Space charge-induced electrofluorochromic behavior for C12-BTBT-based thin-film devices

Abstract

An organic field effect transistor (OFET), regarded as one of the next generation flexible devices, is developing towards higher computational power and multi-functionality, for which improvements in charge transport properties and diversified collaborative response need to be achieved. 2,7-Docosenyl[1]benzothiofuran[3,2-b][1]thiophene (C12-BTBT) is an established organic small molecule semiconductor with high mobility. However, the current research on C12-BTBT and its devices only reports photoelectric responses as photodetectors and artificial synapses. The limited stimulus response characteristics restrict its functional expansion, thereby hindering its broader application in flexible sensing, display, and logical computation. Here in this work, the electrofluorochromic (EFC) characteristics of C12-BTBT based transistors are realized and a quantitative fluorescence modulation strategy is proposed based on space charge regulation in the dielectric layer. A reversible switching of the fluorescence intensity ratio over 20 is realized by introducing positive/negative space charge accumulation in polymeric dielectrics, modulated by gate voltage, enabling optical output along with the excellent electrical performance of high field-effect mobility over 9.6 cm² s⁻¹ V⁻¹. Such discovered optical-electrical dual output characteristics are further utilized by establishing a 2-input 2-output integrated logic gate computing system, triggered by UV light and electrical signals as input stimuli, with current response and fluorescence signals serving as output indicators, demonstrating the wide application potential of C12-BTBT devices in molecular logic computation and multi-stimulus sensors.

---

Introduction

For a wide range of applications such as sensing, bioimaging, storage media and displays, the precise modulation of the optical properties of materials in response to external stimuli holds paramount significance.^1,2 In recent years, stimulus-responsive fluorescent materials have garnered significant attention due to their potential applications.^3–6 The typical stimuli that have the ability to modify the fluorescence behavior of materials include chemicals, temperature variations, light exposure, pH changes, electrical forces, and mechanical influences.^7–10 Among the various methods available, a more ideal approach to control the fluorescent properties is through electrical manipulation, which is commonly referred to as electrofluorochromism (EFC).³ To explore the mechanism of EFC and achieve more functionalities, various EFC materials have been disclosed, including REDOX active molecules, multilayer/bulk heterojunction composites, and electroactive fluorescent polymers.^11–14

The development of next-generation OFET devices is aimed at multi-response devices, which are critical to advancing integrated systems in areas such as sensing and logical computing.^15,16 EFC materials exhibit reversible changes in fluorescence upon the application of an external electric field, making them crucial for a range of applications such as smart windows, displays, and flexible sensors.^11–14 These materials, when incorporated into OFETs, offer additional capabilities, such as the ability to modulate fluorescence and light emission in response to electric fields, enhancing device interactivity and functionality. While EFC materials have significant potential for integration into organic electronics, several limitations persist, particularly in terms of their compatibility with high-performance OFETs. Many EFC materials suffer from lower charge carrier mobility compared to other organic semiconductors, which can negatively impact the performance of OFETs.^17,18 Besides, most of the existing implementation schemes for EFC materials rely on REDOX reactions and intramolecular charge transfer mechanisms and are based on liquid environments,^11–14,17 which are incompatible with the structure of OFETs, making them difficult to apply for the preparation of OFET devices. Consequently, there is a scarcity of materials and devices that concurrently exhibit both high OFET performance and EFC characteristics. It is necessary to explore new methods and find suitable materials to prepare high performance OFET devices while realizing the EFC phenomenon.

Recent studies have revealed that the accumulation of space charge beneath an external electric field can exert significant influence on the excited state charge transition process during the generation of fluorescence, thereby facilitating the emergence of the EFC effect.^3,19–21 In the process of photoluminescence, electrons and holes are respectively generated in the conduction band and valence band.¹⁶ If the shielding is sufficiently weak, an electron and a hole are attracted by the Coulomb interaction between them, and a bound quasiparticle known as a neutral exciton will be formed.^19,20 When the space charge tunnels to the valence band and acts on the neutral exciton, the neutral exciton can be further charged by combining with an additional electron or hole to form a charged three-body exciton, thereby continuing to play a role in the process of fluorescence generation.^16,17 Space charge injection in OEFT devices can charge neutral excitons, which has been shown to be an effective way to manipulate the photoelectric properties of many materials.¹⁶ The space charge effect provides a more tunable and reversible mechanism for modulating the fluorescence properties of EFC materials, especially in OFET applications, making them more suitable for use in practical organic electronic devices. However, the EFC behavior of organic semiconductor materials with an OFET structure has not been reported.

Benzothiophene (BTBT) and its derivatives are renowned P-type organic small molecule semiconductors, which are commonly utilized in the fabrication of OFETs due to their high mobility and long-term stability.^22–25 As an organic semiconductor with a simple planar building, BTBT exhibits strong π–π orbital overlap and significant sulfur interatomic interactions, leading to higher charge mobility in its derivatives.^23–27 The electroluminescence properties of BTBT derivatives and their applications in light-emitting OFET devices have been reported.²⁸ Light-emitting behavior can be realized under continuous current conduction by using an asymmetric source and drain electrode layout structure, indicating the positive impact of external charge on the luminous process. Hence, injecting and extracting space charges might be an effective approach to modulate the fluorescence intensity of BTBT. Nevertheless, the EFC behavior of BTBT has not been studied or reported.

In this paper, 2,7-didodecyl[1]benzothieno[3,2-b][1]benzothiophene (C12-BTBT) is used to investigate the EFC characteristics of BTBT films in OFET devices under an applied electric field. We prepared OFET devices using the classic bottom-gate/top-contact structure. The devices with no organic dielectric layer use Si/SiO₂ as the gate, and the vacuum evaporation method is used to prepare an 80 nm thick C12-BTBT film as the semiconductor layer directly on the gate. For devices with an organic dielectric layer, the dielectric layer was positioned between the gate and semiconductor through solution processing. Specifically, polystyrene (PS) and atactic poly(4-fluorostyrene) (FPS) dielectric layers were spin-coated onto the gate electrodes prior to semiconductor deposition. FPS has better electronegativity than PS after introducing F atom with strong electronegativity.²⁵ Hence, in this paper, we take advantage of the fact that PS has better capacity for capturing positive charges than FPS and compare the effects of gate dielectric layers with different electronegativity on EFC performance. Photodetector performance of OFET devices based on C12-BTBT under 365 nm illumination is evaluated through the measurement of mobility, photosensitivity, photoresponsivity, and detectivity. Based on these, a logic computing system with 2-input and 2-output is constructed to improve the complexity, diversity and integration of the OFET computing system. In this system, two stimulus input signals of 365 nm ultraviolet light and applied electric field are configured, with current response and fluorescence signals serving as output indicators, demonstrating the wide application potential of C12-BTBT devices in molecular logic computation and multi-stimulus sensors.

Results and discussion

Basic EFC properties of C12-BTBT OFET devices

Fig. 1a illustrates the molecular configuration of 2,7-didodecyl[1]benzothieno[3,2-b][1]benzothiophene (C12-BTBT), while also presenting the space charge injection model for a transistor that has a standard bottom-gate top-contact architecture when a positive gate voltage is applied. The EFC behavior of C12-BTBT is primarily attributed to the injection of space charges and the subsequent charging of neutral excitons. In the single-layer C12-BTBT films which possess P-type semiconductor characteristics, the external electric field generated by the gate voltage induces the intrinsic electron drift of C12-BTBT, regardless of the polarity of the electric field. This leads to the recombination of neutral excitons with electrons arriving from the electrode, ultimately forming charged excitons. When a significant number of neutral excitons are re-charged, the concentration of valence band free electrons will undergo a notable increase. As a result, the number of electrons excited by 365 nm UV radiation experiences a substantial surge, leading to an enhanced fluorescence intensity, as clearly depicted in Video S1 (ESI†). Nonetheless, under protracted illumination, C12-BTBT demonstrates robust photoelectric responsiveness. The liberated electrons facilitate energy conversion through the photocurrent, thereby reducing the number of electrons available for energy conversion via fluorescence. Macroscopically, this results in a decrement of the observed fluorescence intensity as shown in Video S1 (ESI†).

Fig. 1 Fluorescence modulation of the C12-BTBT film by an electric field. (a) Molecular structure of C12-BTBT, electron transport process under gate voltage and fluorescence quenching diagram. (b) Schematic diagram illustrating the mechanism of fluorescence modulation induced by electron transport in the C12-BTBT film under an electric field.

To prolong the EFC effect and induce variant EFC phenomena upon disparate polarities of charge injection, an insulating polymer can be intercalated between the gate and the C12-BTBT layer, serving as a dielectric layer that modulates and preserves charge injection with varying polarities over an extended period. Polystyrene (PS), a prevalent polymer insulator with superior charge trapping capabilities, is extensively utilized in transistors as a charge capture and storage layer^21–23 and upon integration into the device can form an insulating dielectric layer atop the gate. The dielectric properties of the PS used in this work are shown in Fig. S1 (ESI†). PS has a dielectric constant of approximately 2.5 to 2.6 in the temperature range −80 °C to 100 °C at all frequencies. This relatively low dielectric constant is beneficial for reducing parasitic capacitances, which can improve the switching speed of the transistor. The stable dielectric constant across frequencies suggests reliable gate control over the semiconductor channel, enhancing device performance. The loss tangent (tan δ) is relatively low (under 0.001) across the measured frequency range, indicating minimal energy dissipation and efficient charge storage. A low dielectric loss tangent ensures minimal energy dissipation, contributing to the overall efficiency of the OFET. The capacitance per unit area (C/A) of the PS dielectric layer can be calculated using the formula:

C/A = ε₀ε_r/d (1)

where ε₀ is the permittivity of free space, ε_r is the dielectric constant of PS, and d is the thickness of the dielectric layer. For a typical PS dielectric layer thickness of around 300 nm, the capacitance per unit area is in the range of 10–20 nF cm⁻². For breakdown voltage, PS has a high breakdown voltage of over 350 kV mm⁻¹. This high breakdown voltage makes PS suitable for applications requiring high operational voltages and ensures robust performance under high electric fields. High breakdown voltage provides a safety margin for the device operation, preventing accidental dielectric breakdown during voltage spikes or transient conditions. It also allows for higher gate voltages, which can improve the charge carrier density in the semiconductor channel and enhance the overall performance of the OFET. Certainly, PS is studied in this work as a representative insulating dielectric layer material for EFC devices. Other insulating dielectric materials, such as PMMA and PVDF, can also be utilized for this application.

The configuration as shown in Fig. 1a efficaciously injects charges of divergent polarities, enabling their prolonged residence within the device,^27,29 thereby sustaining the EFC phenomenon for an extended duration and allowing EFC variations corresponding to the charge polarity. Fig. 1b shows the principle of different EFC phenomena when charges of different polarities are injected. As shown by the AFM test results in Fig. S2 (ESI†), regardless of the presence or absence of a gate dielectric layer, the surface morphology of the C12-BTBT film prepared by vacuum evaporation is about 3–4 nm, with good surface film uniformity. The preparation of the C12-BTBT film always introduces defects located on the surface and inside the material.^28,30 The defect density in the device can be qualitatively characterized by analyzing the subthreshold swing (SS) value. The average trap state densities (TSD) can be calculated using the following formula to characterize the number of traps inside C12-BTBT films:²⁸

where C_gate is the unit area capacitance of the gate dielectric layer. As evidenced by the comparative data in Table S2 (ESI†), C12-BTBT-based OFET architectures fabricated in this work demonstrate defect concentrations maintaining comparable orders of magnitude to those reported in previous studies.³⁰ These electronically active defects manifest as discrete energy states within the semiconductor bandgap, functioning as efficient charge carrier traps, which exert significant influence on the OFET performance³⁰ and photoluminescence process.²⁸ Upon 365 nm UV excitation, electrons from the VB or intermediary traps ascend to higher energy levels, with some becoming ensnared by traps, creating quenching sites. Application of a positive gate voltage results in the entrapment of injected negative charges by these traps, and as the traps become saturated with negative charges, the transition process sees a diminished electron capture by the traps. Concurrently, an influx of negative charges into the valence band augments the valence band electron concentration. This phenomenon significantly increases the number of electrons transitioned to the CB under UV light, enhancing the film's overall fluorescence (Fig. 2a). Conversely, the injection of positive charges leads to recombination of valence band electrons and excited electrons with these positive charges, significantly curtailing the number of electrons excitable by UV light for transition to the CB. Macroscopically, this culminates in the fluorescence bleaching process, evident from the fluorescence near the top electrode under negative gate voltage, as shown in Fig. 2a. However, carriers in deep traps may not be entirely liberated or recombined through charge injection, suggesting that achieving complete fluorescence bleaching under electrical modulation is challenging. The EFC performance varies when different materials are employed as dielectric layers, attributable to the disparate electron affinity of these layers. As depicted in Fig. S4 (ESI†), the fluorescence intensity of PS exhibits a more pronounced decrease than that of FPS upon the application of a negative electric field. This observation can be attributed to the superior capacity of PS to capture positive charges compared to FPS.²⁵ To better illustrate the difference between PS and FPS for positive charge capture, we apply +100 V gate stress to the OFET devices with PS and FPS as the gate dielectric layer, respectively, within 60 seconds. The charge capture capability can be qualitatively characterized using the following equation:²⁵where Δn is the trapped charge density, ΔV_th is the shift of threshold voltage, C_i is the dielectric capacitor, and e is the elementary charge. As shown in Fig. S3 and Table S1 (ESI†), the results demonstrate that after identical gate stress treatments under +100 V for the same duration, OFET devices employing PS as the gate dielectric layer exhibit significantly higher ΔV_th and Δn compared to their FPS counterparts. This comparative analysis reveals superior capability of PS in attracting and retaining positive charges, as evidenced by the more pronounced ΔV_th shift and increased charge trapping density observed in PS-based devices. Consequently, when positive charges are introduced, devices equipped with a PS dielectric layer are more effective in directing the influx of positive charges, thereby enhancing the EFC effect.

Fig. 2 EFC performance of C12-BTBT film devices. (a) Changes in the fluorescence intensity of C12-BTBT films under different gate voltages. (b) Fluorescence response of C12-BTBT films under continuous alternating voltage of 100 V, at 1 Hz for 1000 cycles.

Fig. 2b presents the ratio of the fluorescence intensity to the initial fluorescence intensity of the sample under a gate voltage of 100 V at 1 Hz for 1000 cycles. The quenching and restoration of EFC films to their original emission state is a rapid and reversible process. The repeatability of EFC is quite reversible and stable, with negligible changes in fluorescence intensity even after 1000 cycles. The data indicate that the changes in fluorescence intensity respond rapidly and exhibit a robust contrast over the course of 2000 consecutive testing cycles involving the injection of positive and negative charges, as demonstrated in Video S2 (ESI†).

In order to further verify the effect of space charge on the EFC phenomenon of C12-BTBT, gate bias stress is applied to C12-BTBT-based devices. Gate bias stress is pivotal in controlling the performance of OFETs, which are extensively utilized in the programming and erasure processes. After gate bias stress is applied to OFET devices, the PS gate insulation layer will trap space charge and form an electret and retain these charges for a period of time.^31–33 In this study, we subjected a C12-BTBT/PS device to a −100 V gate stress. As depicted in Fig. 3a, negative gate voltage stress effectively shifts the transistor's transfer characteristics in the negative direction. Subsequently, as illustrated in Fig. 3b and c, we examined the source–drain current (I_d) at V_g = 0 V and V_d = 60 V for an array of 3 × 4 devices both before and after applying the −100 V negative gate voltage stress. These results demonstrate that the PS layer effectively captures holes, resulting in a corresponding negative shift in the devices’ transfer characteristics.²⁹ Upon gate stress imposition, the PS dielectric layer traps charges, forming an electret,^34–37 while concurrently inducing heteropolar charges in the C12-BTBT layer, as shown in Fig. 3a. These charges contribute to the electron transition process and modulate the fluorescence enhancement or quenching phenomenon when the device is illuminated with 365 nm UV light. Therefore, as Fig. 3c and d reveals, the fluorescence intensity of the C12-BTBT layer within the T-shaped region is augmented post negative gate stress application, attributable to the positive charges in the PS layer inducing negative charges in the C12-BTBT layer. Nonetheless, these trapped charges in the PS layer gradually dissipate over time as they attract and neutralize with ions in the air.^26,38,39Fig. 3f and Fig. S5 (ESI†) demonstrate that after 15 days, the PS layer's captured charges almost entirely dissipate, restoring the transistor's transfer curve to its near-original state, concurrently diminishing the fluorescence intensity enhancement caused by the internal induced charges in the C12-BTBT layer.

Fig. 3 Fluorescence regulation of the C12-BTBT film based on charge injection and the corresponding transistor performance. (a) For the OFET with C12-BTBT as the semiconductor transport layer, the electron distribution diagram of charge injection completion after −100 V voltage is applied to the gate, and the transistor transfer curve before and after the gate stress is applied. (b) and (c) The current at V_g = 60 V before and after the gate stress as shown in the T-graph image of the 3 × 4 matrix. (d) and (e) Fluorescence intensity before and after gate stress as shown in the T-graph image in the 3 × 4 matrix. (f) Transfer curves and fluorescence intensity photographs of the devices at the beginning of space charge injection and after 15 days.

Construction of logic gates based on EFC devices

C12-BTBT, a typical photoelectric semiconductor material, exhibits robust absorption at the 365 nm UV wavelength. The OFET devices fabricated using C12-BTBT demonstrate superior performance, as referenced in recent studies.^24,25,40Fig. 4a and Fig. S6 (ESI†) display the transfer characteristics of the photodetectors under both dark conditions and 365 nm UV illumination. During these measurements, the gate voltage (V_g) was swept from 10 V to −45 V while the drain voltage (V_d) remained constant at −60 V. The current of this OFET reaches 4.26 × 10⁻⁴ A in the saturated state (V_d = −60 V) with an on/off ratio of 1.95 × 10⁶. The expression for mobility (μ) is given bywhere W and L are the channel width and length, respectively; C_i is the capacitance per unit area of the dielectric layer. Based on the statistics of the 20 devices, the average mobility is 9.6 cm² s⁻¹ V⁻¹.

Fig. 4 The optical switching characteristics of the OFET with C12-BTBT as the semiconductor layer under different light intensities. (a) Transfer characteristics of the C12-BTBT-based photodetector measured under 365 nm UV light at different illumination intensities. (b) Typical output curves in the dark and under illumination. (c) Photocurrent as a function of illumination intensity. (d) Photosensitivity, (e) responsivity, and (f) detectivity of the OFETs under various incident illumination intensities with a fixed wavelength of 365 nm. (g) and (h) Image of current mapping of the 3 × 4 matrix tested in the dark and under 365 nm illumination with a “T” shape of the commercial hand lamp.

The photodetectors exhibited a marked increase in the drain current (I_d) and a positive shift in the threshold voltage (V_th) upon exposure to 365 nm UV light, suggesting that the C12-BTBT photoactive layer is highly efficient in carrier generation through photon absorption. Fig. 4b presents the photodetector's typical output characteristics in darkness and under UV illumination. As depicted in Fig. 4c, the photocurrent's dependence on light intensity was plotted, which demonstrated a direct correlation. Notably, at high light intensities, the rate of increase in photocurrent begins to plateau, potentially due to the progressive filling of surface traps.

The expression for photosensitivity (P) depicted in Fig. 4d is given by

P = ΔI/I_dark = (I_ph − I_dark)/I_dark (5)

where I_ph represents the channel current when illuminated and I_dark corresponds to the current in the absence of light. Photosensitivity serves as a critical metric for assessing photodetector performance and a higher P value signifies an enhanced responsiveness to light. The data reveal a progressive increase in photosensitivity concomitant with rising incident light intensity, indicating a direct proportional relationship. Notably, the positive shift in V_th under light exposure culminates in a peak photosensitivity of 1.87 × 10⁶ at a 1.21 mW cm⁻² irradiation level, when the gate voltage (V_g) is approximately 0 V.

In addition, the responsivity (R) of photodetectors, which measures their efficiency while reacting with incident light, is described as:

R = (I_ph − I_dark)/E_eWL (6)

where E_e, W, and L are the incident light power density, channel width, and channel length, respectively. It is observed that R increases with V_g at a constant light intensity, attributable to a rise in the photocurrent. Additionally, we have delineated the variation of R with light intensity, as depicted in Fig. 4e. At a constant voltage, an increase in light intensity results in a reduction of R. Consequently, when V_g is set to −45 V and the optical power is at 70 μW cm⁻², R attains its peak value of 1.03 × 10⁴ A W⁻¹.

To further investigate the photosensitive properties, detectivity (D*) serves as a crucial parameter in the assessment of photodetectors. The formula for D*, as a function of incident light intensity and applied voltage, is as follows:

where S denotes the effective working area, while e signifies the elementary electron. As depicted in Fig. 4f, the device exhibits a remarkable detection capability of 1.17 × 10¹⁵ Jones at a gate voltage (V_g) of 0 V and an optical power of 70 μW cm⁻², attributable to its low dark current and pronounced sensitivity. Notably, the D* experiences a marginal decline with the intensification of light, concomitant with the application of a negative voltage that aligns with the photocurrent variation. The parameters P, R, and D*, pertinent to the photoelectric sensor formulated with C12-BTBT, collectively affirm its exceptional sensitivity to the photoelectric response elicited by 365 nm UV light. In order to better compare the photodetector performance of devices based on C12-BTBT in this work, we listed the performance of several representative state-of-the-art organic photodetectors in Table S3 (ESI†).^41–49 Most organic photodetectors reported in previous literature typically exhibit high responsiveness or detectability, but improving both parameters simultaneously has always been a significant challenge. Fortunately, compared to other organic photodetectors, our C12-BTBT-based device exhibits excellent overall photodetector performance in P, R, and D*.

To further investigate the capabilities of image information capture, we constructed a photodetector matrix comprising 3 × 4 pixels. The uniformity of device performance is demonstrated in Fig. 4g, where the dark current, measured at a gate voltage (V_g) of −40 V, is depicted. As shown in Fig. 4h, the current of each device was measured at V_g = −40 V when the C12-BTBT array was exposed to the ‘T’ optical pattern projected by the photodetector array under the irradiation of a commercial hand-held lamp emitting 365 nm light. Compared to the unexposed pixels (6–8 × 10⁻⁴ A), the channel current in the exposed area was approximately 2 × 10⁻³ A, demonstrating sufficient contrast to illustrate the feasibility of high-resolution imaging and its potential applications in cameras and fax machines.^20,34–36

To expand the application of flexible electronic devices, we fabricated devices on flexible PI substrates using aluminum gate electrodes and PVDF/PMMA as dielectric materials (Fig. 5a). The reason for choosing PVDF/PMMA rather than PS as the dielectric layer of flexible devices is to consider flexibility and mechanical properties. PS is more brittle and less flexible than PVDF and PMMA and is prone to cracking under mechanical stress, making it less suitable for flexible devices.^16,50 In contrast, PVDF is known for its excellent flexibility and mechanical strength. It can withstand repeated bending and bending without breaking or losing its dielectric properties, which is essential for flexible electronics applications. PMMA also has good flexibility and mechanical strength and can be used in combination with other materials to improve its dielectric properties. In the flexible device, we have achieved good mobility and low voltage operation, as shown in Fig. 5b and c. To test the device performance under bending conditions, we attached the flexible devices to cylinders of different diameters. In Fig. 5d, the initial flat device has a bending radius of ∞. The transmission curves of the flexible OFETs between dark and light show no significant difference with different bending radii (r = 2.5 mm, 5 mm, and 20 mm). The results of the output characteristics, shown in Fig. 5e, reflect the sensitive light response performance of the flexible device. The variation of μ, R, P, and D* with bending radius is shown in Fig. 5f and g. The flexibility tests of the device at different bending radii indicate that the flexible devices exhibit great mechanical flexibility with almost no performance degradation.

Fig. 5 Photoelectric response test results of flexible devices. (a) Schematic diagram of flexible photodetectors based on C12-BTBT. (b) and (c) Typical transmission and output characteristics on PI substrates. (d) Transfer characteristics of different bending radii under dark and 365 nm light conditions. (e) Typical output curves in the dark and under illumination. (f) and (g) Evolution of μ, R, P, D* and R.

Based on the experimental results, a logic gate with 2-inputs and 2-outputs is constructed (Fig. 6a). In this system, electric potential and 365 nm excitation light are used as two input signals, which are defined as input A and input B, respectively. Output A is defined as the absolute value of the source drain current picked at V_g = −60 V (|I_d|), while output B is defined as the FL₄₂₀ (|L − L₀|/L₀). The detailed definitions of the two inputs and two outputs along with their corresponding thresholds are given in Fig. 6b. Fig. 6c and d provide all combinations of the two inputs and the corresponding four outputs. The constructed 2-input and 2-output logic system can be regarded as a combination of OR and AND gates. For the combination of the three inputs (10), (01), and (11), |I_d| is greater than the threshold value of 10⁻⁸, so all of the output A is 1. For the other input combination (00), the potential for V_g = 0 V (input A = 0), |I_d| is less than the threshold value of 10⁻⁸, so the output A = 0. For the input combination of (11), the ratio of FL₄₂₀ is higher than 0.3, so the output B is 1. This is because when the voltage of −60 V (input A = 1) is added, positive charge is injected, which results in fluorescence bleaching under the excitation light of 365 nm (input B = 1). For the other three input combinations of (00), (10) and (01), the ratio of FL₄₂₀ is less than 0.3, so output B = 0. To sum up, a 2-input 2-output logic gate has been established.

Fig. 6 The 2-input 2-output logic gate circuit model incorporating EFC characteristics. (a) Schematic diagram of a 2-input 2-output logic gate. (b) Definition of input and output items for logic gates. (c) Current and (d) fluorescence intensity responses of the device with gate stress (V_g = −60 V) and 365 nm UV illumination as input conditions.

Experimental

Materials

C12-BTBT (Fig. S7, ESI†) and poly(4-fluorostyrene) (FPS) were synthesized according to the procedure described in the literature.^24,25 Unless stated otherwise, chemicals and solvents were commercially obtained. PS (M_w = 2000 kDa, PDI = 1.3) and PI film were purchased from Sigma-Aldrich. All materials were used as received without further modification.

EFC device fabrication

Bottom-gate/top-contact OFETs were fabricated on n-type-doped Si/SiO₂ (300 nm) substrates. The 250 μm thick silicon film was cut to obtain a small square piece of 30 mm × 30 mm. The silicon films were washed with water, acetone and isopropanol and then dried with nitrogen before device fabrication. For devices with PS or FPS as the dielectric layer, PS or FPS was dissolved in o-dichlorobenzene to achieve a concentration of 10 mg mL⁻¹ and then spin coated at 2000 rpm onto the Si/SiO₂ substrate. For the C12-BTBT film, 20 mg of C12-BTBT was placed in the crucible, and the crucible was placed in the resistance heating furnace of the vacuum evaporation instrument. When the vacuum degree is less than 10⁻⁵ Pa, the temperature is about 130 °C, at a steady rate of 0.1 Å s⁻¹ (measured by a quartz crystal microbalance), an 80 nm C12-BTBT film was deposited onto PS, FPS or Si/SiO₂via thermal evaporation under a vacuum atmosphere. All the spin-coating and vacuum evaporation processes were conducted in a nitrogen glove box. After forming the dielectric and semiconductor layers, 60 nm gold as the source/drain electrodes was deposited in a vacuum through a shadow mask with pressure below 10⁻⁵ Pa. The OFET characteristics were measured by using an Agilent Keysight B2900A Quick IV measurement system under ambient conditions.

Flexible device fabrication

40 μm of thick PI film was cut to obtain 30 mm × 30 mm small squares. Wash with water, acetone, and isopropanol and then blow dry with nitrogen. 60 nm aluminum was deposited on the surface of the PI film under vacuum less than 10⁻⁵ Pa. PVDF was dissolved in DMC at a total concentration of 10 mg mL⁻¹, and PMMA was dissolved in DMC at a total concentration of 40 mg mL⁻¹, then mix the two solutions in a ratio of 1 : 1. PVDF/PMMA films with a thickness of about 50 nm were prepared on aluminum substrates by solution spin coating in a glove box. 20 mg of C12-BTBT was placed in the crucible, and the crucible was placed in the resistance heating furnace of the vacuum evaporation instrument. When the vacuum degree is less than 10⁻⁵ Pa, and the temperature is about 130 °C, at a steady rate of 0.1 Å s⁻¹, 80 nm thick C12-BTBT was steamed onto the PVDF/PMMA film. Finally, 60 nm gold as the source/drain electrodes was deposited in a vacuum through a shadow mask with pressure below 10⁻⁵ Pa. The flexible OFET characteristics were measured by using an Agilent Keysight B2900A Quick IV measurement system under ambient conditions.

Dielectric characterization

The capacitance, relative permittivity, dielectric loss, and impedance of PS were obtained by applying a 1 V AC small signal onto the sample with the test system of broadband dielectric spectroscopy (Novo control concept 80, Germany) under a nitrogen atmosphere (with Alpha-A sample cell) in the frequency range from 10⁻¹ to 10⁶ Hz, and the temperature range from −80 to 100 °C. Electrical breakdown characteristics of PS were obtained using a sphere-plate electrode clamped to a 150 mm thick sample, and the sphere electrode is connected to a high voltage source for up to 100 kV DC voltage, with the plate electrode grounded. During the test, the sample with the electrode was immersed into insulating oil to avoid surface discharging.

Device characterization

The OFET characteristics were measured by using an Agilent Keysight B2900A Quick IV measurement system under ambient conditions. The AFM images were measured by using the atomic force microscope devices (SPM-9700HT). The EFC characteristic was measured by using a fluorescence spectroscopy detection system (Shaanxi Puguangweishi Technology Co., Ltd), which used a small laser to emit 365 nm continuous light to excite the sample. All fluorescence signals were collected through a 50× objective lens (NA = 0.55) in a confocal fluorescence detection system built with multiple optical components. CCD-19250 receives signals and uses Andor Solis 32-bit software to control the spectrometer (Andor, Newton, SR500i) with a grating of 300 l mm⁻¹ and a focused spot diameter of approximately 2 μm. Collect changes in spectral signals using a CCD (charge-coupled device) camera.

Conclusions

In summary, the EFC phenomenon in C12-BTBT OFET devices is achieved through space charge modulation. By incorporating a dielectric layer to improve the capture and storage of space charges, the EFC characteristics of C12-BTBT films are regulated and maintained. The EFC behavior of C12-BTBT devices is further modulated by the controlling polarity of injected space charges, and when a positive or negative voltage is applied to the gate, the fluorescence intensity can be enhanced or quenched, respectively. The EFC behavior achieved is highly reversible and exhibits long-term cycle stability. The fluorescence intensity switches from the emission state to the quenching state in one second, with the ratio switches by more than 20. Additionally, OFETs based on C12-BTBT exhibit a maximum average mobility of 9.6 cm² s⁻¹ V⁻¹ and demonstrate good photodetector performance under 365 nm illumination with photosensitivity, detectivity and responsivity reaching 1.87 × 10⁶, 1.17 × 10¹⁵ Jones and 1.03 × 10⁴ A W⁻¹, respectively. The dual response to electrical and optical stimulation allows EFC devices to function as basic logic gates. Utilizing these capabilities, a 2-input 2-output logic gate with ‘AND’ and ‘OR’ computing functions is constructed. Two stimulus input signals of 365 nm ultraviolet light and applied electric field are configured, with current response and fluorescence signals serving as output indicators, demonstrating a potential application for OFET and EFC devices in molecular logic computing and multi-stimulus sensors.

Author contributions

Y. Zhu: conceptualization, supervision, resources, methodology, and writing original draft; Y. Jiang: investigation, validation, formal analysis, data curation, methodology, and writing original draft; F. Cao: validation, data curation, and methodology; P. Wang: investigation, formal analysis, data curation; J. Ke: investigation, supervision, and methodology; J. Liu: investigation, formal analysis, and data curation; Y. Nie, G. Li, Y. Wei: data curation; G. Lu: resources; and S. Li: supervision and resources.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (52477028), the Open Research Fund of State Key Laboratory of Polymer Physics and Chemistry (PPCL2023-13), the Youth Fund of State Key Laboratory of Electrical Insulation and Power Equipment (EIPE23407), and the Fundamental Research Funds for the Central Universities (sxzy012023180).

Notes and references

1. A. Tang, J. Wang, J. Xu, J. liu, T. Xu, Q. Zhang and Y. Tian, Biosens. Bioelectron., 2025, 278, 117317.
2. Y. Wang, Y. Wang, L. Wei, A. Li, Y. Fang, L. Li, Q. Li and K. Wang, Chem. Eng. J., 2025, 507, 160849.
3. B. Doppagne, M. C. Chong, H. Bulou, A. Boeglin, F. Scheurer and G. Schull, Science, 2018, 361, 251.
4. S. Xue, Y. Li, W. Pan, N. Li and B. Tang, Chem. Commun., 2023, 59, 3040.
5. W. Zhang, Y. M. Zhang, F. Xie, X. Jin, J. Li, G. Yang, C. Gu, Y. Wang and S. X. Zhang, Adv. Mater., 2020, 32, 2003121.
6. J. Zhang, S. Zhang, J. Liu, Y. Ren, J. Chen, W. Hu and H. Yang, Chem. Eng. J., 2025, 505, 159583.
7. D. Khan, F. Li, G. Day, X. Zhang, F. Jiao and W. Hu, J. Mater. Chem. C, 2024, 12, 11060.
8. Y. Yan, N. Sun, F. Li, X. Jia, C. Wang and D. Chao, ACS Appl. Mater. Interfaces, 2017, 9, 6497.
9. B. Sk, M. Sarkar, K. Singh, A. Sengupta and A. Patra, Chem. Commun., 2021, 57, 13590.
10. W. Xu, X. Li, P. Wu, W. Li, Y. Wang, X. Xu, X. Wang, J. Chen, H. Yang and W. Wang, Angew. Chem., Int. Ed., 2024, 63, e202319502.
11. Z. Lin, H. Pan, Y. Tian, J. Tang, C. Zhang, P. Zhang, H. Yang and J. Chen, Dyes Pigm., 2022, 205, 110588.
12. Y. Jiang, J. Ma, Z. Ran, H. Zhong, D. Zhang and N. Hadjichristidis, Angew. Chem., Int. Ed., 2022, 61, e202208516.
13. Z. Xiong, Y. Li, J. Liang, S. Xiang, Y. Lv and Z. Zhang, ACS Appl. Mater. Interfaces, 2022, 14, 38098.
14. K. Lou, Z. Hu, H. Zhang, Q. Li and X. Ji, Adv. Funct. Mater., 2022, 32, 2113274.
15. R. Isci, O. Yavuz, S. Faraji, D. Gunturkun, M. Eroglu, L. Majewski, I. Yilmaz and T. Ozturk, J. Mater. Chem. C, 2025, 13, 472.
16. Y. Ding, X. Xu, Y. Wu, H. Zhang, L. Shao, Z. Wang, H. Zhang, Y. Zhao and Y. Liu, SmartMat, 2025, 6, e70003.
17. C. Ma, W. Cai, T. Zhou, C. Zhang, Q. Zhang, H. Niu and W. Wang, Sol. Energy Mater. Sol. Cells, 2024, 272, 112884.
18. R. Huang, X. Jia, M. Zhou, Y. Xie and D. Chao, Chem. Eng. J., 2023, 459, 141664.
19. Z. He, Y. Sheng, Y. Rong, G. Lee, J. Li and J. Warner, ACS Nano, 2015, 9, 2740–2748.
20. J. Ross, S. Wu, H. Yu, N. Ghimire, A. Jones, G. Aivazian, J. Yan, D. Mandrus, D. Xiao, W. Yao and X. Xu, Nat. Commun., 2013, 4, 1474.
21. E. Rothenberg, M. Kazes, E. Shaviv and U. Banin, Nano Lett., 2005, 5, 1581–1586.
22. W. Lu, J. Cao, C. Zhai, L. Bu, G. Lu and Y. Zhu, ACS Appl. Mater. Interfaces, 2022, 14, 23709.
23. Y. Zhu, N. Qiao, S. Dong, G. Qu, Y. Chen, W. Lu, Z. Qin, D. Li, K. Wu, Y. Nie, B. Liu, S. Li and G. Lu, Chem. Mater., 2022, 34, 6505.
24. N. Qiao, P. Wei, Y. Xing, X. Qin, X. Wang, X. Li, L. Bu, G. Lu and Y. Zhu, Adv. Mater. Interfaces, 2022, 9, 2200713.
25. Y. Zhu, Y. Fan, S. Li, P. Wei, D. Li, B. Liu, D. Cui, Z. Zhang, G. Li, Y. Nie and G. Lu, Mater. Horiz., 2020, 7, 1861.
26. P. Wei, S. Li, D. Li, H. Yu, X. Wang, C. Xu, Y. Yang, L. Bu and G. Lu, Nano Res., 2018, 11, 5835.
27. P. Wei, Z. Shen, X. Qin, P. Zhang, L. Bu, Q. Chen, S. V. Roth and G. Lu, Small, 2022, 18, e2105896.
28. M. Ullah, R. Wawrzinek, R. C. R. Nagiri, S. Lo and E. B. Namdas, Adv. Opt. Mater., 2017, 5, 1600973.
29. H. Un, P. Cheng, T. Lei, C. Y. Yang, J. Y. Wang and J. Pei, Adv. Mater., 2018, 30, e1800017.
30. T. Hawly, M. Johnson, A. Späth, H. N. Jäkel, M. Wu, E. Spiecker, B. Watts, A. Nefedov and R. H. Fink, ACS Appl. Mater. Interfaces, 2022, 14, 16830.
31. W. Feng, W. Zheng, X. Chen, G. Liu and P. Hu, ACS Appl. Mater. Interfaces, 2015, 7, 26691.
32. S. Park, S. H. Kim, H. H. Choi, B. Kang and K. Cho, Adv. Funct. Mater., 2020, 30, 1904590.
33. P. Wei, X. Li, L. Wang, N. Liu, S. He, Y. Ren, Y. Zhu, Y. Yang, G. Lu and L. Bu, Adv. Electron. Mater., 2020, 6, 1901156.
34. Y. Chiu, I. Otsuka, S. Halila, R. Borsali and W. Chen, Adv. Funct. Mater., 2014, 24, 4240.
35. W. Li, F. Guo, H. Ling, P. Zhang, M. Yi, L. Wang, D. Wu, L. Xie and W. Huang, Adv. Sci., 2017, 4, 1700007.
36. C. H. Chen, Y. Wang, H. Tatsumi, T. Michinobu, S. W. Chang, Y. C. Chiu and G. S. Liou, Adv. Funct. Mater., 2019, 29, 1902991.
37. L. Wang, C. Zheng, J. Fu, J. Hua, J. Chen, J. Gao, H. Ling, L. Xie and W. Huang, Adv. Electron. Mater., 2022, 8, 2200155.
38. Z. Wang, G. Yang, X. Cao, Z. Li, M. Li, T. Zhang, C. Liu, Y. Zhou and Y. Cheng, High Voltage, 2024, 9, 1021.
39. L. Meng, Y. Zhou, R. Wang, X. Cao, Z. Wang and J. Huang, Chem. Eng. J., 2024, 501, 157623.
40. G. Wu, C. Chen, S. Liu, C. Fan, H. Li and H. Chen, Adv. Electron. Mater., 2015, 1, 1500136.
41. Y. Dong, Y. Sun, J. Liu, X. Shi, H. Li, J. Zhang, C. Li, Y. Yi, S. Mo, L. Fan and L. Jiang, Adv. Sci., 2022, 2106085.
42. S. Radhe, J. Subhajit, M. Takaaki and P. Rajiv, Sens. Actuators, B, 2024, 418, 136190.
43. K. Liu, Y. Bian, J. Kuang, X. Huang, Y. Li, W. Shi, Z. Zhu, G. Liu, M. Qin, Z. Zhao, X. Li, Y. Guo and Y. Liu, Adv. Mater., 2022, 34, 2107304.
44. G. Wu, C. Chen, S. Liu, C. Fan, H. Li and H. Chen, Adv. Electron. Mater., 2015, 1, 1500136.
45. J. Zhao, Q. Du, X. Zheng, Y. Liu, S. Li, W. Wang, F. Wang and S. Qin, ACS Photonics, 2023, 10, 2955–2963.
46. S. Qin, X. Qin, Q. Du, Y. Gan, Y. Zhang, A. Wang, X. Yan, R. Dong, Y. Liu, S. Li, C. Liu, W. Wang and F. Wang, J. Mater. Chem. C, 2022, 10, 11710.
47. M. Li, S. Qin, X. Zheng, Q. Du, Y. Liu, S. Li, H. Li, W. Wang and F. Wang, ACS Appl. Mater. Interfaces, 2023, 15, 48442.
48. C. Han, X. Liu, X. Han, M. He, J. Han, H. Zhang, X. Hou, H. Zhou, H. Yu, Z. Wu, J. Gou and J. Wang, Adv. Funct. Mater., 2022, 32, 2209680.
49. S. Fu, R. Song, Y. Wang, Y. Han, C. Gao, J. Ma, H. Xu, B. Li, A. Shen and Y. Liu, Adv. Mater. Interfaces, 2023, 10, 2202488.
50. B. Yang, Y. Lu, D. Jiang, Z. Li, Y. Zeng, S. Zhang, Y. Ye, Z. Liu, Q. Ou, Y. Wang, S. Dai, Y. Yi and J. Huang, Adv. Mater., 2020, 32, 2001227.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tc00564g

---