Bifunctionally Driven Organic Photonic Conversion Devices Facilitated by Minimalistic Synthesis‐Based Interfacial Energetic Alignment

Abstract

Bifunctional integration of indoor organic photovoltaics (OPVs) and photodetectors (OPDs) faces fundamental challenges because of incompatible interfacial thermodynamics: indoor OPVs require unimpeded charge extraction under low‐light conditions (200–1000 lx), whereas OPDs require stringent suppression of noise current. Conventional hole transport layers (HTLs) fail to satisfy these opposing charge‐dynamic requirements concurrently with commercial practicality (large‐area uniformity, photostability, and cost‐effective manufacturability). This study introduces benzene‐phosphonic acid (BPA)—a minimalist self‐assembled monolayer (SAM)‐based HTL with a benzene core and phosphonic acid anchoring group—enabling cost‐effective synthesis and excellent ITO interfacial properties such as energy alignment, uniform monolayer, and stability. This molecular design resolves core limitations and achieves high indoor OPV efficiency (28.6% PCE at 1000 lx LED 2700 K), maintains 93% PCE retention when scaled by ≈220× area, and delivers competitive self‐powered (V = 0 V) OPD performance (noise equivalent power = 584 fW at bandwidth = 1 Hz and wavelength = 730 nm; 3 dB frequency = 103 kHz). Simplified synthesis of BPA reduces production costs by 720% ($0.042 cm⁻²) and achieves 9× higher power‐per‐cost ratio (19.25 mW∙$⁻¹) relative to its counterpart SAM. Synergy between performance and commercial practicality positions BPA‐HTL as a transformative enabler for self‐powered IoT and wearable optoelectronics.

A benzene‐phosphonic acid self‐assembled monolayer (BPA) simultaneously optimizes hole extraction and noise suppression, enabling bifunctional organic photovoltaics–photodetectors that deliver superb indoor performance and commercial viability. BPA‐devices achieve 28.6% PCE at 1000 lx, retain 93% efficiency on large‐scaled devices, and reach 584 fW NEP with a 103 kHz 3dB bandwidth, while cutting production costs by 720%.

1. Introduction

The critical need for indoor energy‐autonomous architectures in next‐generation optoelectronic systems has driven intensive research into organic semiconductors.^{[ 1} , ² , ³ , ⁴ , ⁵ , ^{6 ]} These materials offer inherent advantages, including mechanical flexibility, solution processability, and bandgap‐tunable optoelectronic properties, that enable both indoor power generation (via organic photovoltaics (OPVs)) and spectrally selective photodetection (via organic photodetectors (OPDs)) within unified platforms.^{[ 7} , ⁸ , ^{9 ]} Both indoor OPVs and OPDs typically exploit overlapping narrow‐bandgap semiconductors (absorption range: ≈380–900 nm), which suggests their inherent compatibility for functional integration.^{[ 10} , ¹¹ , ¹² , ^{13 ]} Despite this fundamental synergy, technological developments have been independent. This artificial segregation imposes severe penalties: redundant fabrication inflates costs, multiple process steps increase complexity, and duplicated architectures compromise spatial efficiency in miniaturized systems. Consequently, the persistent decoupling of these complementary fundamentals constrains the development of next‐generation platforms, necessitating bifunctional systems capable of concurrent energy harvesting and photodetection.

Foremost among the challenges in “OPV–OPD” bifunctional integration is the inherent electronic‐level difference between photovoltaic and photodetective operational regimes. While bifunctional OPV–OPD architectures theoretically circumvent discrete component limitations, their self‐powered operation is fundamentally constrained by conflicting charge transport kinetics. Indoor OPVs require unimpeded carrier extraction for sustaining high power conversion efficiencies (PCEs) under weak illuminance (E _v = 200–1000 lx), whereas OPDs necessitate stringent noise current suppression and bias‐free photocarrier acceleration—paradoxical requirements at shared interfacial junctions.^{[ 14} , ¹⁵ , ^{16 ]} This inherent conflict is particularly acute within the electron and hole transport layers (ETLs and HTLs, respectively). Since conventional ETL/HTL configurations are individually optimized for either OPVs or OPDs, they struggle to simultaneously fulfill the conflicting demands of both within a single architecture. Such compromises not only degrade device longevity and stability but also inflate fabrication costs caused by incompatible layer‐by‐layer processing requirements.

The ubiquity of conventional HTLs such as poly(3,4‐ethylenedioxythiophene) (PEDOT:PSS) and self‐assembled monolayers (SAMs) (e.g., 2‐(9H‐carbazol‐9‐yl)ethyl]phosphonic acid (2PACz)) in contemporary architectures stems from their work‐function tunability and interfacial compatibility with narrow‐bandgap photoactive systems.^{[ 17} , ¹⁸ , ^{19 ]} As paradigmatic examples, PEDOT:PSS enables exceptional hole extraction in indoor OPVs, achieving high indoor PCEs (ca., 25–30%) under 500–1000 lx LED illumination, whereas 2PACz‐based HTLs demonstrate unparalleled oxidative stability for OPDs, which is critical for sustaining sub‐pA dark currents.^{[ 1} , ²⁰ , ²¹ , ²² , ^{23 ]} However, hygroscopic sulfonic acid moieties in PEDOT:PSS induce irreversible ITO corrosion via proton‐mediated etching, accelerating sheet resistance increases—a fatal flaw for moisture‐sensitive IoT deployments.^{[ 24} , ²⁵ , ²⁶ , ²⁷ , ^{28 ]} On the other hand, the bulky carbazole in 2PACz has difficulty forming sufficient coverage on metal oxide surfaces on a large scale, thereby creating inefficient charge selectivity in optoelectronic devices.^{[ 29} , ³⁰ , ³¹ , ^{32 ]} Other traditionally used HTLs like MoO_x and V₂O₅ offer high work function and thermal stability but require complex, costly vacuum deposition, limiting scalability.^{[ 24 ]} Likewise, emerging HTLs like NiO_x provide chemical stability and tunability but often need high‐temperature processing, reducing compatibility with flexible substrates and increasing costs.^{[ 33 ]} Consequently, an ideal HTL must concurrently provide 1) energy level alignment with a photoactive layer for unimpeded hole‐selective contact in the OPV mode, 2) charge blocking capability for minimizing noise current in the OPD mode, 3) robust ambient stability combined with simple and scalable manufacturability, and 4) system‐level economic viability, reflected in a high power‐per‐cost ratio under real‐world indoor operating conditions. No existing HTL material satisfactorily addresses this multifaceted challenge set, which creates a significant barrier to the practical deployment of high‐performance, low‐cost, and self‐powered optoelectronic systems.

Our study presents a novel SAM molecule, diethyl(3‐phenoxypropyl)phosphonic acid (BPA), which achieves performance metrics surpassing conventional HTLs under low‐lighting conditions while addressing the cost and scalability limitations. BPA features a minimalistic design that eliminates redundant functional groups while preserving essential charge‐transport motifs. It simply comprises a benzene core and a phosphonic acid (PA) group, which enables ultra‐uniform monolayer formation and robust interfacial stability through its small molecular size and conformal PO₃ ²⁻⋯In/O linkage on the ITO surface. Furthermore, BPA enables π–π stacking distances, stabilized by high interfacial energy, to suppress interfacial defects and recombination hotspots inherent to OPV‐OPD bifunctional systems. This mitigation of carrier‐loss mechanisms eliminates operational degradation during dual‐function operation. As a result, the BPA‐based indoor OPV mode delivered a PCE of 28.6% under 1000 lx LED (2700 K) illumination, which outperforms PEDOT:PSS (PCE = 24.9%) and 2PACz (PCE = 27.5%). Crucially, the simple molecular structure of BPA reduces synthesis complexity, which lowers its production cost to $0.042 cm⁻², a >720% reduction compared to 2PACz ($0.351 cm⁻²) (note that PEDOT:PSS is $0.135 cm⁻²). This cost advantage, combined with a power‐per‐cost ratio of 19.25 mW∙$⁻¹ (compared to 5.26 mW∙$⁻¹ for PEDOT:PSS and 2.25 mW∙$⁻¹ for 2PACz), positions BPA as a commercially viable alternative. Remarkably, devices retained 93% of their initial PCE when scaled by ≈220× in area (from 0.045 to 1.0 cm²). In stark contrast, PEDOT:PSS and 2PACz suffer significant efficiency drops of >10% and >13% under identical scaling conditions, respectively, due to interfacial inhomogeneity. In the OPD mode, the BPA‐based device achieves a low noise equivalent power (NEP) of 584 fW (λ = 730 nm) and a specific detectivity (D ^*) of 5.41 × 10¹¹ cm∙Hz^0.5∙W⁻¹ (device area (A) = 0.1 cm² and bandwidth (Δf) = 1 Hz) at a self‐powered state. In addition, the measured 3 dB frequency of the BPA‐based OPD was 103 kHz, demonstrating a high photodetection speed. This provides a balanced performance in both the OPV and OPD modes, positioning it as a unique and practical solution for self‐powering systems.

2. Results and Discussion

2.1. Preparation of HTL Materials for Devices

This study explores the effect of HTLs on bifunctional organic photonic conversion devices (i.e., indoor OPVs and OPDs). First, HTL materials and device architecture employ an ITO/HTL/PM6:Y6/PDINO/Al structure, as shown in Figure 1a, in which the HTL systematically varied among PEDOT:PSS (AI 4083), 2PACz, and BPA (cross‐sectional scanning electron microscopy image is presented in Figure S1, Supporting Information). The nomenclature and synthesis protocols of the materials are detailed in the Methods section. The absorption spectrum and chemical structure of the PM6:Y6 bulk heterojunction (BHJ) photosensitive layer are presented in Figure S2 (Supporting Information). The energy levels of the SAM molecules were characterized via cyclic voltammetry (CV) and UV–vis measurements (Figure S3, Supporting Information), and the resulting energy‐level diagrams are shown in Figure 1b. The work function (WF) of the ITO/HTLs was quantified using ultraviolet photoelectron spectroscopy (UPS) measurements (Figure S4, Supporting Information).^{[ 34 ]} Compared with bare ITO, ITO/HTLs prepared by the deposition of PEDOT:PSS, 2PACz, or BPA exhibited increased WF values. However, ITO/BPA showed the largest shift in WF (0.77 eV), surpassing PEDOT:PSS (0.35 eV) and 2PACz (0.47 eV). This can be attributed to the minimized molecular structure and PA group of BPA, leading to a strong anchoring ability and the formation of hole‐favored interfacial dipole on the ITO surface. Moreover, the greater shift in WF observed with ITO/BPA compared to ITO/2PACz would be affected by the lower highest occupied molecular orbital (HOMO) energy level of BPA (–6.04 eV) relative to that of 2PACz (–5.38 eV).^{[ 35} , ^{36 ]} The WF variations in ITO/HTLs were further confirmed by Kelvin probe force microscopy measurements (Figure S5a, Supporting Information). Surface contact potential differences relative to bare ITO were –0.46 V for ITO/PEDOT:PSS, –0.53 V for ITO/2PACz, and –0.81 V for ITO/BPA, which are consistent with the WF shifts observed in the UPS measurments (Figure S5b, Supporting Information).

Figure 1.

a) Device architecture, b) energy‐level diagram and the chemical structures of 2PACz and BPA, c) optical properties.

To elucidate the chemical interaction between the ITO/HTLs, X‐ray photoelectron spectroscopy (XPS) measurements were conducted. The analysis of ITO/PEDOT:PSS was omitted because the relatively thicker film (30–40 nm) of PEDOT:PSS hinders the accurate investigation of the electronic state of O 1s for metal oxides (instead, PEDOT:PSS showed a strong O 1s spectrum for C‐O‐C).^{[ 37 ]} The relative atomic percentages of ITO/SAMs are shown in Figure S6 and Table S1 (Supporting Information). Both ITO/2PACz and ITO/BPA showed the presence of phosphorus, as supported by energy dispersive X‐ray spectroscopy mapping for phosphorus (Figure S7, Supporting Information). Importantly, the relative carbon‐to‐indium (C/In) and oxygen‐to‐indium (O/In) ratios increased significantly for ITO/BPA (C/In ≈ 0.603, O/In ≈ 0.604) compared to those of ITO/2PACz (C/In ≈ 0.328, O/In ≈ 0.306). The increase in oxygen content is possibly caused by the Ar─O─C bond from BPA, whereas the higher carbon content is presumed to result from the denser deposition of BPA on ITO than 2PACz. CV measurements were conducted with varied scan rates to determine the areal density of chemisorbed constituents on ITO (Figure S8, Supporting Information). The calculated molecular density of the absorbed SAM molecules on ITO was 8.74 × 10¹³ molecules cm⁻² for BPA and 5.06 × 10¹³ molecules cm⁻² for 2PACz, which is consistent with the XPS data.

Narrow XPS scans for O 1s were analyzed to further investigate the chemisorption of the SAMs on ITO (Figure S9 and Table S2, Supporting Information). The O 1s signal could be deconvoluted into three notable peaks: lattice oxygen (M─O─M), which is concurrent with PA‐metal bonding (P─O─M) at a binding energy of 530.1 eV, oxygen vacancy (V_O) and P═O bonding within PA at 530.8 eV, and hydroxides (M─OH) at 532.5 eV. Further, the peak at a binding energy of 531.2 eV for BPA is attributed to Ar─O─C bonding. Therefore, as shown in Figure S10 (Supporting Information), the relative O 1s constituents were compared based on the M─O─M or P─O─M contents. Based on the calculated values, the content of V_O or P═O for ITO/BPA (1.01) was found to be higher than that of ITO/2PACz (0.82) and bare ITO (0.504). In parallel, the number of M─OH species was significantly lower for ITO/BPA (0.69) than that for ITO/2PACz (1.49) and bare ITO (1.79). These results confirm an increased amount of PA units on the ITO/BPA surface, which contributed to a more effective energy‐level shift of the ITO electrode.^{[ 38} , ^{39 ]} The electrochemical resistance of the ITO/SAMs was evaluated through a series of 25 CV scans, where ITO/2PACz showed a larger change in the CV response than ITO/BPA (Figure S11, Supporting Information). A smaller change in the CV response of BPA indicates that its composition on the ITO surface maintains structural and chemical stability under repeated electrochemical stress. This stability could offer advantages for OPV or OPD devices in terms of charge collection efficiency, interfacial resistance, and long‐term operational lifetime.

Finally, the optical transmittance (T(%)) of the HTL films (Figure 1c) revealed that the SAM‐based HTLs induced only minimal optical signal attenuation, originating from their monolayer structure with Ångström‐scale thickness. In contrast, PEDOT:PSS resulted in a pronounced reduction in T(%) in the 450–500 nm and 780 nm regions. These observations underscore the superior optical transparency of SAM‐based films across the photoactive spectrum, which facilitates more efficient photon transmission and makes them strong candidates for enhanced device performance, particularly under low‐illuminance conditions.

2.2. Morphological Properties of Thin Films

The morphological characteristics of the PM6:Y6 BHJ films on ITO/HTLs were characterized using grazing‐incidence wide‐angle X‐ray scattering (GIWAXS) and atomic force microscopy (AFM) to elucidate the effect of HTLs on molecular packing in the photoactive layer. Lamellar stacking distances (d _lamellar) (100) peaks in the OOP direction, which are primarily affected by the crystallite features of the donor polymer (PM6), for PEDOT:PSS (3.634 Å), 2PACz (3.610 Å), and BPA (3.548 Å) showed negligible variations (Figure 2a), following line‐cut profiles (Figure 2b; Figure S12, Supporting Information). Polymer orientation was further evaluated by calculating the proportion of intensity between χ = 45° and 90° relative to the overall intensity from χ = 0° to 90°, based on azimuthal angle profiles of the (100) diffraction peak at q ≈ 0.28 Å⁻¹ (Figure 2c). Herein, BPA revealed the highest face‐on ratio of 0.70, surpassing those of 2PACz (0.66) and PEDOT:PSS (0.64). This result is likely associated with the surface energy difference between the ITO/SAMs,^{[ 40 ]} specifically the higher surface energy of ITO/BPA (65.6 mN m⁻¹) compared to ITO/2PACz (54.3 mN m⁻¹) (Figure S13, Supporting Information). A higher substrate interfacial energy tends to preserve the dominance of face‐on crystallite orientation.^{[ 41 ]} Although the high surface energy of PEDOT:PSS could promote crystallization trends similar to those observed on ITO/BPA, the acidic (anionic) nature of SO₃ ⁻ introduces electrostatic interactions with BHJ blends, which are minimized when employing ITO/SAMs.^{[ 42 ]} Face‐on oriented crystallites are reported to be more efficient in vertical charge transport than edge‐on counterparts.^{[ 43 ]} Additionally, π–π stacking distances (d _π‐π) obtained from (010) peaks, which primarily represent Y6 crystallites, were similar (3.61 Å) for BHJ films deposited on ITO/PEDOT:PSS and ITO/2PACz. Notably, a reduced d _π‐π of 3.55 Å was observed for ITO/BPA, suggesting that the BPA SAM facilitates denser Y6 crystallite packing through uniform SAM coverage and suitable surface energy.

Figure 2.

Characterization of surface morphology and molecular packing. 2D grazing incidence wide‐angle X‐ray a) scattering images of PEODT:PSS/PM6:Y6, 2PACz/PM6:Y6, and BPA/PM6:Y6, b) line‐cut profiles in the OOP direction, and c) azimuthal angle scans of the (100) diffraction peaks. d) 2D topographic atomic force microscopy profiles of ITO, ITO/PEDOT:PSS, ITO/2PACz, and ITO/BPA.

AFM topography images of the ITO/HTLs are displayed in Figure 2d. The root‐mean‐square (RMS) roughness decreased from 2.3 nm for bare ITO to 1.9 nm after desposition of PEDOT:PSS, reflecting the smooth film‐formation capability of a relative thick polymer coating. In contrast, ITO/2PACz showed an increased RMS roughness of 3.2 nm, whereas ITO/BPA maintained a similar roughness (2.4 nm) to bare ITO, indicative of its ultra‐uniform SAM formation. AFM height images of BHJ films (Figure S14, Supporting Information) further revealed RMS roughness values of 15.5 nm for ITO/PEDOT:PSS, 20.0 nm for ITO/2PACz, and 15.1 nm for ITO/BPA. The exceptional uniformity of chemisorbed BPA and the smoother BHJ film morphology would enhance charge extraction and transport.^{[ 44 ]}

2.3. Performance of BPA‐Based Indoor OPVs

Next, the performance of the indoor OPV devices was evaluated to investigate variations in film properties (Figure 3 ). The devices were assessed under diverse indoor lighting conditions, including LED sources with two color temperatures (2700 and 6500 K) and a fluorescent (FL) lamp (6500 K). Figure 3a illustrates the representative current density‐voltage (J–V) characteristics of the indoor OPVs, and their corresponding indoor photovoltaic parameters are summarized in Table 1 .

Figure 3.

a) Current density‐voltage (J–V) characteristics and maximum power output density (P _out) of PEDOT:PSS, 2PACz, and BPA‐based OPV devices under LED 1000 lx (2700 K) (left), LED 1000 lx (6500 K) (middle), and FL 1000 lx (6500 K) (right) illumination. b) Spectral photon‐flux densities and calculated photocurrent densities (J _ph) of PEDOT:PSS, 2PACz, and BPA‐based OPVs under the same illumination conditions. c) Average P _out of PEDOT:PSS, 2PACz, and BPA‐based OPVs under LED (2700, 6500 K) and FL (6500 K) illumination at varying intensities. d) Average PCE and P _out of PEDOT:PSS, 2PACz, BPA‐based OPVs of different cell sizes (0.045 and 1.0 cm²) under LED 1000 lx (2700, 6500 K) illumination. e) HTL‐specific cost comparison for PEDOT:PSS, 2PACz, and BPA. f) Device stability under LED 1000 lx (2700 K).

Table 1.

Photovoltaic parameters of the indoor OPVs (A = 0.045 cm² and averaged over ten devices).

Light sources	HTL	V OC [mV]	J SC [µA∙cm−2]	FF [%]	PCE [%]	P out [µW∙cm−2]
LED (2700 K) 1000 lx	PEDOT:PSS	644.0 ± 3.3	156.9 ± 1.3	70.8 ± 0.7	24.9 ± 0.5	71.4 ± 0.5
	2PACz	663.1 ± 1.0	162.3 ± 3.7	73.5 ± 2.0	27.5 ± 0.1	78.9 ± 0.1
	BPA	668.0 ± 0.8	165.1 ± 3.3	74.5 ± 1.9	28.6 ± 0.7	82.0 ± 0.7
LED (6500 K) 1000 lx	PEDOT:PSS	640.6 ± 4.0	148.1 ± 2.2	69.9 ± 0.6	22.7 ± 0.6	66.0 ± 0.6
	2PACz	660.8 ± 0.7	147.1 ± 0.2	73.7 ± 0.6	24.5 ± 0.2	71.2 ± 0.2
	BPA	665.3 ± 1.0	152.0 ± 1.6	73.9 ± 0.4	25.6 ± 0.3	74.4 ± 0.3
FL (6500 K) 1000 lx	PEDOT:PSS	636.2 ± 4.7	146.5 ± 0.7	69.9 ± 0.6	21.4 ± 0.4	64.9 ± 0.4
	2PACz	658.6 ± 1.3	148.4 ± 1.7	73.5 ± 0.5	23.6 ± 0.5	71.6 ± 0.5
	BPA	663.0 ± 1.8	150.9 ± 0.8	73.7 ± 0.4	24.3 ± 0.3	73.6 ± 0.3

Selected indoor light sources adhere to specific color temperatures designated by the Commission Internationale de l'éclairage (CIE) in accordance with the International Electrotechnical Commission (IEC) 60904‐3 guidelines. In addition, the indoor light sources were calibrated using CIE‐spectrums “B‐1 (2733 K),” “B‐5 (6598 K),” and “FL‐10 (6500 K)” to ensure corresponding color temperatures.

The indoor OPV performance under LED (2700 K) (E _v = 1000 lx and irradiance (E) = 0.287 mW∙cm⁻²) was first systematically evaluated (device area (A) = 0.045 cm² using an aperture mask). The spectral‐E for standard and measured indoor light sources with different color temperatures is shown in Figure S15 (Supporting Information). Indoor OPVs incorporating PEDOT:PSS exhibited a PCE of 24.9 ± 0.5% (output power density (P _out) = 71.4 ± 0.5 µW∙cm⁻²), a fill factor (FF) of 70.8 ± 0.7%, a short‐circuit current density (J _SC) of 156.9 ± 1.3 µA∙cm⁻², and an open‐circuit voltage (V _OC) of 644.0 ± 3.3 mV (Figure 3a). In contrast, 2PACz‐based OPVs achieved a PCE of 27.5 ± 0.1% (P _out = 78.9 ± 0.1 µW∙cm⁻²), an FF of 73.5 ± 2.0%, a J _SC of 162.3 ± 3.7 µA∙cm⁻², and a V _OC of 663.1 ± 1.0 mV. The enhanced V _OC and FF in 2PACz‐based OPVs can be attributed to the ultrathin (< 3 nm) SAM on ITO, facilitating improved charge‐collection pathways and suppressed surface recombinations. Meanwhile, BPA‐based OPVs demonstrated a PCE of 28.6 ± 0.7% (P _out = 82.0 ± 0.7 µW∙cm⁻²), an FF of 74.5 ± 1.9%, a J _SC of 165.1 ± 3.3 µA∙cm⁻², and a V _OC of 668.0 ± 0.8 mV, reflecting a slight improvement in J _SC and FF compared to their counterparts. This enhancement is primarily ascribed to the simple structure of BPA, which ensures high molecular uniformity and minimizes structural defects. Moreover, as discussed previously, GIWAXS analysis revealed that BPA promotes higher surface energy and, therefore, enhanced crystallinity of photoactive molecules, through tightly packed π–π stacking in the OOP direction. This structural order contributes to improved charge transport and higher J _SC by facilitating efficient charge carrier transport across the HTL, as confirmed by space‐charge‐limited current measurements and charge loss analyses (Figures S16 and S17, Supporting Information). These trends were consistently observed across different color temperature LED lamps (6500 K; E _v = 1000 lx, E = 0.290 mW∙cm⁻²) and FL (6500 K; E _v = 1000 lx, E = 0.304 mW∙cm⁻²) sources. On a side note, the superior PCE under LED 2700 K illumination can be ascribed to the optimal spectral overlap between the spectral power distribution of the light source and the absorption range of the PM6:Y6 photoactive layer. Figure S18 (Supporting Information) presents the PCE and P _out statistical values for each HTL‐based OPV under 1000 lx LED illumination (2700 K). In addition, the J–V characteristics at varying illuminance levels (200–500 lx) under 2700 K and 6500 K LEDs are provided in Figure S19 (Supporting Information). Furthermore, the performance of each HTL‐based device under a near‐infrared indoor source (halogen lamp) at 500 lx (E = 2.97 mW∙cm⁻²) and 1000 lx (E = 4.72 mW∙cm⁻²) is illustrated in Figure S20 (Supporting Information), which demonstrates consistency with that under other light sources. Insights into OPV performance are provided by light‐induced shunt resistance (R _P) analysis (Figure S21, Supporting Information). The elevated R _P observed in BPA‐based devices is a direct consequence of the increased surface energy, which mitigates interfacial defects under low‐light conditions and enhances PCE. In contrast, PEDOT:PSS‐based devices characterized by a thicker interfacial layer (ca., >40 nm) exhibit a lower R _P, inducing surface recombination and pronounced FF degradation.

The spectral photon‐flux density for each light source was used to calculate the photocurrent density (J _ph) (Figure 3b) by multiplying the spectral power distribution of the light source by the external quantum efficiency (EQE), whose spectra are provided in Figure S22 (Supporting Information). The calculated J _ph deviated by ≈5% from the measured J _SC, typically smaller, attributed to aperture usage and sidewall edge effects in the OPV devices. Figure 3c compares P _out of the devices under standard indoor light E _v levels (200 and 1000 lx), highlighting that BPA‐based OPVs consistently achieved higher efficiencies across all tested conditions.

The superior performance of BPA‐based OPVs was further validated through large‐scale device evaluation, where cell sizes were scaled up by ≈220× from 0.045 cm², as shown in Figure 3d (please see detailed J–V characteristic in Figure S23, Supporting Information). BPA‐based OPVs exhibited a significantly lower rate of performance degradation with increasing cell size than the other HTLs, retaining 93% of their initial PCE (under LED 1000 lx (2700 K)). In contrast, PEDOT:PSS and 2PACz‐based devices exhibited significant PCE losses exceeding 10% and 13%, respectively, under identical scaling conditions. This degradation resilience is attributed to the uniform film formation and high morphological stability of BPA, which ensures uniform charge extraction and minimal defect formation across larger areas. In contrast, 2PACz‐based OPVs, though belonging to the same SAM family, demonstrated a more significant decline in performance with increasing cell size. This behavior aligns with previous studies highlighting the difficulty of achieving uniform coating and molecular ordering with 2PACz over large areas because of its greater sensitivity to deposition conditions and a higher tendency to form aggregation‐induced defects.^{[ 45 ]}

Figure 3e shows that a modest improvement in the PCE of BPA‐based OPVs contributes to a competitive advantage in terms of cost‐efficacy. The cost per unit area (C _HTL ($∙cm⁻²)) is defined, where C _HTL represents the HTL‐specific cost determined by the synthesis protocols (please see detailed cost‐calculation information in the Methods section). In principle, the cost evaluation for 2PACz was based on reported literature, while that for BPA was calculated experimentally.^{[ 36 ]} Specifically, C _HTL per unit area for PEDOT:PSS, 2PACz, and BPA were calculated at 0.135, 0.351, and 0.042 $∙cm⁻², respectively. The low cost of BPA is attributed to its simple chemical design, along with cost‐effective reagents. Moreover, the usage of a better wetting processing solvent for SAMs (ethanol) than for PEDOT:PSS (water) grants comparable values to that of the C _HTL value. Consequently, the W∙$⁻¹ under LED (2700 K) illumination (1000 lx) for BPA is ≈9× higher than that of its counterpart SAM (19.25 mW∙$⁻¹ for BPA vs 5.26 mW∙$⁻¹ for PEDOT:PSS and 2.25 mW∙$⁻¹ for 2PACz), emphasizing its economic advantage. This combination of improved efficiency and reduced cost underscores the potential of BPA‐based OPVs to achieve both performance and price competitiveness in practical applications.

Finally, the photostability of indoor OPV is evaluated under ambient conditions at room temperature (Figure 3e). PEDOT:PSS and 2PACz‐based OPVs exhibited shorter lifetimes compared to those incorporating BPA owing to the relative instability of their functional interfaces. PEDOT:PSS suffers from hygroscopicity and acidity that gradually degrade the ITO interface by promoting corrosion and delamination. Similarly, 2PACz can experience weaker anchoring to ITO because of suboptimal molecular interactions, which leads to interfacial defects and accelerated degradation under ambient humidity and oxygen exposure. In contrast, the superior stability of BPA‐based OPVs can be attributed to the simplified chemical structure of BPA, which enhances chemisorption by forming a dense PO₃ ²⁻⋯In/O binding on the ITO surface, minimizing steric hindrance and ensuring a chemically robust anodic interfacial layer. In addition, GIWAXS analysis (Figure 2a) revealed that BPA promotes highly ordered molecular packing, which likely contributes to more cohesive and stable interfaces of the photoactive layer, further enhancing device longevity.

Next, to evaluate photostability, devices were subjected to continuous illumination using a calibrated LED 2700 K light source (1000 lx, 0.287 mW cm⁻²), consistent with the J–V characterization conditions, with PCE measured at ≈24 h intervals for over 1000 h under ambient conditions (25 ± 2 °C, 40–50% RH). Consequently, BPA‐based OPVs maintained 86.9% of their initial PCE after 1000 h of operation, whereas PEDOT:PSS and 2PACz OPVs demonstrated 68.7% and 79.2%, respectively (Figure 3f).

Next, we characterized charge transport and collection efficiency using J _ph​ versus effective voltage (V _eff) to elucidate the origin of FF enhancements via suppressed recombination. Here, J _ph = J _illum – J _d​, and V _eff = V ₀ – V _s (where J _illum is illuminated photocurrent density, J _d is dark current density, V ₀​ is the voltage at J _ph ​ = 0, and V _s is applied voltage). Figure S24a (Supporting Information) presents J _ph–V _eff​ characteristics for all devices under multiple indoor light sources. Under 1000 lx LED illumination, BPA‐based OPV exhibited sharply rising J _ph curves, achieving saturation at significantly lower V _eff​ values (0.218 V). In contrast, PEDOT:PSS and 2PACz‐based OPVs required higher driving voltages (0.226 and 0.305 V, respectively) for current saturation. This accelerated saturation in BPA signifies minimized diode current losses and superior charge extraction efficiency, which is correlated with their enhanced FF. Consistent trends were observed under 1000 lx LED (6500 K) and FL (6500 K) lighting, with BPA reaching saturation ≈28% faster than its counterpart devices. To probe trap‐assisted recombination dynamics under low‐light conditions, we assessed the intensity dependence of V _OC. Figure S24b (Supporting Information) shows the linear correlation between V _OC and ln(P _in), where the slope is theoretically equivalent to kT/q (k: Boltzmann's constant, T: temperature, q: elementary charge). When interfacial defects induce trap‐assisted recombination or impede charge transport at the photoactive layer/electrode junction, V _OC exhibits stronger illumination dependence, yielding slopes below 2 kT/q. Experimentally, the slope for BPA (1.029) approached the theoretical kT/q limit more closely than those for 2PACz (1.329) and PEDOT:PSS (1.588); this implies that the BPA‐modified interface substantially suppressed trap‐assisted recombination and minimized transport losses, resulting in the lowest voltage deficit. To investigate the origin of J _SC improvement in indoor lighting conditions, parasitic resistance characteristics were systematically analyzed using an equivalent circuit model (please refer to the supplementary section). J _SC​ can be approximated as a function of the series‐to‐shunt resistance ratio (R _S/R _P​), where lower values correlate with reduced recombination losses (if J ₀ ≈ 0 (nA∼pA scale) and R _S/R _P ≈ 0, then J _SC ≈ –J _ph).^{[ 46} , ^{47 ]} Figure S24c (Supporting Information) shows that under 1000 lx LED (2700 K) illumination, the BPA‐OPV achieved an exceptionally low R _S/R _P​​ ratio of 0.00194, over 20‐fold lower than that of PEDOT:PSS (0.01095) and 2PACz (0.0085). This reduction arises primarily from the high R _P​ of BPA​, which suppresses interfacial recombination pathways. The minimized R _S/R _P​ ratio signifies enhanced charge extraction efficiency and diminished trap‐assisted losses at the electrode interface, which directly explains the optimal J _SC performance of BPA.

2.4. Performance of BPA‐Based OPDs

Following the characterization of indoor energy‐harvesting performance in the OPV configuration, the optical detection properties of the identical device were subsequently investigated. As illustrated in Figure 4 , the detection capabilities of the three HTL‐based OPDs were quantitatively investigated using thorough noise equivalent power (NEP) measurements for assessing their suitability for indoor environments, where low‐optical power (ϕ) is prevalent. This evaluation is essential to understand how the shared device structure can efficiently balance energy harvesting and optical detection functionalities. NEP represents the ϕ at which a PD achieves a signal‐to‐noise ratio (SNR) of 1. To quantify noise current, steady‐state current transients(I(t_j)) were analyzed as shown in Figure 4a. In each HTL‐based OPD, transient charge dynamics were minimized for accurate I(t_j) characterization by attaining steady‐state conditions after thermal stabilization. A calibrated low ϕ (λ = 730 nm) was applied, inducing a transition to the light‐induced current (I _ph) state (pink region in the figure). Here, the precisely modulated ϕ using a neutral density filter spans six distinct levels to determine the ϕ→NEP.

Figure 4.

a) Steady‐state current transition in dark and under illumination, b) root mean square noise current via background subtraction by I(t_j)–I(t), c) signal‐to‐noise ratio under varying optical power, and d) I–V characteristic of BPA‐based OPD, e) linear dynamic rangers, f) response speeds.

I(t_j) was fitted with a bi‐exponential decay function to model the average dark current (I(t)) in the self‐powered state (0 V) in the thermal stabilization and steady‐state regions. The resultant curves enabled the extraction of the root‐mean‐square noise current (I _rms) via background subtraction (I(t_j)–I(t)). Note that measurement Δf was approximated as 1 Hz based on the sampling frequency of 1/Δt, where median I _rms was derived from multiple measurements (Figure 4b). For OPDs utilizing PEDOT:PSS, 2PACz, and BPA, the I _rms values were 14.2 pA, 21.6 fA, and 17.3 fA, respectively. Notably, SAM‐based HTLs (2PACz and BPA) achieved femtoampere‐scale noise levels, with BPA‐based devices exhibiting marginally lower noise. This suppression is likely attributed to the enhanced molecular ordering and smoother surface morphology induced by BPA, as evidenced by GIWAXS and AFM analyses, which collectively reduce interfacial defects and associated noise contributions. The white noise component of the RMS current (I _rms,white) was calculated using the expression I _rms,white = [(I _rms,shot ²+I _rms,thermal ²)^0.5] = [(4q∙I _d∙A)²+(4k∙T∙R _P ⁻¹)²]^0.5, where I _d is dark current (please see the supplementary information section for full derivations of the noise equations). The computed I _rms,shot values were 66.1 fA for PEDOT:PSS, 767 aA for 2PACz, and 416 aA for BPA, where I _d was derived from steady‐state I(t_j) conditions (Figure S25, Supporting Information). These values are approximately two to three orders of magnitude higher than the measured I _rms, suggesting that the observed I _rms is dominated by low‐frequency noise (e.g., pink noise; Δf = 1 Hz) rather than by shot noise. Note that the bias‐dependent dark current characteristics and frequency‐dependent noise measurements are shown in Figure S26 (Supporting Information). The state‐of‐the‐art performance of these OPV and OPD characteristics can be found in Table S3 (Supporting Information). The superiority of BPA was further confirmed in dual‐functional devices using other photoactive materials, as detailed in Figure S27 and Table S4 (Supporting Information).

Using the acquired median I _rms, NEP was determined under conditions approaching an SNR of 1 (i.e., I _ph–I(t_j)/I _rms = 1), which corresponds to R(ϕ → NEP) with ϕ attenuated (Figure 4c). Given the practical difficulty in achieving an exact SNR of 1, NEP was estimated at the closest achievable point to this threshold.^{[ 48 ]} For BPA‐based OPDs, an NEP of 584 fW was recorded (V = 0 V, λ = 730 nm), yielding a specific detectivity (D ^*) of 5.41 × 10¹¹ cm·Hz^0.5·W⁻¹ (A = 0.1 cm², Δf = 1 Hz). Note that R(ϕ → NEP) was consistent with the nonlinear R behavior observed in prior studies. By comparison, PEDOT:PSS and 2PACz‐based OPDs exhibited D ^* of 1.71 × 10⁹ and 3.27 × 10¹¹ cm·Hz^0.5·W⁻¹, respectively, under the same conditions, owing to their high I _rms values and harsh interfacial properties, which exacerbate NEP from recombination and surface defect‐related mechanisms.

Figure 4d presents the J–V characteristics of the BPA‐based OPD, which shows its on‐current under varying ϕ and off‐current conditions. Analogous J–V curves for PEDOT:PSS‐ and 2PACz‐based OPDs are fitted in Figure S28 (Supporting Information). Building on these results, Figure 4e schematically illustrates the linear dynamic range (LDR) performance of the devices. The LDR values followed a hierarchical trend: 94 dB (PEDOT:PSS), 119 dB (2PACz), and 127 dB (BPA). The superior LDR of the BPA‐based OPD, particularly its enhanced linearity at low ϕ, is attributed to surface smoothness and higher surface energy, mitigating charge recombination losses. Detailed J–V curve fits supporting the LDR calculations are provided in Figure S29 (Supporting Information). Figure 4f shows the transient response characteristics of the BPA‐based OPDs. OPDs with PEDOT:PSS and 2PACz exhibited comparable response speeds, with rise times (T _rise, 10–90%) of 5.699 and 5.589 µs, respectively (Figure S30, Supporting Information). In contrast, BPA OPD demonstrated a markedly accelerated T _rise of 2.756 µs, indicative of improved charge extraction efficiency arising from reduced interfacial recombination enabled by the BPA interlayer.^{[ 49} , ^{50 ]} 3 dB bandwidth measurements of the devices are in coherence with the obtained T _rise (Figure S31, Supporting Information). Paradoxically, this device displayed the slowest fall time (T _fall) of 3.09 µs, which is significantly longer compared to PEDOT:PSS and 2PACz (both ≈1 µs) (Figure S32, Supporting Information). This asymmetric behavior suggests that while rapid charge separation minimizes T _rise, prolonged T _fall may result from capacitive effects, delayed recombination dynamics associated with low noise, or residual current persistence during carrier depletion following illumination termination.^{[ 51 ]} Yet, the BPA‐based OPD exhibited the highest switching stability, robustly maintaining rise and fall times across 30 000 cycles (Figure S33, Supporting Information).

3. Conclusion

We presented BPA, a minimalist benzene core‐PA group‐based SAM, engineered to resolve the fundamental interfacial thermodynamic conflict in bifunctional OPV‐OPD systems. BPA achieves a small molecular size and conformal PO₃ ²⁻⋯In/O linkage on the ITO surface by leveraging a synthesis‐by‐design approach that eliminates non‐essential moieties, which simultaneously enables efficient hole extraction for indoor OPVs and stringent dark current suppression for OPDs. This molecular design mitigates interfacial defects and recombination hotspots via stabilized BHJ's π–π stacking, thereby eliminating operational degradation in dual‐function mode. Crucially, BPA enables outstanding performance, achieving a PCE of 28.6% (1000 lx LED 2700K) as an indoor OPV, while retaining 93% efficiency at ≈220× scaled A (1.0 cm²). As a self‐powered OPD, it demonstrates an NEP of 584 fW (λ = 730 nm), D ^* of 5.41 × 10¹¹ cm· Hz^0.5·W⁻¹, and a 3 dB frequency of 103 kHz. The streamlined synthesis reduces production costs by 720% ($0.042 cm⁻²) and achieves a power‐per‐cost ratio of 19.25 mW∙$⁻¹, which is nine times higher than a conventional 2PACz‐based device. Furthermore, the ultrathin, Ångström‐scale BPA monolayer suggests potential for mechanical stability and flexibility in IoT and wearable applications, likely reducing interfacial strain compared to thicker HTLs like PEDOT:PSS, and is compatible with flexible substrates like PET, pending future bending tests. This synergy of high efficiency, scalability, stability, and economic viability positions BPA as a transformative enabler for self‐powered IoT and wearable optoelectronics.

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

S.O., H.C.K., J.H.L., and T.H.K. contributed equally to this work. S.O. performed the experiments, measurements, data curation, and investigation, and drafted and revised the manuscript. H.C.K. performed the experiments and measurements. J.H.L. synthesized, processed, and characterized the materials and electrodes, and drafted the manuscript. T.H.K. performed data curation and investigation and wrote the manuscript. O.H.K. assisted in experiments. E.S.S. assisted in thin‐film analyses. J.W.J. and J.W.S. provided supervision throughout.

Supporting information

Supporting Information

Oh S., Kim H. C., Lee J. H., et al. “Bifunctionally Driven Organic Photonic Conversion Devices Facilitated by Minimalistic Synthesis‐Based Interfacial Energetic Alignment.” Adv. Mater. 38, no. 1 (2026): e12209. 10.1002/adma.202512209

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.