Molecular N‐Type Doping Unlocks Low‐Threshold Nanosecond Lasing in a Microcavity‐Integrated OLED Toward Electrically Pumped Organic Lasers

ABSTRACT

Organic semiconductors are attractive for the development of flexible, wavelength‐tunable lasers. However, most reported organic micro/nanolasers rely on femtosecond‐pulsed optical pumping, which is impractical for real‐world applications. This limitation has urged the pursuit of electrically pumped organic lasers; yet their realization remains a long‐standing challenge primarily due to a fundamental materials dilemma, in which high‐gain organic semiconductors often suffer from poor, unbalanced charge transport. Here, we demonstrate that this intrinsic trade‐off can be effectively alleviated through a molecular doping strategy. Employing a high‐gain spirofluorene derivative as the emissive layer, we introduce an n‐type doped layer to construct an organic light‐emitting diode (OLED), achieving more balanced charge transport while preserving outstanding optical gain. Consequently, singlet‐polaron annihilation is significantly suppressed, as evidenced by reduced efficiency roll‐off and electrically pumped transient absorption measurements. When integrated with a distributed feedback (DFB) resonator, the resulting device exhibits ultra‐narrow (∼2 nm) electroluminescence under pulsed current injections and delivers low‐threshold nanosecond lasing under an optical–electrical co‐pumping configuration, thereby demonstrating a practical architecture for implementing organic laser diodes. Our work provides a general strategy to overcome the intrinsic paradox where high‐gain organic semiconductors struggle to maintain balanced charge transport, illuminating a pathway toward light amplification under electrical excitation.

Through introducing an n‐type doped layer into the organic light‐emitting diode (OLED), we overcome the intrinsic paradox where high‐gain organic semiconductors struggle to maintain balanced charge transport. As a result, singlet‐polaron annihilation is significantly suppressed while outstanding optical gain is preserved. When integrated with a distributed feedback resonator, the resulting device delivers low‐threshold nanosecond lasing under optical–electrical co‐pumping, demonstrating a practical architecture for implementing electrically pumped organic lasers.

1. Introduction

Organic semiconductors have attracted sustained interest in a wide range of optoelectronic devices owing to their tunable electronic structures and excellent solution processability [1, 2, 3, 4, 5]. Notably, organic solid‐state lasers (OSLs) are particularly appealing for applications in optical communication [6], laser display [7, 8], and chemical sensing [9, 10], benefitting from broad spectral tunability, mechanical flexibility, and high sensitivity of organic semiconductors [11, 12, 13, 14]. Since the early demonstration of optically pumped lasing from organic molecular crystals [15], substantial progress has been made in the development of new gain materials and resonator architectures, ultimately leading to wavelength‐tunable organic lasers spanning from violet to near infrared with greatly improved performance [16, 17]. Nevertheless, OSLs still rely on femtosecond‐pulsed optical pumping, limiting their integration into portable, on‐chip photonic systems [18, 19, 20]. Electrically pumped organic lasers are therefore highly desirable yet remain elusive even under intense nanosecond‐pulsed electrical excitation [21, 22, 23]. This is fundamentally constrained by a long‐standing molecular design dilemma, where the intrinsic trade‐off between achieving high optical gain and maintaining balanced charge transport in organic semiconductors [24, 25, 26]. High optical gain would ensure sufficient singlet excitons generation, which typically requires strong spatial overlap between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO). In contrast, balanced hole/electron transport, which is critical for minimizing polaron accumulation and polaron‐induced exciton losses, demands spatial separation of these frontier molecular orbitals [27, 28, 29]. This incompatibility prevents the establishment of population inversion in organic materials under nanosecond‐pulsed electrical pumping [24, 25, 26].

To address this gain‐transport trade‐off, ambipolar organic gain materials have been developed through molecular engineering incorporating donor and acceptor substituents [30, 31, 32, 33, 34]. Although such strategy could improve charge balance, they will inevitably compromise optical gain due to the limited overlap of frontier molecular orbitals [27, 28, 34]. As a result, it is essential to carry out device‐level engineering to balance charge transport without sacrificing optical gain [24], thus mitigating polaron‐related losses [25, 34, 35, 36]. For instance, conventional organic light‐emitting diode (OLED) architectures employ tailored hole‐ and electron‐transport layers to balance charge injection and transport [37, 38]. However, due to limited carrier mobilities of organic transport layers [39], OLEDs still suffer from unbalanced charge transport, particularly under ultrahigh current densities (∼kA cm⁻²) required for lasing [25, 26, 34, 40]. Doping, a long‐established approach for tuning electrical properties of inorganic semiconductors, has recently emerged as a key enabler to improve charge mobilities and electrical conductivity by orders of magnitude in organic semiconductors [41, 42, 43, 44, 45, 46]. In this context, a targeted doping strategy could significantly enhance minority‐carrier transport [41, 43], providing a more effective solution to achieve charge balance in high‐gain organic materials.

In this work, we demonstrate a molecular n‐type doping strategy that resolves the long‐standing trade‐off between high optical gain and balanced charge transport in organic semiconductors, which enables low‐threshold lasing at nanosecond timescale in a microcavity‐integrated OLED, bridging the gap between femtosecond‐pulsed optically pumped lasing and electrically pumped lasing. A spirofluorene derivative, 2,7‐bis[4‐(N‐carbazole)phenylvinyl]‐9,9'‐spirobifluorene (CzPVSBF), was selected as the light‐emitting gain layer because of its excellent gain characteristics and negligible triplet‐related losses. By designing an n‐type‐doped OLED architecture where lithium 8‐hydroxyquinoline (Liq) doped CzPVSBF serves as electron transport layer, we simultaneously achieved outstanding optical gain and balanced charge transport, effectively suppressing polaron‐related losses compared with a conventional multilayer OLED, as validated by electrically pumped transient absorption measurements. When integrated with a distributed feedback (DFB) grating, the resulting microcavity device exhibits low‐threshold nanosecond lasing under optical pumping and ultranarrow electroluminescence (full‐width at half‐maximum = 2 nm) under extremely high current densities of exceeding 10 kA cm⁻². Crucially, optical–electrical co‐pumping experiments directly verify that n‐type doping mitigates the gain‐transport trade‐off, which is beneficial to alleviate polaron‐induced losses, thereby achieving robust optically pumped nanosecond lasing from an operating OLED integrated with a DFB resonator.

2. Results and Discussion

Spirofluorene derivatives, composed of two fluorene units bridged by a central sp³‐hybridized carbon atom, have been extensively developed and employed in organic optoelectronics community, because they not only inherit excellent optoelectronic features from fluorene derivatives but also exhibit enhanced thermal and chemical stability [47]. Among these materials, CzPVSBF [48] stands out for its exceptional optical gain and negligible triplet‐related losses (Figure 1). According to natural transition orbitals (NTOs) analysis (Figure 1a), the rigid π‐conjugated backbone of CzPVSBF leads to substantial overlap between the highest occupied NTO (HONTO) and lowest unoccupied NTO (LUNTO), yielding a high emission oscillator strength (f _em) of 2.8, and finely structured absorption and photoluminescence (PL) spectra (Figure 1b). Therefore, neat CzPVSBF films and corresponding DFB devices exhibit outstanding gain characteristics under 400‐nm femtosecond laser excitation (Figures S1 and S2). At low pump fluences, the PL spectrum of the neat CzPVSBF film closely matched its steady‐state emission profile (Figure 1b). As the pump power gradually increased, an intense emission peak abruptly emerged at around 466 nm, accompanied by a narrowing of full‐width at half‐maximum (FWHM) to below 8 nm (Figure 1c). This spectral evolution indicates the occurrence of amplified spontaneous emission (ASE) with a low threshold (P _th) of around 4.91 µJ cm⁻² (Figure 1d).

FIGURE 1.

Photophysical characteristics of neat CzPVSBF films. (a) Molecular structure (top) and NTOs at optimized S₁ geometry (bottom) of CzPVSBF. (b) Normalized absorption and PL spectra of a CzPVSBF thin film. (c) Emission spectra of a 100‐nm‐thick CzPVSBF film on a quartz substrate under femtosecond pulsed excitation (400 nm, 1 kHz) at various pump fluences. (d) Plots of emission intensity and FWHM as a function of pump density. (e,f) Transient absorption spectra measured with femtosecond (e) and nanosecond (f) timescale for an encapsulated CzPVSBF neat film.

More importantly, CzPVSBF exhibited negligible triplet absorption within its gain region, which is essential to avoid the triplet‐induced quenching of singlet excitons under electrical excitation. Figure 1e,f show the transient absorption spectra of an encapsulated CzPVSBF film under 355 nm laser excitation, measured with femtosecond and nanosecond timescales, respectively. The negative band of around 455–500 nm corresponds to stimulated emission window of CzPVSBF, which is fully separated from its triplet absorption band at 504–750 nm (Figure S3). This complete spectral separation ensures that singlet excitons could efficiently contribute to optical gain without being depleted by triplet excitons. Taken together, high optical gain and the spectral separation of stimulated emission from triplet absorption position CzPVSBF as an ideal model compound for elucidating how the gain‐transport trade‐off in electrically pumped lasers can be alleviated.

To address the paradox that high‐gain organic semiconductors intrinsically struggle to maintain balanced charge transport, we designed an n‐type doped OLED architecture considering that electron mobilities of spirofluorene derivatives are typically an order of magnitude lower than their hole mobilities [39, 47]. The device structures of conventional OLED (hereafter referred to as the reference OLED) and n‐type doped OLED are shown in Figure 2a,b, respectively. The effective electroluminescence (EL) area of both OLEDs is around 300 × 400 µm² (Figure S4). In the reference device, N,N'‐bis(3‐methylphenyl)‐N,N'‐diphenylbenzidine (TPD) and 1,3,5‐tris(3‐pyridyl‐3‐phenyl)benzene (TmPyPB) films function as hole and electron transport layers, respectively. In the n‐type doped device, Liq was introduced as an n‐type dopant in CzPVSBF to ensure efficient electron transport [41] (Figures S5 and S6), and MoO₃ layer serves as a hole‐injection layer. It is worth mentioning that this n‐type doping strategy does not obviously harm the optical gain property of CzPVSBF (Figure S7). The EL spectra of both OLEDs under direct current (DC) operation (Figure 2c,d) exhibited similar curve profiles to the PL spectrum of the CzPVSBF film (Figure 1b), confirming that carrier recombination under electrical excitation occurs within the CzPVSBF emitting layer. Moreover, the EL peak positions of both devices remained essentially unchanged with increasing forward bias voltage, indicating a stable carrier recombination zone without shifts in either OLED.

FIGURE 2.

EL performances of reference and n‐type doped OLEDs under DC operation. (a,b) Schematic diagrams of reference (a) and n‐type doped (b) OLEDs. (c,d) EL spectra of reference (c) and n‐type doped (d) OLEDs at different forward biases. (e,f) Current density‐voltage‐luminance (J‐V‐L) curves (e) and EQE‐J curves (f) of reference and n‐type doped OLEDs. Inset: a photograph of an n‐type doped OLED under DC operation.

Figure 2e,f illustrates the corresponding current density–voltage–luminance (J–V–L) and external quantum efficiency‐current density (EQE‐J) characteristics of both devices. In comparison with the reference OLED, the n‐type doped OLED exhibited markedly improved EL characteristics, including a lower turn‐on voltage of around 4.8 V, a higher maximum luminance of 5041 cd m⁻², and a larger maximum EQE of 1.21%, compared to 6.7 V, 1095 cd m⁻², and 0.93%, respectively, for the reference device (Figure 2e,f). Additionally, the n‐type‐doped device exhibits greatly suppressed EQE roll‐off at high current densities, with the EQE decreasing to half of the maximum value at approximately 1031 mA cm⁻², compared with only 138.2 mA cm⁻² for the reference OLED (Figure 2f). Notably, the n‐type doped device exhibits improved operational stability under both DC and pulsed operation at high current densities compared to the reference device (Figure S8). The improved performance observed in n‐type doped devices should be ascribed to balanced charge transport that effectively suppresses polaron accumulation and polaron‐induced exciton losses [43].

We then fabricated corresponding hole‐only devices (HODs) and electron‐only devices (EODs) to investigate their charge transport behaviors (Figure S9). As shown in Figure S10, the J–V curves of HOD and EOD for reference device progressively diverged with increasing voltage, suggesting unbalanced hole/electron transport. Conversely, the J–V curves for n‐type doped device closely overlapped across the entire voltage range, indicating substantially more balanced hole/electron transport. Based on the J–V data, carrier mobilities were extracted using the space‐charge‐limited current model [49], J = (9/8)ε _r ε ₀ µ(V ²/L ³), where ε _r is the relative permittivity, ε ₀ is the vacuum permittivity, µ is the carrier mobility, and L is the thickness. As shown in Figure 3a,b, fitting these curves with the above equation yielded hole mobilities (µ _h) of 3.96 × 10⁻⁴ cm² V⁻¹ s⁻¹ for the n‐type doped device and 4.45 × 10⁻⁴ cm² V⁻¹ s⁻¹ for the reference device, along with electron mobilities (µ _e) of 3.16 × 10⁻⁴ cm² V⁻¹ s⁻¹ and 8.85 × 10⁻⁵ cm² V⁻¹ s⁻¹, respectively. This corresponds to hole‐to‐electron mobility ratios of approximately 1.25 for the n‐type doped device and around 5.03 for the reference device. In other words, n‐type doping ensures more comparable carrier mobilities, efficiently reducing polaron accumulation and singlet‐polaron annihilation at high current densities [40, 50]. Hence, the n‐type doped OLED exhibited more efficient EL and suppressed EQE roll‐off compared with the reference OLED (Figure 2e,f).

FIGURE 3.

Balanced charge transport suppresses polaron accumulation. (a,b) Charge transport properties of reference (a) and n‐type doped (b) structures. (c) Illustration of the electrically pumped transient absorption (E‐TA) setup. (d,e) Two‐dimensional map of E‐TA spectra for reference (d) and n‐type doped (e) OLEDs at different time delays under pulsed electrical excitation (10 µs pulse width, 1 kHz repetition rate) with a current density of around 1 A cm⁻². (f) The current‐density dependent polaron absorption amplitude in the reference and n‐type doped OLEDs. (g,h) Schematic illustration of the influence of charge transport behaviors on singlet‐polaron annihilation in reference (g) and n‐type doped (h) OLEDs.

Balanced charge transport, along with suppressed polaron accumulation and exciton losses, was further demonstrated by electrically pumped transient absorption (E‐TA) spectroscopy [34, 51, 52, 53, 54]. In the E‐TA measurements (Figure 3c), the OLEDs were driven by a pulsed voltage (10 µs pulse width, 1 kHz repetition rate), with current injection starting at time zero and transient absorption spectra recorded as a function of time delay between the initiation of current injection and spectral acquisition (see Methods for characterization details). As shown in Figure 3d, upon current injection, a prominent positive peak emerged at approximately 436 nm in the reference OLED. This peak intensity gradually increased over time and saturated after about 100 ns (Figure S11). The band‐shape and build‐up timescale of this peak are the characteristics of polaron absorption [25, 34]. To further identify the origin of this feature, we performed E‐TA measurements on single‐carrier devices and transport‐layer‐based devices, as well as under reverse bias (Figures S12–S14). These results collectively establish that the observed signal originates from polaron absorption in CzPVSBF, rather than from charge‐transport layers or the Stark effect. In contrast, no obvious polaron absorption signal was observed in the n‐type doped OLED at 1 A cm⁻² (Figure 3e), attributable to more balanced hole/electron mobilities achieved through n‐type doping (Figure 3a,b). This improved transport balance is further supported by the current‐density‐dependent polaron absorption profiles (Figures 3f and S15). Across all the current densities, the reference device already exhibits pronounced polaron absorption at low current densities, indicating an early onset of polaron accumulation, whereas the n‐type doped device shows a reduced polaron population, particularly at high current densities. In the reference device, the imbalance in charge carrier mobilities resulted in the accumulation of hole polarons within the emissive layer, inducing severe singlet‐polaron annihilation and depleting the population of bright singlets (Figure 3g). The n‐type doping strategy effectively enhances electron transport, thereby establishing more balanced charge transport (Figure 3h). This effectively inhibits excessive hole‐polaron accumulation and the associated exciton quenching, which are otherwise strongly exacerbated at high current densities (Figure 3f) [25, 26, 34]. Overall, this doping‐assisted transport engineering alleviates the intrinsic gain‐transport trade‐off in organic semiconductors and could maintain sufficient singlet populations under electrical excitation, holding the potential for pursuing electrically pumped lasers.

Second‐order DFB gratings were then integrated into the two OLEDs to probe their lasing performance under current injection (see methods for fabrication and characterization details). As shown in Figure 4a, an around 80‐nm SiO₂ layer deposited on the ITO glass substrate was patterned into a DFB grating with a period of 260 nm and an area of 200 × 200 µm², and the functional layers of two OLEDs were subsequently vacuum‐deposited to complete the fabrication of OLED@DFB devices (Figures S16 and S17). Under pulsed current injection with a pulse width of 20–35 ns and a repetition rate of 1 kHz (Figures S18 and S19), the n‐type doped OLED@DFB device exhibited narrow‐band EL emission with a FWHM of around 2 nm, whereas the EL spectra of the reference device showed a broader FWHM of about 7 nm (Figures 4b, S20 and S21). It is worth noting that all reported pulsed current densities are effective average values, obtained by integrating the current density over time and dividing by the pulse width. The rise times of the electrical pulses and EL signals are nearly identical, and the pulse width closely matches the EL emission duration, ensuring that the measured current density is free from numerical inaccuracies arising from transient current overshoot (Figure S22). As the pulsed current density increased beyond 10 kA cm⁻², neither device displayed a pronounced nonlinear increase in EL intensity, and the FWHM remained nearly unchanged, indicating that electrically generated singlet density has not yet reached the population inversion threshold (Figures 4c, S20 and S21) [25]. This limitation could be primarily ascribed to the incomplete suppression of singlet‐polaron annihilation and intolerable thermal degradation at high current densities [17, 25]. Notably, the maximum current density achieved in n‐doped device reaches ∼15 kA cm⁻², placing it among the highest reported for OLEDs to date and approaching the regime required for electrically pumped lasing in organic semiconductors [20, 22, 25, 26].

FIGURE 4.

Characterization of microcavity integrated OLEDs. (a) Schematic device structure of an n‐type doped OLED@DFB device. (b) Evolution of EL spectra for an n‐type doped device under pulsed electrical excitation at various current densities. Inset: EL image of the operating device. Scale bar is 100 µm. (c) Plots of EL emission intensity and FWHM as a function of current density for the corresponding device. (d) Schematic diagram of optical–electrical co‐pumping measurement setup. Inset: cross‐sectional SEM image of n‐type doped OLED@DFB device with a grating period of 260 nm. (e,f) Evolution of PL spectra under nanosecond‐pulsed optical excitation in reference (e) and n‐type doped (f) OLED@DFB devices. (g) Normalized nanosecond lasing intensity of n‐type doped and reference OLED@DFB devices as a function of optical pump fluence without and with pulsed electrical excitation at a fixed current density of around 9 kA cm⁻². This fixed current density was used to inject a considerable amount of polarons and excitons without degrading the device's performance.

To probe the carrier dynamics at nanosecond timescale and evaluate how close these OLED@DFB devices are to an electrically pumped laser, we performed optical–electrical co‐excitation measurements [55, 56] on two microcavity devices (Figure 4d, see methods for characterization details). In this system, a 1‐ns pulsed laser (355 nm, 1 kHz) was employed to closely match the pulse width of electrical excitation, allowing a detailed investigation of the contribution of electrically generated polarons and excitons on population inversion. Figure 4e,f presents corresponding PL spectra of reference and n‐type doped OLED@DFB devices under optical‐only excitation, respectively. When increasing the fluence of 1‐ns pulsed laser, the initial spontaneous emission from CzPVSBF gradually evolved into distinct lasing emission at around 465 nm in both devices (Figure S23). Noticeably, the n‐type doped device exhibited a substantially lower P _th of 39.6 µJ cm⁻², compared with 285.4 µJ cm⁻² for the reference device. This reduction is primarily attributed to the removal of organic charge‐transport layers that would otherwise introduce additional optical losses.

Under synchronized co‐excitation using 1‐ns optical pulse and 20‐ns electrical pulse at an identical repetition rate of 1 kHz (Figure S24), nanosecond lasing was still observed from both OLED@DFB devices. Notably, ∼95 V and ∼142 V were applied to the n‐type doped and reference devices, respectively, ensuring a comparable injected current density of ∼9 kA cm⁻² in both microcavity devices (Figure S18). For the reference device, the lasing threshold P _th increased by approximately 38‐fold, rising from 285.4 µJ cm⁻² under optical‐only excitation (Figures 4g and S23) to 10901.8 µJ cm⁻² under optical–electrical co‐excitation (Figures 4g and S25). This pronounced threshold escalation is attributed to severe singlet‐polaron annihilation induced by excessive carrier injection at such high current densities [25, 34]. By contrast, after effective suppression of singlet‐polaron annihilation via n‐type doping (Figure 3e), the n‐doped device exhibits only an ∼6‐fold increase in P _th from 39.6 µJ cm⁻² under optical‐only excitation (Figures 4g and S23) to 238.8 µJ cm⁻² under optical–electrical co‐excitation (Figures 4g and S25). To the best of our knowledge, this is the first report of organic nanosecond laser with such a low threshold in microcavity‐integrated OLEDs (Figure 4f). The pronounced difference in lasing thresholds highlights that effectively coordinating the gain‐transport trade‐off through n‐type doping of organic semiconductors benefits the lasing performance of the operating OLED@DFB devices. These results emphasize that n‐type doping strategy offers a potential route to overcoming the intrinsic gain‐transport paradox of organic semiconductors, resulting in excellent dual‐function (nanosecond lasing and ultra‐narrow‐band EL) performance in tailored microcavity OLEDs. Further co‐optimization of gain materials, device architectures, and optical cavities should be done to sustain ultrahigh current densities while suppressing triplet accumulation, polaron‐induced losses, and thermal degradation, which altogether are expected to bring electrically pumped organic lasers within practical reach.

3. Conclusion

In summary, we develop an n‐type doping strategy to alleviate the intrinsic gain‐transport trade‐off in high‐gain organic semiconductors, thereby enabling low‐threshold nanosecond lasing in OLED@DFB devices. Unlike conventional OLED architectures, the n‐type doped OLED simultaneously delivers high optical gain and well‐balanced charge transport. When integrated with DFB gratings, the resulting microcavity device yields ultra‐narrow‐bandwidth EL under pulsed current densities on the magnitude of ∼10 kA cm⁻² and, for the first time, demonstrates nanosecond lasing in this tailored OLED@DFB configuration. More importantly, optical–electrical co‐pumping measurements confirmed that n‐type doping strategy effectively mitigates the gain‐transport trade‐off and suppresses polaron‐induced losses, thus promoting robust nanosecond lasing from an operating OLED@DFB device. This work establishes molecular doping‐assisted transport engineering as a general strategy to overcome a central bottleneck in the development of organic semiconductor lasers, offering a clear pathway toward electrically pumped organic lasers.

Author Contributions

Wei Cheng: investigation, validation, visualization, formal analysis, data curation, methodology, writing – original draft, writing – review and editing. Bo Peng: investigation, methodology, validation, data curation. Chenmiao Zhao: investigation, methodology, validation. Leshen Lin: investigation, methodology. Yuhao Xie: investigation, methodology. Zihao Xu: formal analysis, methodology. Boning Wu: investigation, methodology, formal analysis, writing – review and editing. Wenming Tian: investigation, methodology, formal analysis. Yongli Yan: formal analysis, methodology. Jiannian Yao: formal analysis, resources. Kang Wang: conceptualization, funding acquisition, writing – original draft, writing – review and editing, visualization, supervision, validation, methodology, resources, formal analysis. Yong Sheng Zhao: funding acquisition, conceptualization, writing – original draft, writing – review and editing, resources, supervision, formal analysis, methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: anie73111‐sup‐0001‐SuppMat.pdf.

Data Availability Statement

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