA-based OPDs with thin (88 nm), medium (114 nm) and thick (148 nm) active layer. (K) Average (n = 12) on–off ratio for an optimized photodetector operating at −0.1 V. Irradiance (730 nm) was set to 10 mW cm−2.

In contrast to PSCs where TPA-azaBODIPY-TPA was used as a standalone, ultra-thin HTL, the material function as electron donor, further blended with PC₇₁BM, requires additional optimization steps for the active layer to ensure sufficient absorption, balanced charge mobilities, and optimal optoelectronic performance.⁶¹ This is usually achieved by changing the electron donor:acceptor ratio in solution as well their concentration (Figure 4C, details in Table 2). Hence, we first studied the TPA-azaBODIPY-TPA:PC₇₁BM ratios (1:1, 1:2, and 1:3) to avoid imbalanced charge carrier mobilities in the bulk heterojunction, that have been previously found to be detrimental for device efficiency.⁶² From the J–V characteristics in dark and under illumination (730 nm, 10 mW cm⁻²), an average device with the 1:3 ratio yielded simultaneously the best photocurrent and the lowest dark current at −1 V (Figure 4D). The corresponding increase in average responsivity (n = 6 devices) is significant (38, 62, and 67.5 mA W⁻¹, for 1:1, 1:2, and 1:3 ratio, respectively), with dark currents dropping by 44.3% for the best ratio (Figure 4E). An even further increase in electron acceptor content (1:4 ratio) yielded a reduced responsivity of 42.5 mA W⁻¹, which we attribute to imbalanced charge carrier mobilities⁶² (Figure S12). These findings from J–V curves are further confirmed with the spectral measurements of the photodetector responsivity showing a maximum responsivity of 67.7 mA W⁻¹ at 690 nm, with the 10% edge detection at 890 nm (Figure 4F). The EQE measurements further confirm the suitability of the 1:3 ratio for NIR detection, with the maximum of 12.2% being measured at 680 nm, and the 10% edge detection recorded at 875 nm (Figure 4G). The low EQE response indicates a non ideal energy offset with PC₇₁BM, therefore pairing it with suitable acceptors would make a better prospect. Apart from balanced mobilities, atomic force microscopy (AFM) images of the active layer spin-coated with 1:1 and 1:3 electron donor:acceptor ratios also demonstrated the superiority of the 1:3 ratio (RMS of 0.48 and 0.38 nm, for 1:1 and 1:3 ratio, respectively) (Figure 4H), suggesting that the active layer morphology also plays an important role in enabling optimal performance.⁶³

TABLE 2 Comparison of photodetection parameters of TPA-azaBODIPY-TPA-based near-infrared (NIR) organic photodetectors (OPDs)

D:A ratio	Thickness	Rmaxa (mA W−1)	Jdb (μA cm−2)	EQEmaxc (%)
1:1	49 nm	38 ± 2d	52.1 ± 10	3.6
1:2	64 nm	62 ± 5	39.5 ± 18	6.6
1:3	Thin (88 nm)	67.5 ± 3.5	29 ± 12	12.2
	Medium (114 nm)	68 ± 3	17 ± 12	11.9
	Thick (148 nm)	23 ± 10	8 ± 10	N/Ae

^aResponsivity at −1 V.

^bDark current at −1 V.

^cEQE at −1 V.

^dStandard deviation (SD) measured for at least six devices.

^eN/A, parameter not measured.

Having established the best TPA-azaBODIPY-TPA:PC₇₁BM ratio (1:3), we next evaluated the role of the active layer thickness on device performance. Although it has been well reported that a thick active layer can significantly reduce the device dark current, it can also lower the device photocurrent and responsivity, owning to an increase in recombination of photogenerated carriers in thicker layers.^64,65 To find the trade-off between dark current and responsivity, we fabricated OPDs with a thin (88 nm), medium (114 nm) and thick (148 nm) active layers. From the J–V characteristics, the device dark current at −1 V drops by 22% and 68% with an increasing active layer thickness: thin to medium and thin to thick, respectively (Figure 4I). Furthermore, a small increase of 11% is responsivity at −1 V is measured when switching from a thin to a medium device. The responsivity significantly drops by 45% for the thick device, suggesting that a medium thickness (114 nm) improves dark current while preserving high photocurrent. Our statistical test (n = 6 devices) further confirmed that the medium thickness is optimal (Figure 4J). The best configuration (ratio 1:3, medium thickness) yielded an average (n = 12 devices) on/off ratio, that is, the ratio between the OPD photocurrent and the dark current for a given voltage bias and illumination, of 1000 (Figure 4K), illustrating the TPA-azaBODIPY-TPA strong capability to sense light.

Our optimized configuration showed a robust photoresponse to an increasing light intensity and a decreasing voltage bias (Figure 5A). When biased at −1 V, light intensity as low as 100 μW cm⁻² can be detected by the OPD featuring an EQE of 12% (Figure 5B). Impressively, in the short-circuit condition (0 V), the TPA-azaBODIPY-TPA photodetector is able to sense irradiance levels as low as 100 pW cm⁻². The linear response is observed for the accessible irradiance levels of the calibrated NIR LED, that is, from 1 nW cm⁻² to 4 mW cm⁻², featuring a LDR of 132 dB (Figure 5C), similar to previous reports.⁶⁶ To further characterize the optoelectronic performance of optimized TPA-azaBODIPY-TPA-based OPDs, we tested their transient photocurrent response upon NIR illumination. The OPD showed a robust photocurrent response to a 500 μs light pulse at 730 nm, exhibiting a rise time t_rise (time for which device response rises from 10 to 90%) and a fall time t_fall (time for which photodetector response decreases from 90% to 10%) of 66 and 101 μs, respectively. These values are in good agreement with similar NIR OPDs.^67,68 The absence of negative bias significantly slows down the extraction process of photogenerated charge carriers, hence, biasing the OPD would further improve its rise and fall times. To further explore the dynamic behavior of the photodetector, we measured its photocurrent response for light pulses at 5, 10, and 20 kHz. The signal amplitude drops significantly to 85% at 5 kHz, 63% at 10 kHz, and 40% at 20 kHz (Figure 5E). The OPD cut-off frequency f_−3dB, that is, the frequency at which the photocurrent J_ph is reduced to 0.71 of its initial steady-state value J₀, is 7.8 kHz at 0 V (Figure 5F). This value is adequate for most applications in photodetection, human-machine interfaces, and smart surfaces,^69,70 ultimately demonstrating that the TPA-azaBODIPY-TPA is an efficient electron donor material in NIR OPDs.

FIGURE 5. Optoelectrical and temporal performance of optimized organic photodetectors (OPDs) based on TPA-azaBODIPY-TPA. (A) J–V characteristics of an optimized TPA-azaBODIPY-TPA-based OPD under light intensity levels from 0 (“dark”) to 125 mW cm−2 achieved with a monochromatic 730 nm light source. (B) Measured EQE spectra under 0, −0.5, and −1 V. (C) Photocurrent density of the optimized OPD as a function of incident power. A monochromatic light source of 730 nm (near-infrared) is used, and the device operates in the short-circuit condition (0 V). (D) Normalized photocurrent response of the TPA-azaBODIPY-TPA-based device to a 500 μs light pulse at a frequency of 1 kHz. The incident power density is equal to 10 mW cm−2. (E) Transient photocurrent response at pulse frequencies of 5, 10 and 20 kHz. The photodetector is not biased. (F) Measured cut-off frequency (f−3dB) of the optimized device at 0 V. The −3 dB threshold corresponds to the frequency at which the photocurrent signal is attenuated by 1.414. The photograph shows a TPA-azaBODIPY-TPA-based ODP illuminated by a 730 nm light.

CONCLUSION

In this work, we have designed, synthesized, and characterized a novel conjugated small molecule TPA-azaBODIPY-TPA via a D–A–D molecular engineering approach. The molecule offers desirable optoelectronic and photophysical properties for multi-purpose utility in light harvesting and photodetection technologies. The material versatility originates from carefully designed HOMO energy level and lower bandgap arising from donor–acceptor combination which is perfectly suitable for both PSC and OPD devices. Furthermore, the deployment of TPA-azaBODIPY-TPA as a HTL in solar cells or as electron donor in photodetectors is greatly facilitated by its superior solubility and processability. Interestingly, in PSCs, the material outperforms a conventional control HTL PEDOT:PSS improving the overall photovoltaic performance. In NIR OPDs, TPA-azaBODIPY-TPA demonstrates high compatibility with a commonly used electron acceptor PC₇₁BM to offer a superior sensing performance in line with the state-of-the-art devices.

Future developments of this versatile material involve its long-term stability testing in PSCs and optical simulations that are crucial for enabling commercialization.¹⁰ The material's compatibility goes beyond the demonstrated hybrid devices as one could develop TPA-azaBODIPY-TPA-based fully organic solar cells. The improvement pathways for efficient OPDs involve the reduction of the device dark current beyond the active layer optimization. This can be achieved, for example, by the use of effective electron and hole blocking interlayers.^71,72 Finally, the multifunctional material could also find applications in lightning technology, with the potential of replacing PEDOT:PSS in OLEDs.³

In summary, our research demonstrates a new multifunctional material capable of acting as either an efficient HTL in PSCs or as a robust electron donor in NIR OPDs. The established versatility of TPA-azaBODIPY-TPA, combined with its high thermal stability as well as excellent solubility and processability, provides a new guide for developing highly efficient multi-purpose materials for the next-generation optoelectronic devices.

EXPERIMENTAL SECTION Materials Synthesis

Tetrakis(triphenylphosphine)palladium Pd(PPh₃)₄ was obtained from Strem. Sodium sulfate (Na₂SO₄) was purchased from Ajax finechem. Chloroform and ethyl acetate were obtained from Fisher Scientific. Potassium carbonate (K₂CO₃), toluene, sodium sulfate (Na₂SO₄), chloroform, acetone, 4-bromobenzaldehyde, 4-methoxyacetophenone, potassium hydroxide, anhydrous dichloromethane, ethanol, nitromethane, ethyl acetate ammonium acetate, anhydrous n-butanol, diisopropylethylamine (DIPEA), and boron trifluoride diethyl etherate (BF₃OEt₂) were purchased from the chemical supplier Sigma-Aldrich and the chemicals were used without further any purification.

Perovskite solar cells

All the chemicals and materials were purchased and used as received unless otherwise noted, including patterned ITO-coated glass substrates (thickness: ~150 nm, sheet resistance: ≤10 Ω sq⁻¹, Shenzhen Huayu Union Technology Co., Ltd), poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) (Clevious AI 4083), PbI₂ (TCI), MAI (Lumtec), C₆₀ (99.9%, Puyang Yongxin Fullerene Technology Co., Ltd), BCP (99.9%, Ltd and Xi'an Polymer Light Technology Corp.), Chlorobenzene (99.8%, anhydrous, Sigma-Aldrich).

Organic photodetectors

ITO substrates (10 Ω per square) were purchased from Xinyan Technology Limited. Poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) aqueous dispersion (1.3%–1.7%) was purchased from Heraeus and used without any further purification. PC₇₁BM (purity 99%) was purchased from CNano Inc. Barium (purity 99.9%) and silver (purity 99.9%) were purchased from Sigma-Aldrich and used as received.

Synthesis Aza-BODPIY precursors Synthesis of (E)-3-(4-bromophenyl)-1-(4-methoxyphenyl)prop-2-en-1-one (1)

A solution containing 4-bromobenzaldehyde (1.49 g 8.05 mmol), 4-methoxyacetophenone (1.29 g, 8.60 mmol), KOH (62 mg, 1.11 mmol), and ethanol (15 ml) was stirred for 22 h at room temperature. The product was collected by vacuum filtration and washed with cold ethanol resulting in a dried white powder of product 1 (see Figure 1A). Yield: 2.45 g (96.1%). ¹H NMR (600 MHz, CDCl₃) δ: 8.06–8.00 (m, 2H), 7.73 (d, J = 15.6 Hz, 1H), 7.55 (d, J = 8.2 Hz, 2H), 7.53–7.48 (m, 2H), 7.01–6.95 (m, 2H), 3.90 (s, 3H). HRMS (ESI): Found m/z 317.0144, Calc. m/z for [M + H]⁺ = 317.0172.

Synthesis of (Z)-3-(4-bromophenyl)-1-(4-methoxyphenyl)-4-nitrobut-2-en-1-one (2)

A solution of 3-(4-bromophenyl)-1-(4-methoxyphenyl)prop-2-en-1-one (1) (2.54 g, 8.01 mmol), KOH (117 mg, 2.09 mmol), and nitromethane (8.5 ml, 159 mmol) in ethanol (5 ml) was stirred at 60°C for 19 h. The reaction mixture was then cooled to room temperature and the solvent was removed by rotary evaporation under reduced pressure. The resulting residue was dissolved in EtOAc (100 ml) and washed with water (3 × 100 ml) before the combined aqueous layers were extracted with EtOAc (3 × 50 ml). The combined organic layers were washed with brine (100 ml) and dried with Na₂SO₄. The solvent was removed by in vacuo. Product 2 (see Figure 1A) was obtained as an amber oil and used without further purification. ¹H NMR (600 MHz, CDCl₃) δ: 7.95–7.88 (m, 2H), 7.52–7.44 (m, 2H), 7.22–7.16 (m, 2H), 6.98–6.91 (m, 2H), 4.84 (dd, J = 12.6, 6.4 Hz, 1H), 4.68 (dd, J = 12.6, 8.3 Hz, 1H), 4.21 (p, J = 6.9 Hz, 1H), 3.89 (s, 3H), 3.44–3.33 (m, 2H). HRMS(ESI): Found m/z = 378.0327, Calc. m/z for [M + H]⁺ = 378.0335.

Synthesis of 2-(4-((λ¹-oxidaneyl)-λ⁵-methyl)phenyl)-4-(4-bromophenyl)-N-(4-(4-bromophenyl)-2-(4-metho-xyphenyl)-3H-pyrrol-5-yl)-3H-pyrrol-5-amine (3)

3-(4-Bromophenyl)-1-(4-methoxyphenyl)-4-nitrobutan-1-one (2) (3.1 g 8.20 mmol) and ammonium acetate (22.0 g, 285 mmol) were dissolved in anhydrous n-butanol (15 ml). The reaction mixture was refluxed for 23 h. The resulting dark blue reaction was cooled to room temperature, concentrated in vacuo, and filtered. The crude blue-black solid was washed with cold ethanol and air dried resulting in product 3 (see Figure 1A), which was taken straight to the next step without characterization or further purification.

Synthesis of aza-BODIPY

An oven-dried three-neck flask containing the crude azadipyrromethene (3) (1.44 g, 0.288 mmol) was evacuated and flushed with argon. Anhydrous DCM (35 ml) and diisopropylethylamine (DIPEA) (4.2 ml 34 mmol) were added to the flask and the reaction mixture was stirred for 15 min before boron trifluoride diethyl etherate (BF₃. Et₂O) (4.2 ml, 24.1 mmol) was added. The reaction was stirred for 24 h at room temperature. The mixture was washed with water (3 × 200 ml) before the combined aqueous layers were extracted with DCM (3 × 100 ml). The combined organic layers were washed with brine and dried on Na₂SO₄. The solvent was removed in vacuo to isolate the aza-BODIPY (3-(4-((λ¹-oxidaneyl)-λ⁵-methyl)phenyl)-1,9-bis(4-bromophenyl)-5,5-difluoro-7-(4-methoxyphenyl)-5H-5λ⁴,6λ⁴-dipyrrolo[1,2,-c:2′,1′-f[1,3,5,2]triazaborinine) (4, see Figure 1A) as a dark green solid (1.62 g, 28.3% from 3). ¹H NMR (600 MHz, CDCl₃) δ: 8.12–8.04 (m, 4H), 7.94–7.87 (m, 4H), 7.63–7.57 (m, 4H), 7.06–6.98 (m, 5H), 3.89 (s, 6H). ¹³C NMR (151 MHz, CDCl₃) δ 162.28, 131.97, 131.85, 131.46, 130.70, 124.04, 118.89, 114.49, 77.37, 77.16, 76.95, 55.62. ¹¹B NMR (193 MHz, CDCl3) δ: 0.27 (t, J = 29.1 Hz).¹⁹F NMR (565 MHz, CDCl₃) δ −131.99 (d, J = 4.5 Hz), −132.05, −132.10, −132.16. HRMS(ESI): Found m/z 713.0279, calc. m/z for [M]⁺ = 713.0296.

Synthesis of TPA-azaBODIPY-TPA

3-(4-((λ¹-oxidaneyl)-λ⁵-methyl)phenyl)-1,9-bis(4-bromophenyl)-5,5-difluoro-7-(4-methoxyphe-nyl)-5H-5λ⁴,6λ⁴-dipyrrolo[1,2,-c:2′,1′-f[1,3,5,2]triazaborinine (4) (293 mg, 0.41 mmol) and 4-methoxy-N-(4-methoxyphenyl)-N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)ani-line (TPA) (5) (480 mg, 1.11 mmol), and 2 M aqueous K₂CO₃ solution (17 ml) were suspended in degassed toluene (19 ml) under argon protection for 1 h at room temperature. Tetrakis(triphenylphosphine)palladium (22.65 mg, 0.020 mmol) was added to the reaction mixture and stirred at 110°C under argon for 48 h. The reaction mixture was cooled and extracted using chloroform and water. The organic phase was dried over Na₂SO₄ and the solvent was removed under reduced pressure. The crude product was then purified by recrystallization from acetone to obtain black solid TPA-azaBOIDIPY-TPA (4′,4‴′-(3-(4-((λ1-oxidaneyl)-λ5-methyl)phenyl)-5,5-difluoro-7-(4-methoxyphenyl)-5H-5λ4,6λ4-dipyrro-lo[1,2-c:2′,1′-f][1,3,5,2]triazaborinine-1,9-diyl)bis(N,N-bis(4-methoxyphenyl)-[1,1′-biphenyl]-4-amine)) (6, see Figure 1A). Yield: 365 mg (55%). ¹H NMR (600 MHz, CDCl_3, δ): 8.01–8.08 (m, 10H), 7.57–7.58 (d, J = 6 Hz, 5H), 7.41–7.42 (d, J = 6 Hz, 5H), 7.03–7.05 (d, J = 12 Hz, 10H), 6.93–6.96 (m, 12H), 6.78–6.80 (d, J = 12 Hz, 10H), 3.82 (s, 6H), 3.74 (s, 12H). ¹³C NMR (151 MHz, CDCl₃, δ): 161.82, 156.07, 148.59, 145.46, 142.84, 141.65, 140.67, 131.97, 131.57, 130.84, 129.64, 127.57, 126.63, 126.51, 124.83, 124.33, 120.35, 114.77, 114.21, 55.50, 55.43. IR (ATR): ν = 2923, 2852, 1592, 1503 1485, 1438, 1240, 1181, 1097, 1036, 814, 722, and 575 cm⁻¹.

Device fabrication Perovskite solar cells

The ITO-coated glass substrates were ultrasonically cleaned with detergent, acetone, ethanol and deionized water for about 20 min in sequence and then ultraviolet-ozone treated for 15 min. Next, the TPA-azaBODIPY-TPA film was deposited by spin-coating TPA-azaBODIPY-TPA in chlorobenzene solution on the cleaned ITO substrates and then dried at 100°C for 10 min. Four configurations were used to achieve distinct thicknesses: 1 mg ml⁻¹ at 4000 rpm (2 nm), 2 mg ml⁻¹ at 4000 rpm (6 nm), 2 mg ml⁻¹ at 2000 rpm (9 nm), and 4 mg ml⁻¹ at 2000 rpm (28 nm). The PEDOT:PSS film was deposited by spin-coating PEDOT:PSS aqueous (filtered through a 0.45 μm first) on the cleaned ITO substrate, followed by annealing at 145°C for 15 min. Various spin-coating speeds (6000, 4000, 2000, and 1000 rpm, all at 60 s) were employed to obtain the PEDOT:PSS layer with different thickness (20, 30, 40, and 60 nm, respectively). The MAPbI₃ perovskite film was deposited using a co-evaporation deposition method. The substrates were connected to a rotating (10 rpm) base plate in a chamber pumped down to a base pressure of 6 × 10⁻⁶ Torr. The perovskite was deposited by co-evaporating PbI₂ powder and MAI powder in effusion sources. The PbI₂ source was fixed at 348°C and MAI source at 116°C, respectively, the total deposition time was around 120 min. After the co-evaporation, the MAPbI₃ films were heated on a 100°C for 30 min. Finally, 50 nm C₆₀, 6 nm BCP, and 100 nm Ag were thermal evaporated (under a vacuum of <5.0 × 10⁻⁴ Pa) in sequence on the perovskite films though a shadow mask with an active area of 0.060 cm².

Organic photodetectors

The ITO-coated glass substrates were first sequentially cleaned in an ultrasonic bath with Alconox detergent (Sigma-Aldrich), acetone, ethanol and isopropanol (15 min each), then dried using nitrogen gas gun and shortly annealed in air to remove any trace of solvent. PEDOT:PSS was spin-coated at 5000 rpm for 60 s on the ITO substrates to form a 30-nm layer, samples were then dried at 110°C for 15 min. TPA-azaBODIPY-TPA was mixed with PC₇₁BM at 1:1, 1:2, and 1:3 weight ratio with overall concentrations of 30, 49, and 54 mg ml⁻¹ in 1,2-dichlorobenzene (99.9%, anhydrous, Sigma-Aldrich). The solutions were stirred overnight at 80°C in sealed vials. The active layers were deposited by spin-coating in a N₂-filled glovebox environment (O₂ and H₂O <0.1 ppm) at 1000 rpm for 60 s to form uniform thin films of 49, 64 and 88 nm. All samples were then annealed at 120°C for 15 min and moved into an evaporation chamber. Finally, 10 nm of barium and 90 nm silver were evaporated under high vacuum (10⁻⁶ mbar) at slow rates of 0.02 and 0.06 nm s⁻¹, respectively. The device active area was 2.25 mm². The initial device with a 1:3 ratio (88 nm, “thin”) was further optimized by increasing the volume concentration of TPA-azaBODIPY-TPA donor and PC₇₁BM acceptor to 62 and 92 mg ml⁻¹, yielding an active layer thickness of 114 (“medium”) and 148 (“thick”) nm, respectively. The medium and thick devices were fabricated by keeping the rest of the parameters unchanged.

Characterization TPA-azaBODIPY-TPA

¹H and ¹³C NMR spectra were performed on a 600 MHz NMR spectrometer (Bruker, UK), using deuterated chloroform (CDCl₃ and tetramethyl silane [TMS]) as the solvent and internal standard, respectively. Fourier Transform Infrared (FT-IR) spectra was acquired on an FT-IR spectrometer (Bruker, Germany) at the wavelength range of 400–4000 cm⁻¹. The absorption spectra were recorded using a Cary 60 UV–Vis Spectrophotometer (Agilent, USA). Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were performed using a simultaneous thermal analyzer (STA 449F3 Jupiter, Netzsch, Germany). High-Resolution Mass Spectrum (HRMS) was acquired on an LTQ Orbitrap Elite mass spectrometer (Thermo Fisher Scientific, Germany). DFT calculations were carried out in Gaussian 09, using the CAM-B3LY P exchange-correlation functional and the Los Alamos National Laboratory (Lanl2dz) basis set.

Perovskite solar cells

The current density-voltage (J–V) curves of the devices were recorded under 100 mW cm⁻² AM 1.5 G simulated illumination by utilizing a Keithley 2400 source meter and a ScienceTech UHE-NSC AAA solar simulator in a N₂-filled glovebox (<1.0 ppm O₂ and H₂O). The IPCE spectra were performed on a Zolix Solar Cell Scan100 measurement system (Zolix Instruments Co., Ltd.). The AFM analysis was conducted on BioScope Resolve from Bruker. The cross-sectional SEM image was obtained by using JEOL JSM-7600F. The UPS data were taken on an AXIS Supra instrument from Kratos Analytical Ltd. The UV–Vis absorption spectra were measured utilizing a UV–Vis NIR spectrophotometer (Cary 5000, Agilent Technologies).

The thickness of the TPA-azaBODIPY-TPA layer as HTL was measured for the optimal configuration (5.6 nm) using AFM system (BioScope Resolve from Bruker), and estimated for the other configurations using relations from the literature.^73,74

Organic photodetectors

Current density–voltage (J–V) characteristics and linearity (LDR) measurements were recorded using a high-power, near-infrared (730 nm) LED from Intelligent LED Solutions (ILH-OW04-FRED-SC211-WIR200). LED was calibrated with a silicon PIN photodiode (Osram BPX61). A Keithley 2604B dual-channel source measure unit was used to power the LED and acquire data (current, voltage) from OPDs. A Newport QANTAX-300 quantum efficiency system was used to measure spectral EQE and responsivity at various voltages. A Keysight InfiniiVision DSOX2004a oscilloscope and an Agilent 33500B function generator were used to perform photocurrent transient measurements. The AFM analysis was performed using Bruker Icon Dimension, Multimodal AFM. Thickness measurements were conducted on Bruker Dektak XT-A Stylus Profiler. Optoelectrical characterization of OPDs was carried out in a double-shielded box (Faraday cage). Measurements were performed either in air with the device being exposed to N₂ flow or in a N₂-filled glovebox. LabVIEW scripts were written to acquire and interpret different input signals.

GraphPad Prism and Adobe Illustrator were used to prepare all figures.

AUTHOR CONTRIBUTIONS

Prashant Sonar conceptualized an idea. Mahnaz D. Gholami, Isabel Mathers, Astrid C. R. Larin, and Thomas Flanagan synthesized precursors materials including starting materials aza-BODIPY and final compound TPA-azaBODIPY-TPA under the supervision of Prashant Sonar, Emad L. Izake, and Kathryn E. Fairfull-Smith; Aza-BODIPY was characterized by Astrid C. R. Larin whereas TPA-azaBODIPY-TPA was completely characterized by Mahnaz D. Gholami; Ajay K. Pandey provided input for the OPD device engineering and OPD results and discussion part; Enkhtur Erdenebileg and Haixia Rao fabricated and optimized the device under the supervision of Annalisa Bruno and Lam Yeng Ming; Haixia Rao characterized the PSCs and the thin films under the supervision of Lam Yeng Ming; Gurudutt Bhat fabricated TPA-azaBODIPY-TPA-based OPDs, optimized their performance and performed optoelectrical characterization assisted by Marcin Kielar with control of OPD acquisition. Amandeep Singh Pannu and Gurudutt Bhat performed AFM analysis. Prashant Sonar, Ajay K. Pandey, Lam Yeng Ming, and Pankaj Sah supervised the project. Marcin Kielar, Gurudutt Bhat, and Prashant Sonar drafted the manuscript and all authors contributed to the interpretation of results and assisted with the preparation of the final manuscript.

ACKNOWLEDGMENTS

Lam Yeng Ming would like to thank the support of the Singapore Ministry of Education Academic Research Fund Tier 2 (Grant No. MOE2019-T2-1-085). Annalisa Bruno and Enkhtur thank the National Research Foundation for the financial support (Grant No. S18-1176-SCRP). We would like to acknowledge the Facility for Analysis, Characterization, Testing and Simulation, Nanyang Technological University, Singapore, for use of their electron microscopy and UPS instrument. We would like to acknowledge Central Analytical Research Facility (CARF) at QUT for the use of laboratories and experimental facilities. Prashant Sonar is thankful to QUT for the financial support from the Australian Research Council (ARC) for the Future Fellowship (Grant No. FT130101337) and QUT core funding (Grant No. QUT/322120-0301/07). Ajay K. Pandey acknowledges support from Australia India Strategic Research Fund (Project AISRF53820).

CONFLICT OF INTEREST

The authors declare no potential conflict of interest.