Nanophotonic structures energized short-wave infrared quantum dot photodetectors and their advancements in imaging and large-scale fabrication techniques

Abstract

Short-wave infrared (SWIR) photodetectors (PDs) have a wide range of applications in the field of information and communication. Especially in recent years, with the increasing demand for consumer electronics, conventional semiconductor-based PDs alone are unable to cope with the ever-increasing market. Colloidal quantum dots (QDs) have attracted great interest due to their low fabrication cost, solution processability, and promising optoelectronic properties. In addition to advancements in synthesis methods and surface ligand engineering, the photoelectronic performance of QD-based SWIR PDs has been greatly improved due to developments in nanophotonic structural engineering, such as microcavities, localized and propagating surface plasmon resonant structures, and gratings for specific and high-performance detection application. The improvement in the performance of photoconductors, photodiodes, and phototransistors also enhances the performance of SWIR imaging sensors where they have been realized and demonstrated promising potential due to the direct integration of QD PDs with CMOS substrates. In addition, flexible manipulation of the QDs has been realized, thanks to their solution-processable capability. Therefore, a variety of large-scale production process methods have been examined including blade coating, flexible microcomb printing, ink-jet printing, spray deposition, etc. which can effectively reduce the cost and promote commercial application in consumer electronics. Finally, the current challenges and future development prospects of QD-based PDs are reviewed and could provide guidance for future design of the QDs PDs.

---

Dan Wu

Dan Wu, Ph.D., Associate Professor, Deputy Director of the Major in Light Sources and Illumination, Shenzhen Technology University (SZTU). She obtained her Ph.D. degree in electronic engineering from Nanyang Technological University (NTU), Singapore in 2018. Her research focuses on coupled opto-electrical design of high brightness electroluminescent light sources and photodetectors for on-chip integration and optical interconnect.

Genghao Xu

Genghao Xu received his Bachelor's degree from the College of New Materials and New Energies, Shenzhen Technology University, Shenzhen, China, in 2022. He is currently pursuing a Master's degree under the supervision of Associate Prof. Lei Ma and Associate Prof. Dan Wu in the College of New Materials and New Energies, Shenzhen Technology University. His current research focuses on PbS quantum dots and shortwave infrared photodetection.

Yilan Zhang

Yilan Zhang received her Bachelor's degree in light source and illumination from Shenzhen Technology University in 2023. She is currently pursuing her Master's degree under the supervision of Associate Prof. Dan Wu in the College of New Materials and New Energies, Shenzhen Technology University. Her current research focuses on 3D light-field detection and perovskite nanocrystal synthesis.

Lei Ma

Lei Ma, Ph.D., Associate Professor for the College of New Materials and New Energies, Shenzhen Technology University (SZTU). He obtained his bachelor's degree from Huazhong University of Science and Technology (HUST) in 2009 and Ph.D. in Energy and Power Engineering from HUST in 2015. Dr Ma joined SZTU in 2017 with a research focus on industrial energy-saving technology, new energy technology, electronic device heat dissipation technology and material thermal analysis.

Wei Chen

Wei Chen is a tenured Associate Professor at Shenzhen Technology University (SZTU). He obtained his Ph.D. (Dr rer. nat.) from the Technical University of Munich (TUM) in 2020. His research focuses on nanostructure-modulated thin films and optoelectronic devices, particularly colloidal quantum dot photodetectors, with grazing-incidence X-ray scattering (GIXS) as a key analytical technique.

Kai Wang

Kai Wang, Ph.D., is currently a Professor at the Southern University of Science and Technology (SUSTech). His research focuses on quantum dot optoelectronic devices and their applications in displays, opto-information and optical interconnect. He has published more than 200 articles in academic journals, with an h-index of 55. His research achievements were awarded the National Award for Technological Invention and the National Ministry of Education Award for Technological Innovation. He is a recipient of the National Outstanding Youth Scholar award, and was also selected in the World's Top 2% Scientists identified by Stanford University.

---

1. Introduction

Photodetectors (PDs) that can convert light into electronic signals are core components of various modern optoelectronic systems,^1,2 playing important roles in optical fiber communications, infrared imaging, health inspection and autonomous driving as well as in several fields such as food inspection, agriculture, pharmaceuticals and biology.^3–9 According to the detection wavelength, PDs can be roughly categorized as ultraviolet (UV), visible (Vis) light, near-infrared (NIR), short-wave infrared (SWIR), mid-wave infrared (MIR), and long-wave infrared (LIR) PDs. The SWIR range, typically with wavelengths between 1000 nm and 2500 nm, can penetrate well through the atmosphere, haze, dust, and silicon (Si)-based materials, and therefore, it has great application prospects in communication, machine vision, SWIR imaging, and multispectral sensing.^3,10,11 Commercialized SWIR PDs are generally based on traditional semiconductor bulk materials with narrow bandgap such as germanium (Ge) and indium gallium arsenide (InGaAs). These PDs have the advantages of high responsivity and fast response speed.¹² However, these materials are difficult to directly integrate with complementary metal-oxide semiconductor (CMOS) technology due to lattice mismatch, high thermal budget, and the generally expensive requirement of epitaxial equipment (metal–organic chemical vapor deposition (MOCVD) or molecular beam epitaxy). In comparison, colloidal quantum dots (CQDs) have exhibited promising potential as future candidates for PDs due to their wide and tunable bandgap, solution processing capability, as well as low thermoelectric noise.^13,14 Currently, lead chalcogenide (PbX, X = S, Se, Te),¹⁵ mercury chalcogenide (HgX, X = S, Se, Te),¹⁶ indium arsenide (InAs),¹⁷ and indium antimonide (InSb)¹⁸ QDs all have the ability to adjust the bandgap to the SWIR, which provides new paths and opportunities for QD-based SWIR detection and imaging.

For the past decades, tremendous advancements have been achieved in the area of QD-based SWIR PDs. For example, the specific detectivity (D*) of PbS QD-based PDs exceeds 10¹³ Jones in all three device structures (photoconductors, transistors, and photodiodes),¹⁹ while the external quantum efficiency (EQE) of PbSe QD-based PDs can reach 120%.²⁰ Reports on HgTe QD PDs demonstrated a photodiode with a longer cutoff wavelength of 2.6 μm in the SWIR.^21,22 The first InSb QD-based photodiode has recently been developed for SWIR photodetection.²³ From the perspective of material synthesis, the rapid progress of CQD-based SWIR PDs in recent years has mainly benefited from the synthesis process and surface ligand exchange strategies.^24,25 The advancements in synthesis technology have enabled CQDs with high monodispersity, tunable bandgap and longtime stability, and the surface ligand exchange strategy of QDs can effectively passivate surface defects and enhance carrier mobility.²⁵ Apart from the above developments, device architecture engineering, especially the optical structures-incorporated device, can also effectively promote carrier separation, transport, extraction, etc. and therefore enable high-performance and selective spectra detection. For the realization of enhancement of light absorption efficiency, microcavities are used to generate light interference and enhance the light absorption of the device. To address the problem of low detectivity in conventional devices, plasmonic structures have been employed to enhance the performance of the devices. From gold (Au) nanoparticles to core–shell nanocubes, the localized surface plasmon resonance is adopted to increase the absorption coefficient in the CQD layer.^26–28 Moreover, surface plasmon resonance structures have been proposed, aimed at inducing surface plasmon polariton modes through a periodic array structure and effectively improving the responsivity of devices. In order to further enhance the performance of the PDs, grating structures have been proposed, which promote the coupling of light in transmission waveguide media and can increase the light absorption rate by 30 times.²⁹

Recently, more efforts have been devoted to the field of QD-based SWIR imaging chips. Monolithic integration of QDs with the readout chip greatly expands the application of SWIR imaging. Georgitzikis et al. integrated PbS QD-based PDs onto CMOS readout integrated circuit (ROIC) and the thin-film stacks were patterned with photolithography, successfully producing a high-resolution monolithic SWIR imager with a pixel pitch of 40 μm, which has opened the way for the new generation of high-resolution monolithic infrared imagers.³⁰ In recent years, the IMEC research institute has successfully obtained a SWIR image sensor with a pixel pitch of 1.82 μm by photolithography patterning and adopting a three transistor (3T) pixel design.³¹ In 2022, Tang et al. designed a CQD photodiode array compatible with CMOS ROIC and obtained a high-resolution, low-cost infrared imager with image quality comparable to commercial InGaAs imaging chips.^32,33 In order to further promote the commercialization of CQD-based SWIR PDs, large-scale production process strategies have been explored. By optimizing the evaporation rate conditions, a uniform single-step spraying (SSC) CQD film can be obtained, and low-cost and large-scale production of CQD film can be achieved using methods such as inkjet printing and blade coating.^34–36

This review examines the latest research progress of CQD-based SWIR PDs. Firstly, the terminology and figures of merit to evaluate the performance of a PDs are summarized, and then the fundamental operating principles, individual merits and drawbacks, and relative trade-offs of different types of PDs are discussed. Secondly, the development of CQD synthesis methods and thin film preparation techniques are briefly introduced. To effectively improve the performance of PDs, various nanophotonic structure optical strategies, including microcavity, plasmonic structure, surface plasmon resonance structure and diffractive grating, are discussed. Furthermore, the latest application progress in CQD-based SWIR imaging chips formed by high-performance PD arrays is introduced. Subsequently, the strategies in the large-scale preparation process of CQD PDs are highlighted to explore the effective preparation of PDs for arrays. Finally, the opportunities for CQD-based SWIR detection technology are discussed. This comprehensive review will provide reference for the further development of the industrial application of SWIR PDs.

2. Classical device architectures and working principles of typical QD-based PDs

Generally, CQD-based PDs can be categorized into three fundamental structures, namely: photoconductors, photodiodes, and phototransistors.³⁷ The device architectures and schematic diagrams of carrier transport and separation are shown in Fig. 1. The five most important merit parameters are listed: responsivity (R), noise equivalent power (NEP), detectivity (D*), response time (τ), and dynamic range (DR). Table 1 provides a brief description of the basic parameters used to characterize PDs.

Fig. 1 Device architecture and energy band diagrams of QD-based PDs, (a) and (b) photoconductor, (c) and (d) photodiode, and (e) and (f) phototransistor.

Table 1 Photodetector metrics of performance

Metric	Definition	Formula	Description	Unit
Responsivity (R)	Defined as the ratio of photocurrent to the incident optical power.		I light is the current under light illumination, Idark is the dark current. Pin is the incident optical power on the effective area of the detector	A W^−1
	Representing the relationship between R and EQE. Parameters R and EQE reflect the light-response limit of the PDs which indicates the light-absorption ability of the active layer.		q is the elementary charge, h is the Planck constant, c is the speed of light. λ is the light wavelength.
Noise equivalent power (NEP)	Defined as the optical power when the signal-to-noise ratio (SNR) is unity, giving the minimum detectable power per square root of bandwidth.		I n is the noise current spectral density, the total noise current in a PDs include thermal noise (Ith), shot noise (Ish), and low-frequency flicker noise (1/f).	W Hz^−1/2
Detectivity (D*)	Defined as the SNR detected when an optical power of 1 W is incident on the PDs, normalized to a noise bandwidth of 1 Hz for a detector area of 1 cm^2		A is the effective area of the PDs, Δf is the bandwidth, and In is the noise current.	cm Hz^1/2 W^−1 (Jones)
Response time (τrise/τdecay)	τ rise is defined as the time taken for the signal to rise from 10% to 90% of the peak value and τdecay is defined the time taken for the signal to fall from 90% to 10% of the peak value.	—	Response time is usually used to characterize the reproducibility of the photocurrent and dark current and the sensitivity of the PDs.	μs
Dynamic range (DR)	Defined as the range in which the photocurrent increases with the increase of incident optical power. It is within this range of optical power that the detector can be used to detect incident signal.		The linear dynamic range (LDR) is the range in which the photocurrent increases linearly with the increase of optical power. Ideally, the responsivity of the device should remain constant as the light intensity increases.	dB

---

Conventional photoconductors have a simple structure which consists of a photoactive layer and two electrodes (Fig. 1(a)).³⁸ The working principle of photoconductors is the temporary change in resistance or conductivity of the CQD layer on irradiation with incident light (Fig. 1(b)). Due to the secondary photocurrent formed by the injection and transit of carriers of device electrodes, photoconductors often have excellent optical gain, exhibiting considerably high responsivity.³⁹

The common QD-based photodiodes are vertical structures composed of electron transport layer (ETL), hole transport layer (HTL), and CQD layer (Fig. 1(c)). The working principle of a photodiode is mainly to separate and collect the photogenerated electron–hole pairs by creating a built-in electric field (Fig. 1(d)). Since photogenerated carriers need to transport multiple layers of different thicknesses to reach the corresponding electrode and be collected, the device thickness of photodiodes is particularly important.² Due to the potential barrier at the junction of the device, only a small number of unwanted charge carriers can pass through, resulting in a very low dark current in the photodiode, leading to a high SNR.^24,40

CQD-based phototransistors are three-terminal devices consisting of three contacts, specifically source, drain, and gate, and CQD photoactive channels (Fig. 1(e)). Within the CQD layer, charge carrier transport can be changed by adjusting the gate voltage.^41,42 The carrier density and conductivity of the CQD can be better controlled by adjusting the source drain voltage under light irradiation, resulting in improvement in low dark current and high gain.^43,44 Phototransistors can reduce photogenerated carrier recombination and greatly extend carrier lifetime.

From previous research, it has been found that traditional device structures have their pros and cons. There exist issues such as Ohmic contact in photoconductors, relatively low responsivity of photodiodes, and unstable work instability of phototransistors. In general, the overall photoelectric performance of most PDs based on CQD materials still lags behind that of traditional semiconductor materials such as Ge or InGaAs.¹² Therefore, many researchers have improved the performance of CQD-based PDs by optimizing CQD synthesis methods from the view of materials scientists which can be found in many reported results.^37,45–49 The focus of this paper is on the optical strategies used to improve the performance of SWIR CQD PDs, and these representative optical strategies will be comprehensively described in subsequent sections.

3. Typical synthesis and film preparation methods

In the past few decades, the research of QDs has made impressive progress in terms of size, morphology, and crystal structure. The improvement of QD synthesis quality is expected to significantly enhance the performance of QD-based devices.^50–52 Two main strategies to synthesize QDs are: physical vacuum-based methods and wet-chemical methods. Vapor phase epitaxy is a mature technology for synthesizing high-quality QDs with adjustable size. However, this method requires strict substrate lattice matching and expensive equipment such as physical growth equipment, and the synthesized QDs are difficult to separate from the substrate.^53–56 Correspondingly, the chemical solution method, which uses low-cost chemical reagents, is also suitable for large-scale synthesis of highly dispersed QDs with lower energy consumption.⁵⁷ In addition, thin film deposition methods based on ligand exchange also have varying degrees of influence on the charge carrier mobility and carrier lifetime of the thin film. A general review of the development of chemical solution methods for the synthesis of SWIR CQDs and film preparation methods is included here. For more comprehensive content, readers can refer to other reviews that have effectively addressed the issues.^57–60

3.1 Hot-injection organometallic synthesis

The hot-injection organometallic synthesis method, which injects a solution of organometallic precursor into a mixture of organic solvents to obtain monodisperse and highly luminescent nanoparticles, is one of the liquid-phase synthesis methods for synthesizing CQDs.²⁴ In 1993, Bawendi et al. first proposed a hot-injection method to successfully prepare high quantum yield and narrow size dispersions CdSe QDs at a temperature of 300 °C.⁶¹ The revolutionary synthesis method made Bawendi one of the three Nobel laureates in chemistry in 2023. The hot-injection method typically uses a three-necked flask as the synthetic apparatus (Fig. 2a). In the hot-injection method, the synthesis of QDs involves two stages: nucleation and the growth of nuclei.⁶² The precursor solution rapidly decomposes and nucleates under specific conditions (such as anhydrous, anaerobic, and high-temperature). By controlling the relevant reaction parameters, the crystal nucleus stably develops into CQD (Fig. 2b). Since then, the hot-injection method has been widely used in the synthesis of CQDs. In 2001, Murray et al. firstly synthesized monodisperse and size-adjustable PbSe QDs using the hot-injection method.⁶³ Hinse and Scholes synthesis is currently a widely used method for synthesizing PbS QDs.⁶⁴ The substitute InAs QDs for lead-free QDs is synthesized using a continuous injection process to produce large-sized InAs QDs with an absorption rate of 1600 nm.⁶⁵ The hot-injection organometallic synthesis method is suitable for manufacturing high-quality QDs. In this technique, the high reaction temperature (between 150 and 350 °C) helps to eliminate crystal defects, and the size distribution of QDs can be simply optimized by temperature or reaction time.⁶⁶

Fig. 2 (a) The simple apparatus for synthesizing CQD using hot-injection method. (b) Nucleation and growth stages in the synthesis of monodisperse CQD (the La Mer model).⁷⁵ Reproduced from ref. 75 with permission from John Wiley and Sons, copyright 2019. (c) Controlling particle size distribution by cation exchange and quantitative Ostwald ripening.⁷⁶ Reproduced from ref. 76 with permission from American Chemical Society, copyright 2017. (d) One-step direct synthesis of QD ink.⁷¹ Reproduced from ref. 71 with permission from Springer Nature, copyright 2019. (e) Schematic illustration of the core/shell CQDs.⁷⁴ Reproduced from ref. 74 with permission from John Wiley and Sons, copyright 2023.

3.2 The cation exchange method

The cation exchange method is considered an effective method for obtaining QDs with better size monodispersity and is applied in various nanomaterial synthesis studies.⁶⁷ More types of QDs are generated by exchanging cations in QDs in the process. Early explorations of cation exchange used cadmium sulfide (CdS) nanowires as sulfur sources and lead chloride (PbCl₂) as lead sources.⁶⁸ In the reaction, Cd ions in QDs are replaced with Pb ions to form PbS QD. According to Ostwald ripening, when two different sized nanoparticles coexist in a solution, particles with radii below or close to the critical dissolution radius will be dissolved, completely releasing monomers for the growth and size aggregation of larger QDs.⁶⁹ This is the main reason for the high monodispersity in cation exchange synthesis (Fig. 2c). At present, the quality of QDs prepared using the cation exchange method is already high, with the primary focus of research being on the preparation of larger-sized QDs.⁷⁰

3.3 The one-step synthesis

CQDs prepared by hot-injection organometallic synthesis and cation exchange synthesis methods are both encapsulated by long-chain ligands. In order to improve the carrier mobility in QD membranes, these long-chain organic ligands must be exchanged for short ligands. In 2019, Wang et al. proposed a new method for directly synthesizing PbS CQD ink, avoiding the tedious ligand exchange process, enabling the synthesized QD ink to be directly applied to thin film deposition (Fig. 2d). The synthesis method plays a crucial role in both the prevention and passivation of defects in QDs. Compared with traditional methods, the number of synthesis steps is significantly reduced, and the cost of ink production is lowered to below 6 $ per g. This method boasts advantages such as simplicity, speed, and cost-effectiveness.⁷¹ This method also has high value for integration with large-scale fabrication process such as blade coating, spray coating etc. which will be examined thoroughly in a later section.

3.4 Core/shell QDs by growing epitaxial shell

The QD structure formed by epitaxial growth of a secondary semiconductor shell on a semiconductor core is called a core/shell type QD (Fig. 2e).⁷² Due to the fact that the shell can passivate the defect states on the surface of the core layer, it avoids oxidation or hydrolysis of the core layer. Kwon et al. demonstrated through the synthesis of PbS/CdS QDs that the self-passivation of CdS shell significantly improved the lifetime of SWIR PDs.⁷³ Seo et al. coated InAs shell on InSb QDs, verifying the photoluminescence of core–shell QDs in the SWIR region and improving their photostability.²² Core/shell QDs possess advantages such as superior optical properties, robust fluorescence performance, and high photostability, rendering them highly applicable in the SWIR region.⁷⁴

3.5 Film preparation methods

In the synthesis process of CQDs, in order to control the nucleation of QDs, the as-synthesized CQDs are passivated with long organic ligands such as oleic acid or oleylamine. This not only helps prevent the aggregation of QDs but also effectively reduces surface defects of QDs, thereby achieving colloidal stability.⁷⁷ However, the synthesized QDs are not favorable to highly efficient charge carrier transport in the thin film. This requires the substitution of long-chain organic ligands by short ones from the surface of CQD through ligand exchange, reducing the transport spacing between QDs, increasing the charge carrier mobility of CQD thin films, and increasing the surface passivation degree to reduce unexpected charge carrier recombination. Ligand exchange can be divided into two types: solid-state ligand exchange and liquid-phase ligand exchange. Solid-state ligand exchange is mainly used for depositing thin films through a layer-by-layer (LbL) deposition process (Fig. 3).⁷⁸ Specifically, solid-state ligand exchange is achieved by spin-coating QDs coated with long-chain aliphatic ligands onto a substrate, then spin-coating appropriate short chain ligands on the surface of the QD membrane for ligand exchange, and finally removing the original long-chain ligands and excess short-chain ligands using a polar solvent.^79,80 However, this method suffers from the drawback of low efficiency in the utilization of QDs, typically below 1%.^78,81 Correspondingly, the use of liquid-phase ligand exchange enables monolithic deposition (Fig. 3). QDs containing long-chain ligands are dissolved in non-polar solvents, while organic or inorganic molecules containing short-chain ligands are dissolved in polar solvents, and the two are mixed to obtain orthogonal solvents. During the mixing process, long-chain ligands are replaced by hydrophilic short-chain ligands (e.g. short-chain molecules, lead halides, perovskites, etc.), and QDs transferred from non-polar solvents to polar solvents, completing ligand exchange.⁷⁸ Liquid-phase ligand exchange reduces the residues of long-chain ligands, resulting in QD films with higher mobility and longer carrier lifetime.

Fig. 3 Deposition of thin films using solid-state ligand exchange and liquid-phase ligand exchange.⁸² Reproduced from ref. 82 with permission from Springer Nature, copyright 2021.

4. Device architecture engineering with nanophotonic structure

In QD-based PDs, thicker QD films are usually used to obtain longer incident light path, thereby capturing more light. However, excessively thick QD film can in turn limit the transport of charge carriers due to the short diffusion length.⁸³ Therefore, capturing more light in the device while ensuring carrier mobility is a pressing issue that urgently needs to be addressed. The photonic structure mentioned here refers to an optical system that precisely controls the propagation and interaction of light based on optical principles. It can be combined with QD devices to improve device performance.^60,84 Recently, the nanophotonic structures that have improved device performance mainly include four typical optical structures: resonant cavity, plasmonic structure, surface plasmon resonance structure and grating. A resonant cavity or microcavity in its simple form is composed of two reflective mirrors, in which the incident light is repeatedly reflected on the inner surface of the two mirrors, generating light interference and enhancing the light absorption of the device.⁸⁵ The cavity allows for prolonged light interaction with the QD layer, significantly enhancing absorption efficiency. The most commonly reported plasmonic structure is the localized surface plasmon resonance (LSPR) generated by electron photoexcitation oscillation on the surface of noble metal nanoparticles (NPs), such as gold or silver.⁸⁶ Plasmonic structures can increase the local electromagnetic field intensity in the near field of QD film, enhancing light–matter interaction, which improves the light absorption of the device. Surface plasmon resonance structures are typically used on metal surfaces, where free electrons couple with incident photons in the surface region to generate surface plasmons. These surface plasmons propagate along the metal surface, concentrating the light in the near-field region and enhancing light absorption.⁸⁶ By introducing periodic array structures such as gratings or nanoholes, the efficiency of light coupling can be further improved. In addition, the grating structure increases the optical path of the incident light through diffraction, and then promotes the coupling of light in the transmit medium such as QDs to enhance light absorption.⁸⁷ The optical properties of four typical nanophotonic structures commonly used in QD-based devices are examined in detail in this section.

4.1 Resonant cavity

A resonant cavity is typically an optical microcavity formed when a thin active layer is sandwiched by two parallel planar mirrors (electrodes). In an ideal situation, light undergoes multiple reflections in microcavity. This repeated reflection creates constructive interference for certain wavelengths, forming a stable resonant mode. The incident light with resonant frequency can be effectively captured in the optical microcavity, thereby enhancing light absorption (Fig. 4a). The resonant conditions of microcavities can be expressed as follows:
where n is the refractive index of the absorption layer, m is a positive integer number, L is the length of the microcavity, and λ is the wavelength of light.^85,88 This equation describes the resonant condition where the optical path length within the cavity is an integer multiple of the wavelength of the light. By designing microcavities to target specific wavelengths, light absorption in QD layers can be maximized, significantly enhancing device performance, especially in narrowband photodetection.

Fig. 4 Schematic diagram of structures of various devices enhanced by optical microcavity. (a) Schematic diagram of a basic simplified microcavity model. (b) The normal QD solar cell with double pass of incident light (top) and the periodic arrangement of folded-light-path QD solar cell with multiple light passes (bottom).⁸⁹ Reproduced from ref. 89 with permission from Springer Nature, copyright 2013. (c) Schematic diagram of a microcavity with DBR. And comparison in EQE spectrums of cavity-based and control devices.⁸⁵ Reproduced from ref. 85 with permission from American Chemical Society, copyright 2016. (d) Schematic diagram of InAs/GaAs QD infrared PDs structure with microcavity structure.⁹¹ Reproduced from ref. 91 with permission from Optica Publishing Group, copyright 2017. (e) Schematic diagram of Fabry–Perot cavity-enhanced flexible detectors and detectivity during concave bending cycles of flexible HgTe CQDs detectors.⁹² Reproduced from ref. 92 with permission from John Wiley and Sons, copyright 2019. (f) Schematic diagram of device thin film stack structure combined with organic optoelectronic diode.¹¹ Reproduced from ref. 11 with permission from John Wiley and Sons, copyright 2022.

In 2013, Koleilat et al. formed a resonant cavity between metal electrode stacks using a folded-light-path (FLP) architecture with a 45° tilt angle, increasing light reflection to multiple passes and effectively absorbing all photons with energy greater than the CQD bandgap. Compared with its standard counterpart, it effectively enhanced the light absorption (Fig. 4b).⁸⁹ In 2016, Ouellette et al. successfully induced multiple optical reflections in a cavity by integrating a distributed Bragg reflector (DBR) mirror on a glass substrate and utilizing the high reflectivity of DBR to specific wavelengths of light. The devices with resonant microcavities increased the infrared absorption rate by a total of 56% and achieved 60% EQE at the exciton peak, verifying the feasibility of enhancing the CQD-based PD characteristics through the resonant cavity optical strategy (Fig. 4c).⁸⁵ In the same year, Zhang et al. introduced an optical spacer layer into the microcavity, causing a change in the thickness of the device cavity and interference with incident light of specific wavelength, thereby maximizing the light absorption within the QD layer and achieving a power conversion efficiency of 7.3% for solar cells based on glass substrates. The possibility of tuning the thickness of the CQD active layer by microcavity structure has been demonstrated by this work.⁹⁰ In 2017, Kim et al. fabricated InAs/GaAs QD PDs on silicon substrates using a metal wafer bonding and epitaxial lift-off process (Fig. 4d).⁹¹ The resonant cavity was form by designing Pt/Au as the back mirror and GaAs/air as the front mirror in the device structure, effectively enhancing device responsivity by nearly twofold from 0.038 A W⁻¹ to 0.067 A W⁻¹. In 2019, Tang et al. integrated Fabry–Perot resonant cavities with HgTe CQDs on a flexible substrate (Fig. 4e), providing enhanced light response through controllable spectral features, resulting in a peak detectivity of 7.5 × 10¹⁰ Jones which is far superior to other flexible infrared detectors.⁹² In 2023, Kim et al. replaced the traditional ITO bottom anode with an ITO/Ag/ITO anode, inducing a strong microcavity effect and demonstrating an efficient top-emitting infrared-to-visible light up-converter device, resulting in device performance with an EQE of up to 15.7%.⁹³

Apart from the classical device architecture, a double junction structure is a PN-NP or NP-PN structure formed by connecting two PN or PIN junctions in series. The working principle of the double junction structure is that QDs in two PN junctions have different bandgaps. By controlling the bias of positive or negative, a different part of the spectrum can be detected. For double-junction structures that can be used for multispectral imaging in SWIR, microcavities can also play an important role. In 2022, Pejović A et al. proposed a dual-band PD for multispectral sensing.¹¹ To address the issue of spectral crosstalk in the device, a complete dual photodiode stack was designed by researchers. A thin layer of Ag was placed between the two photodiodes, creating a microcavity between the back photodiode and the back metal electrode, thus enhancing the light absorption in the front photodiode to reduce spectral crosstalk (Fig. 4f). The obtained device increased the EQE in the SWIR range to 30% and successfully achieved broadband multispectral sensing from visible light to SWIR.

4.2 Localized surface plasmon resonance structures

Surface plasma refers to the collective oscillation of free electrons located at the interface between metal and dielectric. The NPs which present surface plasma are called plasmonic NPs and exhibit the phenomenon of localized surface plasmon resonance (LSPR).⁹⁴ Noble metals Au and Ag are the most common candidates for plasma metal NPs that exhibit strong plasma frequencies under incident light irradiation.⁹⁵ These materials can be shaped into spheres, nanorods, nanodisks, and micro gears to enhance the performance of PDs. When light incidents onto the metal NPs with size smaller than the wavelength, the electron cloud oscillates under the combined effect of electric field and coulomb force generated by charge displacement, causing the surface electron cloud to deviate from the atomic nucleus. The coulomb force also affects unbound electrons, causing charges to accumulate at opposite ends of the structure. These charges will generate a depolarizing field, generating a restoring force on the electron cloud, causing oscillation of the electrons. When the inherent frequency of this oscillation is the same as the frequency of the external incident light wave, LSPR occurs (Fig. 5a). The frequency of LSPR, ω_LSPR, is related to the plasmonic frequency, ω_P, of the metal and the dielectric environment, and can be approximately described by the following equation:
where ω_P is the bulk plasma frequency, ε_m is the permittivity of the metal, and ε_d is the dielectric constant of the surrounding medium.^86,94 A strong localized electromagnetic field is generated near the surface of metal NPs when resonance occurs, which improves the distribution of localized photoelectric fields, giving it the ability to concentrate near-field light fields and enhance far-field light scattering in the vicinity of the metal structures.⁹⁶ Therefore, researchers can manipulate the LSPR of metal nanostructures by designing geometric shapes, sizes, and compositions to capture specific wavelengths of sunlight, thereby improving the absorption of the CQD active layer by effectively adjusting the light field distribution of the device. These NPs can be directly mixed with QDs without the need for additional optical structures, which is easy to achieve while effectively improving device performance.⁹⁷

Fig. 5 (a) Schematic diagram of LSPR principle. (b) Schematic of the HgTe QD photodiode detector with Au nanorods embedded in ZnO layer and representative J–V curves of control devices without Au nanorods, devices with 4.5 and 7.5 nm ZnO-coated Au nanorods, and devices with Au nanorods in direct contact with the QD layer.²⁶ Reproduced from ref. 26 with permission from American Chemical Society, copyright 2014. (c) Schematic demonstration of photoconductive PD with a PbS CQD/Ag NPs composite active layer and responsivity of composite devices with and without the addition of 1% Ag NC.⁹⁸ Reproduced from ref. 98 with permission from American Chemical Society, copyright 2014. (d) Schematic diagram of the PbS QD solar cells with the dual-plasmonic effects of Au and Ag NPs.⁹⁹ Reproduced from ref. 99 with permission from Springer Nature, copyright 2020. (e) Schematic diagram of PD configuration based on Au NSs/PbS.¹⁰⁰ Reproduced from ref. 100 with permission from American Chemical Society, copyright 2023. (f) Schematic diagram of the PD with HgTe CQDs/Ag NPs and absorption spectra of HgTe CQD and HgTe CQD/Ag NP film.¹⁰¹ Reproduced from ref. 101 with permission from Elsevier, copyright 2023.

As early as 2014, Zhao et al. incorporated Au nanorods into a HgTe QD/ZnO heterojunction photodiode PD (Fig. 5b), and the results showed that the plasmonic structure improved the detectivity of the photodiode without sacrificing response time.²⁶ The work validated the feasibility of using plasmonic structures to enhance the performance of PDs. In 2018, He et al. added 0.5% to 1% (by weight) Ag NPs into PbS CQDs thin film, constructing a photoconductive PD with a PbS CQD/Ag NPs composite active layer (Fig. 5c).⁹⁸ The introduction of Ag NPs improved the photocurrent and suppressed the dark current of PD simultaneously, and significantly improved the detectivity of the device. In addition, they also found that even on flexible PDs, the presence of composite materials can still achieve a detectivity of up to 1.5 × 10¹⁰ Jones. In 2020, Hong et al. introduced two different types of plasma NPs, Au and Ag, into the top and bottom interfaces of the device (Fig. 5d). Ag NPs exhibited strong scattering, while Au NPs exhibited strong optical effects in the wavelength region with the strongest light absorption, enhancing the device's effective absorption of incident light.⁹⁹ In 2023, Guan et al. coupled Au nanospheres onto PbS CQDs (Fig. 5e), resulting in a plasmonic effect that increased carrier mobility. The device responsivity was three times higher than that of devices without Au NPs.¹⁰⁰ Recently, Chen et al. introduced an Ag NP layer below the HgTe QD layer to achieve surface plasmon resonances and enhance detector performance (Fig. 5f).¹⁰¹ The results showed that a 10 nm Ag NP layer could effectively increase the light-to-dark current ratio of the PD to 5.7 times. The PD could achieve a detectivity of 8.92 × 10¹⁰ Jones and responsivity of 2 A W⁻¹.

4.3 Propagating surface plasmon resonance structure

At the interface between the planer metal film and the dielectric layer, propagating surface plasmon polaritons (SPPs) are generated when free electrons undergo collective oscillations under the excitation of incident light.¹⁰² The propagating SPPs have greater momentum than the radiation modes in the dielectric, leading to evanescent decay on both sides of the interface. Therefore, periodic arrays are usually constructed on metal surfaces to scatter-out the extra momenta, thereby promoting the excitation of SPPs by incoming light or light radiation from SPPs modes (Fig. 6a).¹⁰³ The propagation SPP has the ability to concentrate the near-field light field and enhance far-field light scattering in the vicinity of metal structures, leading to enhanced light absorption of the QD layer during plasmonic resonance. To excite the propagating SPP, it is necessary to satisfy the momentum matching between the SPP and the incident light, as shown in the following equation:
where k_SPP is the wavevector of the SPPs, k₀ is the wavevector of light in free space, and ε_m and ε_d are the permittivity of metal and dielectric, respectively.⁸⁶ There have been many papers reporting on the applicability of plasma arrays in enhancing the performance of PDs.

Fig. 6 (a) Schematic diagram of SPPs principle. (b) Schematic diagram of the SPP structure with the metal hole array on the QD infrared detector.¹⁰⁴ Reproduced from ref. 104 with permission from American Chemical Society, copyright 2010. (c) Schematic diagram of a 2D plasma grating photodiode.¹⁰⁵ Reproduced from ref. 105 with permission from American Chemical Society, copyright 2014. (d) The SEM image of a BES PD.²⁹ Reproduced from ref. 29 with permission from Springer Nature, copyright 2015. (e) The false color SEM image of the filterless narrow-band infrared detectors consisting of transfer-patterned HgSe QD film, interdigitated electrodes, and the patterned plasmonic disk arrays.¹⁰⁶ Reproduced from ref. 106 with permission from the Royal Society of Chemistry, copyright 2017. (f) Illustration of the HgTe QD photodiode detector with interference cavity and plasmonic disk array.¹⁰⁷ Reproduced from ref. 107 with permission from American Chemical Society, copyright 2018. (g) The schematic of the on-chip detection with the plasmonic–silicon hybrid waveguide system with the inset of the cross-section of the metal–insulator–metal (MIM) waveguide with HgTe QD coating (left). The simulation schematic of the cross-section of the HgTe QD-loaded MIM waveguide and the electric field distribution of the MIM mode (right).¹⁰⁹ Reproduced from ref. 109 with permission from John Wiley and Sons, copyright 2019. (h) Schematic diagram of PbS CQD-based PD with microwheel array.¹¹⁰ Reproduced from ref. 110 with permission from American Chemical Society, copyright 2022.

In 2010, Chang et al. integrated InAs QD-based infrared PDs with a gold 2D hole array (2DHA) structure (Fig. 6b).¹⁰⁴ 2DHA was fabricated at the top of the device through a combination of a standard optical lithography and a metal lift-off process. The 2DHA structure not only supported plasmonic mode but also promoted optical coupling through scattering. The enhancement of infrared light response and detection ability during plasmonic resonance exceeding 100% have been successfully demonstrated. In 2014, Beck et al. designed a 2D plasma grating structure and integrated it within a photodiode to couple incident light to SPP modes propagating on the metal/semiconductor interface, demonstrating that the coupling of SPP modes could enhance the exciton peak absorption of CQD layers at ultra-thin thicknesses (Fig. 6c).¹⁰⁵ The research results provided a simple and effective technical means to improve PD performance. In 2015, Diedenhofen et al. used electron beam lithography technology to manufacture a plasma bull's eye structure (BES), which enabled the device to have color selection and significantly enhanced sensitivity.²⁹ BES is a concentric plasma grating with metal grooves, which can effectively couple incident light to SPP modes, guide incident photons towards the central aperture, and perform phase length interference. By placing PbS QDs as photonic material in the nanoscale central aperture of BES, a planar photoconductive material enhanced by a nanofocusing lens was formed (Fig. 6d). Only when the lattice vector of the plasma grating matches the propagating SPP mode will the incident photon and SPP undergo strong coupling, so the device had high sensitivity. In 2017, in order to realize narrow-band detection, Tang et al. first reported a plasmonic nanodisk array-enhanced filterless narrowband HgSe QD PD (Fig. 6e).¹⁰⁶ Researchers designed the structure of plasmonic nanodisks to form a second-order grating in the disk array, coupling the incident infrared to an in-plane collective photonic mode, thereby achieving a significant increase in responsivity at the center wavelength. Research has shown that after integrating plasma nanodisk arrays, the responsivity of the device increased from 28.76 mA W⁻¹ to 145 mA W⁻¹. In 2018, Tang et al. significantly optimized the performance of HgTe QD photodiodes by integrating the cavity structure and an array of plasmonic gold nanodisks simultaneously.¹⁰⁷ The interference structure at the top (5 nm Au + 20 nm ITO) could maximize the light reflection of light from the electrode to the QD layer, while the gold nanodisk array could enhance the device's absorption of light through plasmonic resonance (Fig. 6f). The PD could increase the detectivity by more than three times by integrating the cavity structure and a plasmonic structure compared with devices without optical structure.

In addition, SPPs can also be generated by introducing subwavelength structures at the metal/dielectric interface.^102,108 In 2019, Zhu et al. used a metal–insulator–metal (MIM) plasma waveguide structure to guide SPP waves to achieve a noise equivalent power (NEP) of 8.7 × 10⁻¹¹ W Hz^−1/2 for HgTe QD PDs in the SWIR band (Fig. 6g). It was an exemplary solution of achieving both plasma enhancement and chip-level integration simultaneously.¹⁰⁹ In 2022, Song et al. developed a PD with a 3D microwheel array based on PbS QD. This subwavelength structure introduced SPP, which limited the light field to a small region and concentrated the current density on each wheel (Fig. 6h), enabling the device to have high response and very low NEP. The responsivity could reach 4.67 A W⁻¹, which was 9 times higher than similar devices. The microwheel structure can further promote the development of PbS QD-based PDs in the field of communication devices, and provide new directions for other devices to improve their performance by utilizing plasmonic structures.¹¹⁰

4.4 Diffractive grating

Diffractive gratings are another optical structure commonly introduced into PDs to enhance light absorption. The gratings work by modulating the incident light's path through a diffraction process, splitting the incoming light into multiple diffraction orders. The relationship between the incident angle, diffraction angle, and the wavelength of light is governed by the grating equation:

mλ = d(sin θ_i + sin θ_m),

where m is the diffraction order, λ is the wavelength of the incident light, d is the grating period, θ_i is the angle of incidence of the light, and θ_m is the angle of the outgoing or diffracted light, corresponding to the m-th order diffraction.⁸⁷ This equation describes how the incident light is split into different diffraction orders, with each order having a distinct propagation direction based on the incident angle and wavelength. The waveguide modes are typically described by their propagation constant β, which is related to the wavevector in the medium and the effective refractive index n_eff:
where n_eff is the effective refractive index of the waveguide mode, and λ is the wavelength of the incident light.¹¹¹ When the wavevector of the diffracted light matches the propagation constant of the waveguide mode, efficient coupling occurs, confining the light within the active region and significantly increasing the light–matter interaction. By using a diffractive grating, the incident light undergoes multiple internal reflections or is trapped in the waveguide modes, enhancing light absorption across a wider spectral range without compromising carrier transport (Fig. 7a).¹⁰³

Fig. 7 (a) Schematic diagram of SPPs principle. (b) Schematic diagram of light absorption process of photodiode enhanced by metal diffractive grating. (c) Schematic structure of a PbS/ZnO heterojunction photodiode with Ag NPs plasmonic grating structure.¹¹² Reproduced from ref. 112 with permission from Optica Publishing Group, copyright 2016. (d) Schematic cross-section of the Au nanoelectrode used to induce GMR. (e) Schematic of the PDs structure deposited on silicon substrate with 1 μm period GMR. (f) Schematic cross-section of the optical path of PDs with GMR.¹¹³ Reproduced from ref. 113 with permission from American Chemical Society, copyright 2019.

In 2016, Beck et al. utilized metal diffractive gratings with the periodic structure of Ag NPs to enhance the light trapping of PbS CQD heterojunction photodiodes (Fig. 7b).¹¹² Enhanced light absorption involves three processes: light scattering into the diffusion order (DO), coupling to guided mode (GM), and absorption enhancement in the active layer (Fig. 7c). When the incident light reaches the Ag grating, due to the excitation of LSP, the incident light will be strongly scattered by the metal NPs, forming DO. According to the changes in the geometric shape and optical properties of the device, diffracted light propagates within the PbS CQD film in a set of discrete GM. When light of different diffraction orders is coupled into this waveguide mode, a guided resonance effect occurs, which enhances the light absorption of the device. By improving the coupling efficiency between DO and GM, the photocurrent measured at the exciton peak at a wavelength of approximately 1000 nm increased by 250%.

In 2019, Chu et al. proposed a cross-finger electrode structure and successfully induced guided-mode resonance (GMR) of SWIR PDs in PbS QD or HgTe QD layers (Fig. 7d and e).¹¹³ On the one hand, this interleaved electrode can be regarded as a diffractive grating, which promoted the coupling of diffraction light waves into the guided modes of the QD layer, and the patterned electrode induced GMR, where the diffracted light was confined in the active layer, resulting in the PbS QD film undergoing multiple light absorptions (Fig. 7f). On the other hand, the shorter electrode spacing between the cross-finger electrodes reduced the carrier transport time, thus achieving photoconductive gain. The research results indicated that compared with devices without GMR structure, the responsivity of PbS QD and HgTe QD photoconductors increased by 250 and 1000 times, respectively. The strategy of using GMR to enhance the light absorption of devices provided a feasible solution to the problem of low absorption efficiency in infrared sensors.

4.5 Summary

In summary, the integration of nanophotonic structures with CQDs has driven notable advancements in device performance. These nanophotonic structures include resonant cavities, localized surface plasmonic resonant structures, propagating surface plasmonic resonance structures, and diffractive gratings which significantly enhance light absorption and improve the interaction between light and the CQD active layer. As a result, there has been a marked improvement in responsivity, detectivity, and EQE across various CQD materials. The combination of these structures with CQD-based PDs has not only optimized light absorption but also enabled the development of high-performance devices with selective spectral detection. The advancements are particularly critical for applications in imaging, communication, and sensing, where high responsivity and detectivity are essential. For the convenience of our readers and to provide a comprehensive overview, Table 2 lists the key figures-of-merit for CQD-based PDs integrated with nanophotonic structures, as well as the degree of performance improvement of these PDs relative to their counterparts without nanophotonic structures.

Table 2 Figures-of-merit of CQD-based PDs based on nanophotonic structures

Materials	Nanophotonic structure	Wavelength	Absorption with nanophotonic structure (enhancement) compared with its planar counterpart	R/D* without nanophotonic structure	R/D* with nanophotonic structure (enhancement) compared with its planar counterpart	Ref.
PbS CQD	Resonant cavity (Folded-light-path)	AM 1.5	—	Photocurrent: 21 mA cm^−2	Photocurrent: 25 mA cm^−2 (22%)	89
PbS CQD	Resonant cavity (Fabry–Perot cavity)	1300 nm	60% (56%)	Photocurrent: 1.6 mA cm^−2	Photocurrent: 2.5 mA cm^−2 (56%)	85
InAs/GaAs CQD	Resonant cavity	7150 nm	—	0.038 A W^−1/3.77 × 10^9 Jones	0.067 A W^−1 (80%)/6.66 × 10^9 Jones (80%)	91
HgTe CQD	Resonant cavity (Fabry–Perot cavity)	2200 nm	≈30%	0.25 A W^−1	0.5 A W^−1 (100%)	92
PbS CQD	Resonant cavity	940 nm	15.7% (222.4%)	—	—	93
PbS CQD	Resonant cavity (Fabry–Perot cavity)	940 nm	70%	—	—	11
HgTe CQD	LSPR (Au nanorods)	1300 nm	—	Photocurrent: 0.37 mA cm^−2	Photocurrent: 1.27 mA cm^−2 (243%)	26
PbS CQD	LSPR (Ag nanocrystals)	900 nm	—	1.5 mA W^−1/2.1 × 10^10 Jones	3.8 mA W^−1 (153%)/7.1 × 10^10 Jones (238%)	98
PbS CQD	LSPR (Au/Ag nanoparticles)	700 nm	9.18% (25%)	Photocurrent: 23.67 mA cm^−2	Photocurrent: 26.16 mA cm^−2 (10.5%)	99
HgTe CQD	LSPR (Ag nanoparticles)	2300 nm	—	0.62 A W^−1/2.55 × 10^10 Jones	2 A W^−1 (210%)/8.92 × 10^10 Jones (250%)	101
InAs CQD	PSPR (Au 2D hole array)	9000 nm	7.7% (492%)	Not mentioned	8 × 10^10 Jones	104
PbS CQD	PSPR (SPP grating couplers)	1080 nm	45% (300%)	Photocurrent: 11 mA cm^−2	Photocurrent: 33 mA cm^−2 (200%)	105
PbS CQD	PSPR (bull's eye structure)	950 nm	Not mentioned (3000%)	4.9 × 10^12 Jones	1.7 × 10^13 Jones (247%)	29
HgSe CQD	PSPR (Au nanodisk arrays)	4200 nm	—	24.1 mA W^−1	148.7 mA W^−1 (517%)	106
HgTe CQD	PSPR (plasmonic disk arrays)	5000 nm	70% (500%)	0.42 A W^−1/1.2 × 10^11 Jones	1.62 A W^−1 (286%)/4 × 10^11 Jones (233%)	107
HgTe CQD	PSPR (silicon waveguide grating couplers)	2300 nm	—	Not mentioned	23 mA W^−1	109
PbS CQD/HgTe CQD	Diffractive grating (guided mode resonance)	1700 nm/2600 nm	70% (290%)	0.004 A W^−1	1 A W^−1 (24 900%)	113

---

5. Application of imaging chips based on SWIR CQD

For decades, researchers have been committed to high-quality imaging in the SWIR band. The PDs achieve imaging by detecting the SWIR light emitted or reflected by the observed target. Traditional narrow-bandgap semiconductor materials are crucial in SWIR PDs and imaging sensors. Nowadays, commercial SWIR technology mainly relies on InGaAs and HgCdTe as photosensitive materials. However, traditional narrow-bandgap semiconductor materials typically require expensive epitaxial equipment to grow on lattice-matched substrates and require complex flip chip bonding processes to connect to silicon readout integrated circuits (ROICs) (Fig. 8a). Only a single die can be processed at a time, resulting in low yield rate, extremely high cost, and small array size for imaging chips.^114,115 CQDs, as one of the new generation of infrared optoelectronic candidates, have great potential in replacing existing materials to achieve full band coverage of SWIR. The solution-processable CQDs have flexible maneuverability in thin-film deposition, enabling large-scale deposition of CQD-based PD arrays. A mature and low-cost deposition method can be utilized to create QDs imaging chips for imaging applications by directly integrating them into a CMOS ROIC (Fig. 8b). This silicon-based ROIC monolithic integration method avoids complex flip chip processes, further improving pixel resolution and reducing the fabrication cost of image sensors.^116,117

Fig. 8 (a) Flip-bonding-based fabrication process of imagers.¹²¹ Reproduced from ref. 121 with permission from American Chemical Society, copyright 2023. (b) Monolithic QDs for the infrared image sensor.¹²² Reproduced from ref. 122 with permission from MDPI, copyright 2017.

Nowadays, there are also reports of lead chalcogenides (PbSe, PbS, and PbTe) QDs and mercury chalcogenides (HgTe, HgS, and HgSe) QD PDs for SWIR imaging applications. PbX QDs exhibit excellent light absorption and photoelectric properties. Among them, PbS QDs have a large Bohr exciton of 18 nm and a high molar absorption coefficient of 1 × 10⁶ M⁻¹ cm⁻¹, making them the most widely investigated thin-film material for infrared sensing and imaging.¹¹⁸ HgX (X = S, Te, Se) QDs have the characteristics of fast response time, high absorption, and low dark current.^28,119 Compared with PbX QDs, HgX QDs have wide bandgap tunability from NIR to LIR.¹⁰⁷ In HgX, HgTe QD was the first material applied to infrared imaging sensor, with the best performance and shortest time response.¹²⁰ Nowadays, there are many schemes for preparing SWIR imaging sensors based on QD PDs. QD photoconductors, photodiodes, and phototransistors have all been investigated for SWIR imagers. In addition, new pixel-free upconversion devices also play an important part. This section will introduce the main achievements in the field of QD-based SWIR imaging.

5.1 QD imager with photoconductors

Due to its simpler planar structure, photoconductors can be used for plane bias application through specially designed readout circuits. This type of imager with photoconductor only requires one step of deposition manufacturing, which can achieve low-cost SWIR imaging. In 2022, Gréboval et al. proposed a strategy for designing infrared focal plane array (FPA) from a single fabrication step.¹²³ Researchers designed readout circuits with 640 × 512 pixels and a pixel pitch of 15 μm. An FPA chip structure based on HgTe QD was constructed by spin coating HgTe CQD thin film onto ROICs. The SWIR images captured by inserting the chip into the camera system exhibited significant contrast compared with visible images of the same scene (Fig. 9a and b). This study provided an effective solution for low-cost SWIR imaging sensors.

Fig. 9 (a) Scheme of a HgTe QD film deposited on the ROIC as the imaging array in a camera system. (b) Visible and SWIR pictures (taken by a HgTe QD-based FPA) of four vials containing different chemical solvents. An ITO-covered glass slide and a two-inch diameter silicon wafer are placed in front of the vials.¹²³ Reproduced from ref. 123 with permission from the Royal Society of Chemistry, copyright 2022. (c) Schematic diagram of 8 × 8 PD pixel array structure and ROIC structure. (d) Comparison of pixel array imaging image and ground truth image.¹²⁴ Reproduced from ref. 124 with permission from American Chemical Society, copyright 2023.

Eye-tracking devices can capture information and have broad application potential. The image sensor for eye-tracking in the SWIR band has important research value. In 2023, Mercier et al. first reported a semi-transparent image sensor suitable for eye-tracking applications.¹²⁴ The sensor was manufactured in the form of eyeglasses, consisting of 8 × 8 semi-transparent photoconductor arrays and electrodes, deposited on a fully transparent quartz substrate, using graphene pixel material sensitized with PbS CQD. Due to the high electron mobility of graphene, these pixels not only achieved optical transparency of 85–95%, but also greatly improved the responsivity of image sensors and their sensitivity under SWIR. In addition, researchers were committed to integrating a single ROIC into the eyeglass frame to achieve line-by-line readout of image information and avoid crosstalk between arrays (Fig. 9c). After projecting the PD pixel array, the image sensor demonstrated strong imaging quality compared with the ground truth (Fig. 9d). The eye-tracking image sensor with photoconductors fully demonstrated the new potential of virtual reality technology and autonomous driving.

5.2 QDs imager with photodiodes

CQD photodiodes are often used as sensing elements to achieve high-performance imaging in the SWIR range due to the advantages of CMOS compatibility, high sensitivity, fast response, high uniformity, and high EQE. As early as 2013, Heves et al. achieved the integration of a 3 × 2 pixel array PbS QD photodiodes on silicon substrates and ROICs by optimizing the PbS CQD thin-film and ligands exchange processes, for SWIR imaging (Fig. 10a).¹²⁵ Under 2 V reverse bias, the diode on the ROIC achieved a responsivity of up to 5.73 A W⁻¹ and a detectivity of 1.42 × 10¹² Jones. The results demonstrate the possibility of implementing a monolithic integrated SWIR FPA.

Fig. 10 (a) Schematic diagram of PbS photodiodes fabricated on the top of emulated ROICs.¹²⁵ Reproduced from ref. 125 with permission from IEEE, copyright 2013. (b) Schematic diagram of QD PD with stack structure integrated on top of ROIC. (c) Image of a Euro banknote acquired using a PbS QDs image sensor with a pixel pitch of 2.5 μm.¹²⁶ Reproduced from ref. 126 with permission from SPIE, copyright 2021. (d) Photographs of fake and real banknotes under SWIR using image sensors.³¹ Reproduced from ref. 31 with permission from IEEE, copyright 2020. (e) Imaging photographs of samples captured under SWIR with a PbS image sensor.¹²⁷ Reproduced from ref. 127 with permission from IEEE, copyright 2021. (f) Integral schematic of the PbS CQD imager. (g) Contrast schematic diagram of images obtained by smartphone silicon imager, InGaAs imager, and PbS CQD imager.³³ Reproduced from ref. 33 with permission from Springer Nature, copyright 2022. (h) Illustration of the structure of a dual-band CQD imaging device. (i) MWIR and SWIR images of a hand behind glass.¹²⁸ Reproduced from ref. 128 with permission from Springer Nature, copyright 2019.

In order to achieve the lowest possible dark current and high responsivity, a photodiode stack was developed. The device structure and manufacturing process of CQD photodiodes have been optimized to address the high dark current caused by the abundant trap states. In 2020, through a series of improvements and optimizations on photodiodes, Malinowski et al. integrated these stacked PbS QD photodiode pixels onto CMOS ROICs to obtain image sensors for SWIR imaging (Fig. 10b).¹²⁶ The pixel pitch of the device was determined by the critical dimensions of the ROIC, which provided the possibility of achieving CMOS image sensors with millions of pixel resolutions. When the pixel size of the array was reduced to 2.5 μm with a resolution of 1024 × 256, the image sensor obtained an image detail of a Euro banknote (Fig. 10c). In 2020, for the purpose of further reducing the pixel pitch of SWIR imagers, the group adopted a 3T pixel structure (composed of reset, source follower, and row selection transistors), and it was integrated into a monolithic ROIC.³¹ The horizontal–vertical symmetrical pixel architecture used in the 3T pixel structure resulted in a 24% reduction in pixel pitch for SWIR image sensors to 1.82 μm, setting a record for the smallest pitch in SWIR pixels at that time. Due to the difference in ink reflectivity under SWIR, the obtained image sensor could distinguish the fake banknotes from the real ones under SWIR (Fig. 10d). However, the 3T pixel structure had a higher negative photodiode bias during reset, resulting in a higher dark current, which resulted in an EQE of only 13% at 1400 nm for the device. In 2021, the group further optimized PbS QDs photodiodes to be compatible with 3T pixel design solutions, creating a superior SWIR image sensor with a pixel array with a 5 μm pixel pitch, and EQE was as high as 40% at 1450 nm.¹²⁷ The sensor could obtain images with high sensitivity and a high level of detail under SWIR (Fig. 10e). In 2022, Tang et al. reported a PbS CQD imager (Fig. 10f).³³ The photodiode achieved top illumination with high-quality junction by improving the top deposition method of transparent conductive oxides. At the same time, the circuit of each individual pixel of the imager had a buffered direct-injection structure, and the transistor switch was controlled by the readout timing diagram, which enabled high-quality imaging. After integrating the photodiodes with the COMS ROIC, an array of 640 × 512 pixels and a pixel size of 15 μm was formed for a high-quality imaging instrument. The detectivity could reach 2.1 × 10¹² Jones, and the EQE exceeded 60% at 940 nm. The obtained images of apples and hands showed that, compared with the silicon imager and InGaAs imager in smartphones, the PbS CQD imager could better display the detailed features of objects, reflecting its advantages in the future SWIR imaging field (Fig. 10g).

In addition, multi-band detectors can provide better object recognition by processing signals from different bands, playing an important role in autonomous driving assistance or industrial detection. In 2019, Tang et al. prepared two back-to-back stacked photodiodes using two different sizes of HgTe CQDs, which enabled the device to rapidly switch between SWIR mode to MWIR mode, achieving dual band imaging (Fig. 10h and i).¹²⁸ The device structure and the design of this multiband imaging provided possibilities for multifunctional scene applications.

5.3 QDs imager with phototransistors

Another class of PDs used for imaging is a phototransistor based on CQDs and graphene. It relies on a photogating effect and induced gain, which means that PDs achieve signal amplification through light and electric field regulation, leading to image sensors with high responsivity, which have great prospects in SWIR imaging. In 2017, Goossens et al. reported a high-resolution broadband image sensor with PbS-CQD/graphene integrated with CMOS electronic circuits.¹²⁹ A layer of graphene was grown on the CMOS using MOCVD. Each pixel of the 388 × 288 array was covered with a layer of graphene connected to the bottom ROIC, and a layer of PbS QDs was spin-coated on top of the graphene (Fig. 11a). After the graphene channels were sensitized with PbS CQDs, electron–hole pairs were generated, and under the effect of the built-in electric field, the electrons were trapped in CQD while the holes were transported into the graphene. These photogenerated carriers circulated in graphene by applying a bias voltage between two pixels contacts. Combined with the high electron mobility of graphene, the responsivity of this device was as high as 10⁷ A W⁻¹, and the detectivity exceeded 10¹² Jones. The obtained SWIR images also demonstrated high quality (Fig. 11b). This work validated new potential for monolithic chip integration of CQD and 2D materials for CMOS SWIR image sensors. In addition, there are also some works focusing on flexible imaging sensors. In 2022, Shultz et al. fabricated a flexible nine-channel PbS-QD/graphene nanohybrid broadband image sensor on a polyethylene terephthalate (PET) flexible substrate (Fig. 11c).¹³⁰ This pixel array was integrated into a low-cost Arduino ROIC to process signals through digital to analog conversion and transmit image data to a computer. The sensor exhibited high and uniform light response. At a bias voltage of 1 V, the maximum responsivity of the device could reach 9.56 × 10³ A W⁻¹. The most significant feature of this hybrid detector array was its flexibility and transferability, enabling it to seamlessly integrate into wearable flexible devices, fully demonstrating the potential development potential of this nanohybrid FPA in flexible devices.

Fig. 11 (a) Schematic diagram of the MOCVD graphene transfer process to a single chip. (b) Images of apples and pears obtained under SWIR light.¹²⁹ Reproduced from ref. 129 with permission from Springer Nature, copyright 2017. (c) Schematic diagram of PbS CQD and graphene hybrid detector array with built-in electric field.¹³⁰ Reproduced from ref. 130 with permission from American Chemical Society, copyright 2022. (d) 3D schematic of a PbS QD/IGZO hybrid phototransistor. (e) Original image and the image obtained by the SWIR imager. (f) Schematic illustration of the image scanning system.¹³¹ Reproduced from ref. 131 with permission from Springer Nature, copyright 2016. (g) 3D schematic cross-sectional view of the PbS/IGZO phototransistor. (h) Schematic of the imaging system consisting of a SWIR flat planar imager with LEDs and (i) the output SWIR duck image.¹³² Reproduced from ref. 132 with permission from American Chemical Society, copyright 2020.

In addition, indium–gallium–zinc–oxide (IGZO) thin-film transistor (TFT) is often used to construct active-matrix arrays in flat-panel imagers due to its high electron mobility and transparency. Each pixel is typically composed of a single phototransistor, which is connected to an external load resistor to create a photogating inverter. The inverter converts incident light signals into voltage signals, achieving high-sensitivity imaging. In 2016, Hwang et al. fabricated a PbS QD/IGZO hybrid phototransistor.¹³¹ PbS CQD was deposited on the top of the prefabricated IGZO TFT array on a glass substrate after ligand exchange (Fig. 11d). The responsivity of the phototransistor could exceed 10⁶ A W⁻¹ and detectivity exceeded 10¹³ Jones under 1.3 μm light irradiation. A photogating inverter formed with this hybrid phototransistor can be used as a SWIR flat panel single-pixel imager. The SWIR output image with the “KIST” logo demonstrated the practicality of the imaging device with TFT (Fig. 11e and f). In 2020, the same group further fabricated another 1 × 6 line scanner consisting of 6 photoconverter pixels (Fig. 11g).¹³² A fully patterned flexible array of PbS QDs was successfully achieved using photolithography stripping technology. Each photoconverter pixel was connected to a semiconductor parameter analyzer using a lead bonding technique to form a SWIR flat planar imager. A duck SWIR image was obtained under 1.3 μm illumination (Fig. 11h and i). This research can provide new ideas for further development of high-resolution large-area flat panel imaging devices.

5.4 Upconversion devices for pixel-free infrared imaging

SWIR to visible upconversion devices have attracted wide attention in recent years due to their compact size and easy imaging and visualization for large-scale processing. The upconversion devices, with a combination of light-emitting diode (LED) and QD PDs together, are a promising method widely used in the field of infrared imaging. The PD in the upconversion devices can capture infrared photons of incident light and convert it into electrical signals, which are then injected into the LED. The whole device can convert low-energy photons into high-energy photons, that is, infrared light into visible light. Therefore, the SWIR signal can be obtained by human eyes or ordinary commercial cameras for imaging, without a pixel connect process.^133–135 As early as 2011, Kim et al. first reported an upconversion device by using nanoscale PbSe nanocrystals as photosensitive layers combined with green phosphorescent organic light-emitting diodes (OLEDs) (Fig. 12a).¹³⁶ Sensitivity to SWIR light could be achieved through this structure, achieving a maximum photon-to-photon conversion efficiency at peak wavelength of 1.3 μm of 1.3%. The green emission of the device was demonstrated at 15 V (Fig. 12b). This work provides a low-cost and efficient design scheme for infrared-to-visible upconversion device, providing new possibilities for the application of QD-based upconversion devices.

Fig. 12 (a) Schematic diagram of PbSe QDs infrared-green upconversion device. (b) Images with and without NIR illumination in an upconversion device.¹³⁶ Reproduced from ref. 136 with permission from American Chemical Society, copyright 2011. (c) Schematic diagram of the layers in a near-infrared to visible upconversion device structure. (d) the on–off behavior of device without and with IR.¹³⁷ Reproduced from ref. 137 with permission from AIP, copyright 2019. (e) Schematic diagram of the upconversion device with CdSe/ZnS core/shell QDs. (f) Energy band diagrams of PDs containing Ag NPs in ZnO thin films, (g) photograph of the upconversion PD with an illumination of 940 nm and 1550 nm SWIR light.³ Reproduced from ref. 3 with permission from Springer Nature, copyright 2020. (h) Schematic diagram of SWIR imaging using an upconversion PD with a spatially defined shadow mask. (i) SWIR imaging images with ITO anodes and multilayer anodes, respectively.⁹³ Reproduced from ref. 93 with permission from John Wiley and Sons, copyright 2023.

In 2019, Zhang et al. fabricated a high-performance upconversion device by integrating a CdSe/ZnS QLED with PbS QD absorption layer.¹³⁷ Due to the much higher electron mobility than the hole mobility in PbS QDs, the device utilized photogenerated electrons from the active layer of PbS QDs to inject into the QLED to generate visible light, greatly improving the response speed of the device. Combined with the advantages of QLEDs with tunable narrow linewidth emission and high color saturation, the device had a maximum conversion efficiency of 3.35% at a peak wavelength of 970 nm and a switching ratio of 8 × 10³ (Fig. 12c and d). In 2020, Zhou et al. proposed a solution-processed infrared upconversion PD with a similar structure (Fig. 12e).³ The ZnO ETL in this device was doped with Ag NPs, enhancing the carrier tunneling effect and generating high photogenerated current under illumination to drive the LEDs (Fig. 12f). After optimization, the upconversion device achieved a photon-to-photon conversion efficiency of 5.4% and the detectivity could be as high as 6.4 × 10¹² Jones. The high imaging quality of a mouse breast cancer sample from the SWIR operating device fully demonstrated its potential for development in the field of SWIR bioimaging (Fig. 12g). In 2023, Yu et al. enhanced the upconversion device through the microcavity effect, resulting in a photon conversion efficiency of 15.7% when the upconversion device emitted from the top (Fig. 12h).⁹³ Meanwhile, the SWIR imaging quality of the device had also been improved. At 20 V, the upconverter with multilayer anode formed brighter and more saturated images of “KIST” and “KHU” characters, fully demonstrating the prospect of coupling the upconversion device with the optical structure (Fig. 12i).

Different from the upconversion photodiode, a high-gain infrared-to-visible upconversion light-emitting phototransistor (LEPT) was proposed by Yu et al.¹³⁸ It was combined with an infrared photoactive gate and integrated with an OLED (Fig. 13a). The underlying mechanism of the LEPT was achieved through gate bias and infrared illumination. Under zero gate bias and no infrared illumination, electron injection from the porous ITO source electrode to the C₆₀ channel layer was blocked by a large potential barrier at the interface. Under positive gate bias voltage, the injection of holes into the ITO gate was blocked, and electrons were attracted to the HfO₂/C₆₀ interface in the porous ITO region, causing the energy band bending. While under the positive gate bias of infrared illumination, the photogenerated electrons were transported to the ITO gate through the ZnO layer, and the photogenerated holes were collected at the PbS/HfO₂ interface, causing a strong field effect through accumulation. The modulated electrons were injected into the OLED through the C₆₀ channel layer from the ITO porous source electrode. At this time, the holes injected from the Al drain combine with the modulated electrons, causing the device to emit light and generate high gain. Therefore, uniform light emission could be observed over the active area of the device under infrared illumination during the operation of LEPT (Fig. 13b). The device had an EQE up to 1 × 10⁵% and detectivity can reach 1.2 × 10¹³ Jones.

Fig. 13 (a) Schematic diagram of the upconverted LEPT structure. (b) Photograph of the sample clamped in the measurement box under room light or with infrared illumination.¹³⁸ Reproduced from ref. 138 with permission from Springer Nature, copyright 2016.

6. Scalable processable method

The attainment of high-volume production capability is crucial for the commercialization of imaging chips based on CQD-based PDs. The development of the solution-phase ligand exchange method realizes the single-step fabrication of CQD films, providing more options for deposition. Moreover, solution-processed CQDs is compatible with various large-scale thin film deposition methods, such as blade coating, flexible microcomb printing (FMCP), inkjet printing and spray casting.^34,139 The blade coating method facilitates the fabrication of large-area QDs film on various rigid or flexible substrates. However, the shear rate of the CQD solution during the coating process is difficult to precisely control, which would lead to poor reproducibility of the blade coating.¹⁴⁰ In addition, the FMCP, which is recently developed, due to its unique microcomb structure can form a uniform thin film and ordered arrangement of QDs. The flexibility of the blade allows it to directly contact the substrate, which has great potential for printing on flexible substrates or curved surfaces.¹⁴¹ Furthermore, the inkjet printing method offers a different approach and allows for the monolithic integration of optoelectronic devices with various functions, as well as reduced material waste during the manufacturing process.¹⁴² However, due to the lack of colloidal stability of CQD inks and printing issues, this may lead to clogging of the nozzle or negatively affect the quality of the printed film, thereby reducing the photoelectric performance of the device.^143,144 Moreover, spray deposition is a simple and potentially large-scale deposition method, and the CQD PDs obtained through spray deposition exhibit better photoelectric performance and lower cost than spin-coated devices.¹⁴⁵ These methods not only improve the process, but also effectively supplement the application in some scenarios where spin-coating cannot be used, such as curved surface deposition. The integration of nanophotonic structures accompanying these deposition methods can further enhance the performance of CQD-based PDs by improving light absorption and guiding light through subwavelength optical structures. Different process methods each have advantages and limitations, and the decision on which method to choose for large-scale industrial production depends mainly on the different requirements of the specific purpose for production. Improved process methods developed in recent years that are suitable for large-scale industrial production are briefly examined in this section.

6.1 Blade coating for QDs PD fabrication

Blade coating is a scalable method that can be used for continuous and large-scale deposition of CQD films. In the process of blade coating, QD ink forms a thin film due to the shear force caused by the constant relative motion between the blade and the substrate (Fig. 14a). The thickness and morphology of the films can be controlled by parameters such as the speed of the blade, the distance between the blade and the substrate, and the deposition temperature.

Fig. 14 (a) Schematic diagram of the blade coating process. (b) Schematic illustration of the blade coating process and device structure.¹⁴⁶ Reproduced from ref. 146 with permission from the Royal Society of Chemistry, copyright 2018. (c) Schematic diagram of ink generated by liquid-phase ligand exchange for blade coating.¹⁴⁰ Reproduced from ref. 140 with permission from American Chemical Society, copyright 2018. (d) Schematic illustration of blade-coating of SWIR CQDs.¹⁴⁷ Reproduced from ref. 147 with permission from American Chemical Society, copyright 2020. (e) Schematic diagram of blade coating process using DFP-based CQD ink.³⁶ Reproduced from ref. 36 with permission from American Chemical Society, copyright 2021. (f) Schematic of high-quality PbS QD films obtained by utilizing leaf coating.¹⁴⁹ Reproduced from ref. 149 with permission from Springer Nature, copyright 2023.

In 2018, Aqoma and Jang firstly used QD ink based on liquid-phase ligand exchange for blade coating to fabricate a 450 nm-thick QD ink layer, resulting in device performance that was comparable to its spin-coated counterparts (Fig. 14b).¹⁴⁶ In the same year, Balazs et al. fabricated CQD film with smooth surfaces and low defect density in a single deposition step utilizing the blade coating method (Fig. 14c).¹⁴⁰ Sargent et al. prepared thick and smooth QD film utilizing blade coating, which showed unique advantages in preparing micrometer-sized films compared with the cracking caused by the spin-coating method (Fig. 14d).¹⁴⁷ In 2020, Sukharevska et al. found that PbS CQD ink maintained high colloidal stability in two different polar solvents, propylene carbonate and 2,6-difluoropyridine (DFP).¹⁴⁸ CQD can maintain colloidal stability for more than 20 months by utilizing the high dielectric constant and relatively low boiling point DFP. Subsequently, in 2021, this group successfully obtained high-quality CQD films with a thickness of 100 to 300 nm at relatively low processing temperatures by optimizing blade-related parameters using this superior ink (Fig. 14e).³⁶ Good device stability under illumination of solar cells made from DFP-based PbS QD inks as well as remarkable air stability was noted. After aging the device in air for 97 days, there was no change in the device parameters after keeping the devices under light soaking for more than 1 hour.

In addition, Tang et al. developed a PbS QD-based mixed solvent system compatible with the blade coating process in 2023.¹⁴⁹ Based on this ink system, researchers obtained a uniform PbS QD film with an area of 100 cm² and a thickness of approximately 350 nm through blade coating (Fig. 14f). At the same time, the above absorber-based device exhibited excellent storage stability. The PCE decreased from 11.13% to 10.64% after storage for 1000 hours under nitrogen atmosphere. This work can effectively reduce fabricating costs and improve device performance by improving the stability of QD ink and the ability to prepare large-area uniform films, strongly demonstrating the feasibility of commercialization of PbS CQD optoelectronic devices. Recently, to solve the problem of poor ink stability, Wang et al. used methylammonium lead iodide ligands to passivate PbS QDs with a size of up to 6.7 nm, allowing the ink to have a shelf-life of several months.¹⁵⁰ Finally, the SWIR photodiode prepared by blade coating with improved ink demonstrated an EQE of 76% at 1300 nm and 1.8 × 10¹² Jones specific detectivity. It can be foreseen that this large-sized stable QD ink could provide a foundation for the industrial development of SWIR PDs.

These works indicate that the blade coating method has great potential in preparing large-area CQD films with smooth surfaces and low defect density, and is more suitable for manufacturing large-area optoelectronic devices. In addition, it can be observed that integrating periodic nanophotonic structures within the coated film or substrate surface can enhance the light management properties of the film, optimizing the photon-to-charge conversion efficiency. The designed nanophotonic structures can act as optical traps, increasing the absorption of light by redirecting it within the active layer of CQD films, which is crucial for improving the device's overall sensitivity and response.

6.2 Flexible microcomb printing for QDs PD fabrication

Compared with blade coating, in 2022, our group proposed a new process solution for flexible microcomb printing (FMCP) (Fig. 15a).¹⁴¹ The blade of the FMCP method is made of flexible PET sheet which can be used for thousands of bending cycles with high durability. In addition, a computational fluid dynamics (CFD) model was created by constructing the blade into an ordered microcomb structure through laser marking micromachining. This facilitated mathematical calculations to analyze the relationship between the printed flow and the morphology of the PbS CQD film (Fig. 15b). In addition, microcomb-printed devices have better stability than spin-coated devices. The stabilities of the devices were measured for 2500 hours in an N₂ glovebox without encapsulation. The microcomb-printed devices showed slower degradation in comparison with the spin-coated devices. In 2023, Liu et al. printed high-quality top electrodes for organic photovoltaics using Field's metal by utilizing the FMCP process, which fully proved the practicality of FMCP in printing deposition (Fig. 15c).¹⁵¹ Recently, our group successfully prepared PbS QD photoconductors using the FMCP process, and the devices exhibited high responsibility of 2.1 A W⁻¹ (Fig. 15d).¹⁵² This fully demonstrated the prospects of FMCP in the future large-scale production of PDs.

Fig. 15 (a) Schematic diagram of FMCP device and SEM image of microcomb blades. (b) Schematic of the solution velocity field between the blade and the substrate simulated from CFD.¹⁴¹ Reproduced from ref. 141 with permission from American Chemical Society, copyright 2022. (c) Schematic diagram of manufacturing process using FM material and FMCP method.¹⁵¹ Reproduced from ref. 151 with permission from John Wiley and Sons, copyright 2023. (d) The fabrication process with the exploded view of the printing head (top) and the SEM image of the micro comb (bottom).¹⁵² Reproduced from ref. 152 with permission from John Wiley and Sons, copyright 2023.

6.3 Inkjet printing for QD PD fabrication

Inkjet printing is a non-contact, micrometer-level printing process that can directly spray nanoscale solutions onto different substrates. In inkjet printing, the shape of the piezoelectric transducer inside the ink chamber will change at the applied voltage. The ink inside the chamber is compressed to achieve the inkjet by generating pressure pulses in the ink chamber. The formation of ink droplets is influenced by surface tension, ink viscosity, and injection pressure. Inkjet printing can precisely control the volume and position of solution deposition and has advantages such as low material waste and ease of patterning. The limitations on quality of inkjet-printed CQDs films are mainly related to the property of the CQD ink. Firstly, the successful implementation of the inkjet printing process requires the use of highly stable CQD ink which maintains colloidal stability and excellent surface passivation. In addition, the deposition process must also be able to regulate the rheological properties associated with the ink to get the best inkjet performance.¹⁴² So far, there are some reports on optimizing QDs ink for inkjet printing systems, such as Yang et al.'s strategy of using a mixed solvent to spray small and uniform ink from the nozzle and provoke Marangoni flows that prevent coffee stain rings.¹⁵³ In addition, they also demonstrated that adding zinc oleate (Zn (OA)₂) in a liquid phase ligand exchange can effectively improve the problem of reduced device efficiency during the printing process.¹⁵³

In 2007, Heiss et al. studied the application of inkjet printing technology in the fabrication of nanocrystal PDs, particularly for photodetection in the infrared spectrum.¹⁴⁴ They utilized inkjet printing to fabricate HgTe QDs PDs, which demonstrated detectivity up to 3.2 × 10¹⁰ Jones and were capable of operating at wavelengths up to 3 μm (Fig. 16a). This early work demonstrated the effectiveness and repeatability of inkjet printing technology in fabricating high-performance PDs. In order to achieve high colloidal stability of ink, the composition of ink is crucial. In early 2019, YousefiAmin et al. developed an ink formula composed of dimethylformamide (DMF), N-methylformamide, and dimethyl sulfoxide in a 2 : 1 : 1 ratio,³⁵ adopting an automated form of full inkjet printing for QDs PDs (Fig. 16b). Also, the stability of the inkjet-printed photoconductors was comparably good in comparison with other PbS nanocrystal devices. The photocurrent only decreased by about 50% after being stored in the air for 3 months. However, it was challenging to expand the detector to an array because PbS CQD is prone to oxidation to PbSO₄, which can cause trap states and many cracks in the thin layer. After adding a small amount of polymer (polyvinylpyrrolidone-PVP) to the ink formula, the researchers observed the disappearance of cracking, a decrease in the trapping state of photoconductivity, and a further increase in colloidal stability. Using the improved CQD ink, a fully printed photoconducting device with a detectivity of 10¹² Jones through inkjet printing was obtained successfully (Fig. 16c). In September 2019, Sliz et al. determined an optimal ink composition composed of N-methyl-2-pyrrolidone (NMP) containing 1 wt% n-butylamine (BTA).¹⁴² Compared with inks using only NMP, inks with BTA exhibited better dispersion and stability. The obtained ink enhanced colloidal stability within the jetting window, maintained the passivation of the CQD surface, and was applied to highly sensitive large-area PDs that can operate in the SWIR and visible regions (Fig. 16d). The optimized ink formulation was used for inkjet printing and a photodiode with a detectivity of up to 1.4 × 10¹² Jones was fabricated (Fig. 16e).

Fig. 16 (a) Top view of the inkjet-printed HgTe nanocrystal PD.¹⁴⁴ Reproduced from ref. 144 with permission from John Wiley and Sons, copyright 2007. (b) Schematic diagram of full inkjet printing detector array. (c) Schematic of a large PbS photoconductive device with 25 cm² effective area.³⁵ Reproduced from ref. 35 with permission from American Chemical Society, copyright 2019. (d) High-stability CQD ink is achieved by using electrostatic stabilization mechanism (left) and steric stabilization mechanism (right). (e) Schematic diagram of PD structure obtained by inkjet printing.¹⁴² Reproduced from ref. 142 with permission from American Chemical Society, copyright 2019. (f) Demonstration of the hybrid-PbS QDs phototransistor on a curved surface. Thumbnail shows the device's schematic and optical microscope image, respectively.¹⁵⁴ Reproduced from ref. 154 with permission from John Wiley and Sons, copyright 2023.

In addition, inkjet printing technology has also been proved to be applicable for fabricating PDs on curved surfaces. In 2023, Kara et al. demonstrated the integration of graphene-PbS QD hybrid IR phototransistors on polymer optical fibers (POFs) using co-solvent ink (α-terpineol and hexane) (Fig. 16f).¹⁵⁴ This device serves as a functional coating that can detect infrared light propagating through POF without interrupting the waveguide. This work demonstrated the potential of inkjet printing technology in manufacturing curved integrated PDs. Moreover, those curved or flexible substrates with nanoimprinted nanophotonic structures cannot be fabricated using the spin-coating method. On the contrary, the inkjet printing technique can provide new opportunities for these situations. Through adoption of inkjet printing, conformal coating can be made on planar, curved, or flexible substrates, paving the way for smart textiles and other wearable devices.

6.4 Spray deposition for QD PD fabrication

Spray deposition is another droplet-based method which can evenly coat large areas of coatings on various substrates. In spray deposition, QD ink is usually atomized into fine particles through high-pressure air flow, and then sprayed onto the substrate at a high speed. The inks coalesce on the substrate according to the surface tension to form a continuous wet film. After evaporation and drying, a solid QD film will be formed.¹⁵⁵ During the spraying process, parameters such as surface tension of the ink, substrate properties, gas pressure, nozzle structure and size, and spraying distance can to some extent affect the formed thin film.

Initially, Chen et al. used spray coating to fabricate a SWIR photoconductor structure out of HgTe QDs (Fig. 17a).¹⁵⁶ The HgTe QD-based photoconductors exhibited good air stability. Even without encapsulation, these devices still exhibited stable performance after continuous testing for several hours under high illumination levels and under ambient conditions. In terms of longer-term stability, after exposing the device to the ambient environment for over a month, the photocurrent even slightly increased, while the 3 dB bandwidth decreased by about an order of magnitude. This method can be scaled up through low-cost and high-throughput manufacturing processes to produce large-area SWIR PD arrays. In order to further improve the performance of spraying devices, Sargent et al. developed a spraying technique for PbS CQD deposition and implemented a fully automated process with near monolayer control, which they named “sprayLD” (Fig. 17b).¹⁵⁷ This technique used a fine mist composed of droplets with a diameter of approximately 20 μm, which helped the solvent evaporate more quickly and uniformly. Compared with the spin-coating process, the sprayLD process improved the passivation and the way of packing CQD film, thereby eliminating electronic defects. The performance of prepared CQD film remained uniform over a large area of 60 cm². In addition, this group also researched the application of spray deposition in large-scale deposition.¹⁵⁸ They achieved device fabrication in a roll-to-roll environment, on flexible substrates and on curved surfaces by spraying CQD on different substrates, demonstrating the feasibility of spraying deposition for large-scale manufacturing of various unconventionally shaped devices.

Fig. 17 (a) Structure diagram of HgTe QD device structure after a few passes of spray-coating.¹⁵⁶ Reproduced from ref. 156 with permission from John Wiley and Sons, copyright 2013. (b) Schematic illustration of sprayLD process.¹⁵⁷ Reproduced from ref. 157 with permission from John Wiley and Sons, copyright 2014. (c) Schematic of one-step deposition of large-area CQD ink using spray deposition.³⁴ Reproduced from ref. 34 with permission from John Wiley and Sons, copyright 2019. (d) Spray deposition process. Electrodes (i–v) are fabricated on the substrate using photolithography; PbS CQD is deposited using spray deposition (vi); ligand exchange treatment of QD film (vii); final device obtained (viii). (d) Schematic structure of PbS QD-based PD obtained by utilizing spray deposition.¹⁴⁵ Reproduced from ref. 145 with permission from Elsevier, copyright 2020. (e) Device architecture of PbS CQD photovoltaic device having a spray coated PbS CQD film as an active layer.¹⁶¹ Reproduced from ref. 161 with permission from American Chemical Society, copyright 2021. (f) Schematic of spray-stencil lithography platform.¹⁶² Reproduced from ref. 162 with permission from John Wiley and Sons, copyright 2021.

However, during the drying process of CQD ink film formed by spray deposition, the solute of CQD droplets underwent spatial redistribution, resulting in unsatisfactory film morphology and affecting the performance of the device.^159,160 In 2019, Choi et al. optimized the solute redistribution mechanism of CQD droplets and efficiently deposited a 100 cm² PbS CQD film on ZnO/ITO substrates using an ultrasonic spraying system (Fig. 17c).³⁴ The root-mean-square roughness value of the film was 11.9 nm, indicating a highly uniform formation of CQD film. In addition, the thickness of the film reached 265 nm, which was much higher than the 150 nm thickness of the spin-coating method. This strategy was applicable to various CQD ink devices and could be used to prepare various large-scale CQD films. In 2020, Chen et al. utilized spray deposition to obtain high-quality PD, and the distance between adjacent QDs on the QDs thin film was 0.2 nm smaller than that of the spin-coating process.¹⁴⁵ The smaller the distance between QDs, the higher the dissociation rate of excitons; and the higher the carrier density, the higher the photocurrent in the solid device circuit. This provided stronger electronic coupling characteristics for the QD solid. Under the illumination power of 63.5 μW cm⁻², the detectivity of the device obtained by spray deposition was 1.4 × 10¹² Jones, and responsivity reached 365.1 A W⁻¹, both of which were higher than those obtained by the spin-coating process (Fig. 17d). In addition, the colloidal stability of QD ink can also affect the effect of spray deposition. In 2021, Yang et al. developed a mixed solvent system consisting of BTA and DMF, which can maintain static charge balance and prevent aggregation to obtain stable QD ink.¹⁶¹ This ink could be used to spray a low-defect QD film with a thickness of 350 nm on a large-sized substrate of 36 cm² (Fig. 17e). However, compared with the results regarding the spin-coated device, the initial PCE performance of the sprayed device rapidly decreased to 50% within 10 hours, possibly attributable to the relatively poor CQD packing in the matrix with respect to the spin-coated counterpart. Further developments are necessary to improve the stability of spray-coated CQD photovoltaics. There are also reports on the research of large-scale spray deposition CQD PD arrays. Zhang et al. realized the large-scale fabrication of a pixelated 10 × 10 HgTe CQD photoconductive array by using the spray-stencil lithography method (Fig. 17f).¹⁶² The properties of the prepared devices proved that spray deposition was an effective method to fabricate high-performance infrared PDs. The integration of spray depositions with stencil lithography facilitated the creation of pixelated arrays, as well as the coverage and patterning of CQDs on a single substrate.

Spray deposition makes it possible to fabricate PDs in large production quantities that are highly scalable, low cost and have competitive performance. By incorporating nanophotonic structures, such as diffractive gratings on the substrates or adding plasmonic nanoparticles into the sprayed film, the light absorption can be further optimized to boost the device's efficiency without increasing the material consumption or material cost significantly. Interestingly, these nanostructures can be tailored to specific wavelengths, thereby enhancing the detectivity of the CQD PDs for the designed wavelength, which can be applied to narrow bandwidth detection or multispectral sensing.

7. Prospects

The QD-based SWIR PDs have progressed tremendously during the past decade, owing to advances in synthesis process, ligand exchange methods, and device architecture engineering.

Since it was found that in the SWIR region QDs naturally suffer significant surface trap states, due to the large surface-to-volume ratio of the nanoparticles, any improvements in the surface formation and post-treatments of QDs will no doubt bring a fundamentally positive influence for the improvement of the device efficiency beyond the device structure optimizations. For instance, in QD fabrication, the dedicated nucleate and growth condition modulations for the nanocrystals in the reaction solvent are promising strategies to achieve high-quality QD materials with naturally fewer surface trap states.¹⁶³ Those surface-optimized QDs can be used in dark current-depressed photodiodes.¹⁶⁴ Moreover, since QDs are sensitive to ambient conditions, particularly during the ligand exchange process, cascade ligand exchange methods combined with multi-solvent engineering can help achieve high-quality QD ink, with both good dispersity and stability.^165,166 Notably, the QD stacking dynamics, in most cases, can be spontaneously optimized via the mentioned strategies based on modifying the QD surface and the solvent.^167,168 In addition, fully understanding the degradation mechanism of the QDs in the solids is still urgent for the development of reliable and usable QD-based PD and imager applications, which needs to draw more attention in the QD community.¹⁶⁹

Further improving the performance of PDs is of utmost importance. Attempts can be made to combine CQDs with more superior nanophotonic structures or 2D materials, not limited to microcavities, plasmonic and grating structures. Furthermore, 2D nanomaterials have excellent optical and physical properties which can effectively solve the problem of low carrier mobility in QDs when combined with QDs, thereby greatly improving device performance.^{129,170–172} Zhu et al. combined photosensitive PbS CQD and Mxene 2D materials to construct skin-like bilayer PD arrays on polyimide substrates, and finally fabricated flexible skin-like PDs.⁴ The development of mass-produced skin-like PDs of this kind could have a profound impact in the fields of bioimaging, machine vision, and artificial intelligence. In addition, one promising avenue for further exploration is the incorporation of halide perovskite-based QDs. This type of QD has a simple ABX₃ structure, exhibiting a unique combination advantage of high absorption coefficients, long carrier diffusion lengths, and small exciton binding energy, and has shown great promise when incorporated into PDs.^173–176 Recent advancements have integrated QDs with vertically aligned graphene arrays to engineer ambipolar, multifunctional PDs. This integration enhances light absorption, facilitates electron transport, and promotes efficient separation of photoinduced electron–hole pairs, yielding outstanding photocurrent responses with higher detectivity and responsivity at specific wavelengths.¹⁷⁷ Moreover, perovskite-based QDs have outstanding broadband photodetection properties that allow them to detect light in a broad range of wavelengths, from ultraviolet to visible and infrared. The integration of QDs with materials like MoS₂ has resulted in robust photocurrents and efficient PDs operating at various wavelengths.¹⁷⁸ Perovskite-based QDs are becoming highly preferred for advanced photodetection applications.^179,180 The most commonly used QDs in SWIR PDs contain toxic heavy metals (such as Pb and Hg), which greatly limit their practical applications. Therefore, it is of great research significance to search for alternative QD materials. Recently, Kim et al. successfully synthesized InAs QDs with uniform particle size distribution and excellent optical properties, solving long-standing obstacles in the research of InAs QDs due to synthesis limitations.⁶⁵ In addition, Sun et al. prepared a PD with an impressive EQE of over 30% at 920 nm by applying InBr treatment to the surface of InAs QDs.¹⁸¹ This fully demonstrates the potential of InAs QDs in heavy metal-free SWIR detection and imaging. The development of non-toxic QDs with high photoelectric performance is beneficial for the widespread application of devices in the market.

After arranging PDs with high photoelectric performance into an array, this is directly integrated with the ROIC to form an imaging chip, eliminating the complex flip-chip process, making it very promising in imaging applications. In terms of scenario applications, flexible semi-transparent image sensors can be applied to glass-like materials to achieve imaging functions in curved transparent screens or car windshields, expanding the application range of CQD imaging chips in unexplored fields.¹²⁴ In terms of functional exploration, adjusting the size of quantum dots allows researchers to control their properties, thereby capturing different parts of the infrared spectrum, which can be used to improve spectrometers and infrared cameras. Tang et al. achieved dual band imaging in SWIR and MWIR using two different sizes of CQDs.¹²⁸ This multi-band imaging device structure and design based on CQD size provides possibilities for multifunctional scene applications. In addition, the PD array can also form a four-quadrant detector, which can capture and track the beam of light by detecting the position distribution of the spot on the detector. The diversity and flexibility of QDs PD arrays play an increasingly important role in future technological development.

At present, some progress has been made in the large-scale preparation technology of QD-based SWIR PDs, such as blade coating, the FMCP process, inkjet printing and spray deposition. Blade coating deposits QD films on the substrate through the shear rate between the blade and the substrate, which has unique advantages in depositing uniform micrometer-level thin films. The recently reported FMCP process promotes high-quality thin films due to its regular micro-comb structure providing high shear rates, which has also shown extraordinary potential in large-scale deposition. Inkjet printing controls the generation of droplets through pressure pulses, which has a series of advantages such as efficient material utilization, non-contact deposition, and high scalability. The quality of the formed QD film depends on the ink formula due to the close correlation between the generation of droplets and the rheological properties of the ink. Spray deposition requires the ink to be sprayed into mist droplets at the nozzle. In addition, the sprayed mist droplets will be subjected to the dynamics of the airflow during flight and deposit extensively on the substrate. Some properties of PDs based on spray deposition are no less than those of spin-coating. Different deposition methods have their own advantages and disadvantages. Further fluid dynamics analysis of deposition methods can help better understand and control the deposition process, laying the foundation for the commercial application of QDs. It is believed that in the future, flexible, fast, low-cost, and widely used CQD PDs will continue to enrich all aspects of people's lives (Table 3).

Table 3 Progress in QD-based SWIR PDs

Year	Photoactive material	Ligands	Device type	Exciton peak [nm]	Spectral range [nm]	Responsivity [A W^−1 ]	Detectivity [Jones]	Rise and decay time or f−3 dB	Ref.
2012	PbS CQD:Ag MNP	EDT	Photoconductor	950	400–1200	—	—	—	182
2013	HgTe CQD	OA	Photoconductor	1300	SWIR	—	>10^10	1 MHz	156
2014	PbS CQD:Au NPs	EDT	Photoconductor	1084	350–1000	1.6 × 10^−3	1.1 × 10^10	1.02 kHZ	183
2014	HgTe CQD:Au Nanorod	OA	Photodiode	1300	NIR–MWIR	—	—	—	26
2015	PbS CQD/MoS2	EDT	Phototransistor	1380	400–1500	6 × 105	7 × 10^14	0.3–0.4s	48
2015	PbS CQD/MoO3	TABI	Phototransistor	950	400–1100	4	2 × 10^10	10/12 μs	184
2015	PbS CQD:P3HT	OA	Photoconductor	1150	UV–Vis–NIR	1 × 1012@600 nm	2.1 × 10^12	0.16 s/0.11 s	45
2015	PbS CQD:Ag NPs	BDT	Photoconductor	1050	400–1200	110	5 × 10^11	200 Hz	185
2015	PbS CQD/c-Si	TBAI	Photodiode	1230	400–1300	0.4	1.5 × 10^11 @ 1230 nm	—	186
2016	PbS CQD	BDT	Photodiode + OLED	1040	400–1400	—	1.23 × 10^13	—	138
2016	PbS CQD/SnS2 nanosheet	EDT	Phototransistor	939	300–1000	1 × 10^5	2.4 × 10^11	—	187
2016	PbS CQD/graphene	EDT	Photodiode + phototransistor	1600	600–1800	4 × 10^7	1 × 10^13	1.5 kHZ	188
2017	PbS CQD	EDT	Photodiode	950	400–1200	—	—	—	189
2017	PbS CQD/graphene	EDT	Phototransistor	1050	500–1200	10	2 × 10^11	—	171
2017	PbS CQD/graphene	EDT	Phototransistor + ROIC	1670	300–2000	1 × 10^7	1 × 10^12	—	129
2017	PbS CQD/Si	TBAI	Phototransistor	1300	400–1600	1 × 10^4@1300 nm	1.8 × 10^12	10 μs	190
2018	PbS CQD	BDT	Photodiode	1440	Vis–SWIR	—	1 × 10^12	—	191
2018	HgTe CQD	EDT	Photodiode	2500	SWIR	—	3 × 10^8	>10 kHz	192
2018	PbS CQD/graphene	—	Phototransistor	1550	Vis–NIR	1 × 10^4@1550 nm	1 × 10^12@1550 nm	3 ms	170
2018	PbS CQD	TBAI	Photoconductor	1550	Vis–NIR	5.15@1550 nm	1.96 × 10^10@1550 nm	—	193
2019	HgTe CQD	OA	Photodiode	—	SWIR–MWIR	—	>10^10	<2.5 μs	128
2019	InAs/GaAs	—	Photodiode	1206	SWIR–MWIR	0.067	6.66 × 10^9	—	91
2019	HgTe CQD	OA	Photoconductor	2200	SWIR	0.5	7.5 × 1010@1550 nm	260 ns	92
2019	PbS CQD:GMR	EDT	Phototransistor	1550	Vis–SWIR	1	1 × 10^9	1 kHZ	113
2019	PbS CQD:PVP	BiI3	Photoconductor	950	NIR	1.5	6 × 10^11	>3 kHz	35
2019	HgTe CQD:MIM	FMT	Photoconductor	2300	SWIR–MWIR	23	—	10 kHz	109
2019	PbS CQD/CH3NH3PbI3	SCN^−	Phototransistor	1400	300–1500	255@365 nm	4.9 × 10^13@365 nm	42 ms	194
2019	PbS CQD/WS2	EDT	Phototransistor	1800	800–2200	1400	1 × 10^12	0.03s	195
2019	PbS CQD	TBAI/BDT	Photodiode	2100	400–2600	0.385@2100 nm	1.5 × 10^11@2100 nm	43/70 μs	118
2019	HgTe CQD:GMR	EDT	Photoconductor	2600	SWIR	1	—	1 kHz	113
2020	PbS CQD/n-Si	EDT	Photodiode	1540	400–1700	0.26@1540 nm	1.47 × 10^11@1540 nm	2.04/5.34 μs	196
2020	PbS CQD	TBAI/EDT	Photodiode + LED	1500	400–1600	20	6.4 × 10^12	—	3
2020	HgTe CQD	OA	Photodiode	2400	Vis–SWIR	0.9	5 × 10^9	13 ns	197
2020	PbS CQD	TBAI	Photoconductor	1250	800–1400	1.4 × 10^12	365.1	—	145
2020	PbS CQD	TBAI/EDT	Phototransistor	1300	700–1400	1 × 10^3–1 × 10^4 @1310 nm	—	<0.5 s	132
2020	PbS CQD	Thiols	Photodiode	940	300–1100		10^12@940 nm	57/86 μs	29
2020	PbS CQD	—	Photoconductor	1400	400–1600	774	—	—	143
2021	PbS CQD	Halide/EDT	Photodiode	1000	600–1200	—	6.7 × 10^12@980 nm	—	198
2021	HgTe CQD	EDT/HCl/IPA	Phototransistor	—	SWIR–MWIR	—	>10^11	6.4 μs	162
2021	PbS CQD	TBAI	Photoconductor	1400	900–1600	1.895 × 10^3	1.51 × 10^12	—	199
2021	PbS CQD/PbSe CQD	TBAI	Phototransistor	1550	980–1550	613@1550 nm	4.8 × 10^11@1550 nm	37.9/92.3 μs	200
2022	PbS:PDDTT	EDT	Photodiode	1850	900–2100	4.3 × 10^13@1550 nm	0.195@1550 nm	11.5/45.5 μs	201
2022	PbS/graphene	TBAI/PBA	Phototransistor	800	600–1000	3 × 10^7		101.64/2.11 s	172
2022	PbS CQD	PbI2/MPA	Photodiode	1550	1000–1700	77@1550 nm	1.5 × 10^11@1550 nm	14/20 μs	202
2022	In(As,P) CQD	DMF/n-BuNH2/MPD	Photodiode	1400	1400–1500	—	10^9	0.6–1.6 μs	203
2022	PbS CQD	EDT	Photodiode + ROIC	970	400–1300		2.1 × 10^12	140 kHz	33
2023	PbS CQD	PbI2	Photoconductor	1300	—	2.1	4.74 × 10^9	209.2 ms/105.7 ms	152
2023	HgTe CQD:Ag NPs	OA	Photoconductor	—	SWIR	2	8.92 × 10^10	840 ns	101
2023	InSb CQD	InBr3	Photodiode	1000–1500	SWIR	0.098	—	550/800 ms	23
2024	HgTe CQD	HgCl2/β-ME/BTA/DMF	Photodiode	1700	1300–5000	—	8.1 × 10^11	3/16 μs	204
2024	InSb/InP	InI3	Photodiode	1240	900–1750	—	4.4 × 10^11	70 ns	205

---

8. Conclusion

In recent years, with the continuous improvement and development of nanomaterial technology, solution-treated CQDs have been proved to be a representative photoelectric material that can be applied in the field of SWIR detection. CQD-based PD arrays can be compatible with CMOS substrates, with convenient manufacturing processes, low costs, and high photoelectric quality factors, opening up new avenues for future high-resolution SWIR detection and imaging. A detailed review of recent advances in CQD-based SWIR PDs has been presented. Firstly, the basic structures and working principles of different types of PDs have been introduced. Then, the latest progress in the synthesis of QDs and ligand exchange methods was introduced. To address the problem of light absorption and carrier extraction, optical strategies to improve device performance have been discussed for four different nanophotonic structures (resonant cavity, localized surface plasmon resonance structure, propagating surface plasmon resonance structure and diffractive grating). These four different nanophotonic structures each have unique advantages, and through in-depth research and rational design, the performance of devices can be significantly improved. At the same time, the latest progress in the application of SWIR imaging chips based on CQD PDs has been introduced in detail. Three types of PDs, namely photoconductor, photodiode, and phototransistor, have made significant progress in the application of SWIR imaging chips. The development of these PDs not only improves the performance of SWIR imaging systems, but also expands their application scope in various fields. In addition, the development of up-conversion devices provides an alternative method for infrared imaging that does not require pixel connection. To promote the commercialization of CQD PDs, the advances in large-scale production processes based on CQD PDs are focused on four aspects, namely blade coating, FMCP, inkjet printing and spraying deposition. By conducting an in-depth analysis of the fluid dynamics and the advantages and disadvantages of each process, the understanding of large-scale production techniques can be accelerated. The continuous development and improvement of these technologies will provide a feasible approach for the industrial production of CQD PDs, laying the foundation for market promotion and application. Finally, the future development direction and opportunities of CQD-based SWIR PDs have been discussed. It is believed that in the future, CQD-based PDs will utilize flexible fabrication methods to further achieve simpler preparation processes and novel functional extensions, to further meet the growing demand for optical detection today.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Shenzhen Stable Support Research Foundation (No. 20220717215521001); Natural Science Foundation of Top Talent of Shenzhen Technology University (No. GDRC202110, GDRC202340); Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515011071, 2021A1515110535); Shenzhen Basic Research General Program (No. JCYJ20190809152411655); The Education Department of Guangdong Province (No. 2021KCXTD045); National Natural Science Foundation of China (No. 12204318); Shenzhen Science and Technology Program (No. RCYX20221008092908030); Shenzhen Key Laboratory of Applied Technologies of Super-Diamond and Functional Crystals (ZDSYS20230626091303007).

References

1. T. Lin and J. Wang, Adv. Mater., 2019, 31, 1901473.
2. R. Saran and R. J. Curry, Nat. Photonics, 2016, 10, 81–92.
3. W. Zhou, Y. Shang, F. P. García De Arquer, K. Xu, R. Wang, S. Luo, X. Xiao, X. Zhou, R. Huang, E. H. Sargent and Z. Ning, Nat. Electron., 2020, 3, 251–258.
4. Y. Zhu, C. Geng, L. Hu, L. Liu, Y. Zhu, Y. Yao, C. Li, Y. Ma, G. Li and Y. Chen, Chem. Mater., 2023, 35, 2114–2124.
5. S. Yakunin, M. Sytnyk, D. Kriegner, S. Shrestha, M. Richter, G. J. Matt, H. Azimi, C. J. Brabec, J. Stangl, M. V. Kovalenko and W. Heiss, Nat. Photonics, 2015, 9, 444–449.
6. Y. Li, Z. Shi, W. Liang, L. Wang, S. Li, F. Zhang, Z. Ma, Y. Wang, Y. Tian, D. Wu, X. Li, Y. Zhang, C. Shan and X. Fang, Mater. Horiz., 2020, 7, 530–540.
7. J. Huang, J. Lee, J. Vollbrecht, V. V. Brus, A. L. Dixon, D. X. Cao, Z. Zhu, Z. Du, H. Wang, K. Cho, G. C. Bazan and T. Nguyen, Adv. Mater., 2020, 32, 1906027.
8. T. Zhang, C. Ling, X. Wang, B. Feng, M. Cao, X. Xue, Q. Xue, J. Zhang, L. Zhu, C. Wang, H. Lu and W. Liu, Adv. Mater. Technol., 2022, 7, 2200250.
9. W. Cheng, W. Tian, F. Cao and L. Li, InfoMat, 2022, 4, e12348.
10. N. Huo, S. Gupta and G. Konstantatos, Adv. Mater., 2017, 29, 1606576.
11. V. Pejović, E. Georgitzikis, I. Lieberman, P. E. Malinowski, P. Heremans and D. Cheyns, Adv. Funct. Mater., 2022, 32, 2201424.
12. Q. Xu, J. Hu and X. Wang, J. Opt. Soc. Am. B, 2021, 38, 194.
13. Z. Wu, Y. Ou, M. Cai, Y. Wang, R. Tang and Y. Xia, Adv. Opt. Mater., 2023, 11, 2201577.
14. W. Gong, P. Wang, D. Dai, Z. Liu, L. Zheng and Y. Zhang, J. Mater. Chem. C, 2021, 9, 2994–3025.
15. A. Shrestha, M. Batmunkh, A. Tricoli, S. Z. Qiao and S. Dai, Angew. Chem., Int. Ed., 2019, 58, 5202–5224.
16. C. Gréboval, A. Chu, N. Goubet, C. Livache, S. Ithurria and E. Lhuillier, Chem. Rev., 2021, 121, 3627–3700.
17. V. Grigel, D. Dupont, K. De Nolf, Z. Hens and M. D. Tessier, J. Am. Chem. Soc., 2016, 138, 13485–13488.
18. W. Liu, A. Y. Chang, R. D. Schaller and D. V. Talapin, J. Am. Chem. Soc., 2012, 134, 20258–20261.
19. M. Vafaie, J. Z. Fan, A. Morteza Najarian, O. Ouellette, L. K. Sagar, K. Bertens, B. Sun, F. P. García De Arquer and E. H. Sargent, Matter, 2021, 4, 1042–1053.
20. T. Zhu, L. Zheng, X. Yao, L. Liu, F. Huang, Y. Cao and X. Gong, ACS Appl. Mater. Interfaces, 2019, 11, 9205–9212.
21. M. M. Ackerman, M. Chen and P. Guyot-Sionnest, Appl. Phys. Lett., 2020, 116, 083502.
22. H. Seo, H. J. Eun, A. Y. Lee, H. K. Lee, J. H. Kim and S. Kim, Adv. Sci., 2024, 11, 2306439.
23. S. Chatterjee, K. Nemoto, B. Ghosh, H.-T. Sun and N. Shirahata, ACS Appl. Nano Mater., 2023, 6, 15540–15550.
24. X. Yin, C. Zhang, Y. Guo, Y. Yang, Y. Xing and W. Que, J. Mater. Chem. C, 2021, 9, 417–438.
25. H. Liu, B. Zhang, Z. Hu, Q. Yan, J. Liu and J. Tang, Chin. Sci. Bull., 2021, 66, 4664–4676.
26. M. Chen, L. Shao, S. V. Kershaw, H. Yu, J. Wang, A. L. Rogach and N. Zhao, ACS Nano, 2014, 8, 8208–8216.
27. S. Baek, J. H. Song, W. Choi, H. Song, S. Jeong and J. Lee, Adv. Mater., 2015, 27, 8102–8108.
28. X. Tang, G. F. Wu and K. W. C. Lai, J. Mater. Chem. C, 2017, 5, 362–369.
29. S. L. Diedenhofen, D. Kufer, T. Lasanta and G. Konstantatos, Light:Sci. Appl., 2015, 4, e234–e234.
30. E. Georgitzikis, P. E. Malinowski, Y. Li, J. Maes, L. M. Hagelsieb, S. Guerrieri, Z. Hens, P. Heremans and D. Cheyns, IEEE Sens. J., 2020, 20, 6841–6848.
31. J. Lee, E. Georgitzikis, Y. Li, Z. Lin, J. Park, I. Lieberman, D. Cheyns, M. Jayapala, A. Lambrechts, S. Thijs, R. Stahl and P. E. Malinowski, in 2020 IEEE International Electron Devices Meeting (IEDM), IEEE, San Francisco, CA, USA, 2020, p. 16.5.1–16.5.4.
32. L. Gao, D. Dong, J. He, K. Qiao, F. Cao, M. Li, H. Liu, Y. Cheng, J. Tang and H. Song, Appl. Phys. Lett., 2014, 105, 153702.
33. J. Liu, P. Liu, D. Chen, T. Shi, X. Qu, L. Chen, T. Wu, J. Ke, K. Xiong, M. Li, H. Song, W. Wei, J. Cao, J. Zhang, L. Gao and J. Tang, Nat. Electron., 2022, 5, 443–451.
34. M. Choi, Y. Kim, H. Lim, E. Alarousu, A. Adhikari, B. S. Shaheen, Y. H. Kim, O. F. Mohammed, E. H. Sargent, J. Y. Kim and Y. S. Jung, Adv. Mater., 2019, 31, 1805886.
35. A. YousefiAmin, N. A. Killilea, M. Sytnyk, P. Maisch, K. C. Tam, H.-J. Egelhaaf, S. Langner, T. Stubhan, C. J. Brabec, T. Rejek, M. Halik, K. Poulsen, J. Niehaus, A. Köck and W. Heiss, ACS Nano, 2019, 13, 2389–2397.
36. N. Sukharevska, D. Bederak, V. M. Goossens, J. Momand, H. Duim, D. N. Dirin, M. V. Kovalenko, B. J. Kooi and M. A. Loi, ACS Appl. Mater. Interfaces, 2021, 13, 5195–5207.
37. K. Xu, W. Zhou and Z. Ning, Small, 2020, 16, 2003397.
38. H. Ren, J. Chen, Y. Li and J. Tang, Adv. Sci., 2021, 8, 2002418.
39. Z. Ren, J. Sun, H. Li, P. Mao, Y. Wei, X. Zhong, J. Hu, S. Yang and J. Wang, Adv. Mater., 2017, 29, 1702055.
40. W. Gong, P. Wang, W. Deng, X. Zhang, B. An, J. Li, Z. Sun, D. Dai, Z. Liu, J. Li and Y. Zhang, ACS Appl. Mater. Interfaces, 2022, 14, 25812–25823.
41. P. Nagpal and V. I. Klimov, Nat. Commun., 2011, 2, 486.
42. H. Zhang, Y. Zhang, X. Song, Y. Yu, M. Cao, Y. Che, J. Wang, J. Yang, H. Dai, G. Zhang and J. Yao, Nanotechnology, 2016, 27, 425204.
43. R. Dong, Y. Fang, J. Chae, J. Dai, Z. Xiao, Q. Dong, Y. Yuan, A. Centrone, X. C. Zeng and J. Huang, Adv. Mater., 2015, 27, 1912–1918.
44. M. He, X. Pang, X. Liu, B. Jiang, Y. He, H. Snaith and Z. Lin, Angew. Chem., Int. Ed., 2016, 55, 4280–4284.
45. J. Yoo, S. Jeong, S. Kim and J. H. Je, Adv. Mater., 2015, 27, 1712–1717.
46. J. W. Lee, D. Y. Kim and F. So, Adv. Funct. Mater., 2015, 25, 1233–1238.
47. J. R. Manders, T. Lai, Y. An, W. Xu, J. Lee, D. Y. Kim, G. Bosman and F. So, Adv. Funct. Mater., 2014, 24, 7205–7210.
48. D. Kufer, I. Nikitskiy, T. Lasanta, G. Navickaite, F. H. L. Koppens and G. Konstantatos, Adv. Mater., 2015, 27, 176–180.
49. Q. Liu, H. Tian, J. Li, A. Hu, X. He, M. Sui and X. Guo, Adv. Opt. Mater., 2019, 7, 1900455.
50. H. Liu, J. S. Owen and A. P. Alivisatos, J. Am. Chem. Soc., 2007, 129, 305–312.
51. T. Sugimoto, Colloids Surf., A, 2000, 164, 205–215.
52. J. S. Owen, E. M. Chan, H. Liu and A. P. Alivisatos, J. Am. Chem. Soc., 2010, 132, 18206–18213.
53. M. Paul, J. Kettler, K. Zeuner, C. Clausen, M. Jetter and P. Michler, Appl. Phys. Lett., 2015, 106, 122105.
54. X. Dai, Y. Deng, X. Peng and Y. Jin, Adv. Mater., 2017, 29, 1607022.
55. J. Owen and L. Brus, J. Am. Chem. Soc., 2017, 139, 10939–10943.
56. A. Surrente, R. Carron, P. Gallo, A. Rudra, B. Dwir and E. Kapon, Nano Res., 2016, 9, 3279–3290.
57. Y. Pu, F. Cai, D. Wang, J.-X. Wang and J.-F. Chen, Ind. Eng. Chem. Res., 2018, 57, 1790–1802.
58. F. O. Silva, L. C. D. S. Viol, D. L. Ferreira, J. L. A. Alves and M. A. Schiavon, Quim. Nova, 2010, 33, 1933–1939.
59. Y. Zhao, J. Li, Y. Dong and J. Song, Isr. J. Chem., 2019, 59, 649–660.
60. D. V. Talapin, J.-S. Lee, M. V. Kovalenko and E. V. Shevchenko, Chem. Rev., 2010, 110, 389–458.
61. C. B. Murray, D. J. Norris and M. G. Bawendi, J. Am. Chem. Soc., 1993, 115, 8706–8715.
62. B. J. Schwartz, E. R. Bittner, O. V. Prezhdo and P. J. Rossky, J. Chem. Phys., 1996, 104, 5942–5955.
63. C. B. Murray, S. Sun, W. Gaschler, H. Doyle, T. A. Betley and C. R. Kagan, IBM J. Res. Dev., 2001, 45, 47–56.
64. M. A. Hines and G. D. Scholes, Adv. Mater., 2003, 15, 1844–1849.
65. T. Kim, S. Park and S. Jeong, Nat. Commun., 2021, 12, 3013.
66. S. Mussa Farkhani and A. Valizadeh, IET Nanobiotechnol., 2014, 8, 59–76.
67. C. R. McCormick, R. R. Katzbaer, B. C. Steimle and R. E. Schaak, Nat. Synth., 2023, 2, 152–161.
68. X. Xing, Q. Zhang, Z. Huang, Z. Lu, J. Zhang, H. Li, H. Zeng and T. Zhai, Small, 2016, 12, 874–881.
69. P. Dagtepe and V. Chikan, J. Phys. Chem. C, 2010, 114, 16263–16269.
70. C. Dong, S. Liu, N. Barange, J. Lee, T. Pardue, X. Yi, S. Yin and F. So, ACS Appl. Mater. Interfaces, 2019, 11, 44451–44457.
71. Y. Wang, Z. Liu, N. Huo, F. Li, M. Gu, X. Ling, Y. Zhang, K. Lu, L. Han, H. Fang, A. G. Shulga, Y. Xue, S. Zhou, F. Yang, X. Tang, J. Zheng, M. Antonietta Loi, G. Konstantatos and W. Ma, Nat. Commun., 2019, 10, 5136.
72. D. Vasudevan, R. R. Gaddam, A. Trinchi and I. Cole, J. Alloys Compd., 2015, 636, 395–404.
73. J.-B. Kwon, S.-W. Kim, B.-H. Kang, S.-H. Yeom, W.-H. Lee, D.-H. Kwon, J.-S. Lee and S.-W. Kang, Nano Convergence, 2020, 7, 28.
74. J. Kim, J. Roh, M. Park and C. Lee, Adv. Mater., 2023, 2212220.
75. M. Alizadeh-Ghodsi, M. Pourhassan-Moghaddam, A. Zavari-Nematabad, B. Walker, N. Annabi and A. Akbarzadeh, Part. Part. Syst. Charact., 2019, 36, 1800302.
76. C. Zhang, Y. Xia, Z. Zhang, Z. Huang, L. Lian, X. Miao, D. Zhang, M. C. Beard and J. Zhang, Chem. Mater., 2017, 29, 3615–3622.
77. W. S. Lee, Y. G. Kang, H. K. Woo, J. Ahn, H. Kim, D. Kim, S. Jeon, M. J. Han, J.-H. Choi and S. J. Oh, Chem. Mater., 2019, 31, 9389–9399.
78. A. Fischer, L. Rollny, J. Pan, G. H. Carey, S. M. Thon, S. Hoogland, O. Voznyy, D. Zhitomirsky, J. Y. Kim, O. M. Bakr and E. H. Sargent, Adv. Mater., 2013, 25, 5742–5749.
79. M. Yuan, K. W. Kemp, S. M. Thon, J. Y. Kim, K. W. Chou, A. Amassian and E. H. Sargent, Adv. Mater., 2014, 26, 3513–3519.
80. G. Konstantatos, I. Howard, A. Fischer, S. Hoogland, J. Clifford, E. Klem, L. Levina and E. H. Sargent, Nature, 2006, 442, 180–183.
81. J. Shi, P. Zhao and X. Wang, Adv. Mater., 2013, 25, 916–921.
82. B. K. Jung, W. Kim and S. J. Oh, J. Korean Ceram. Soc., 2021, 58, 521–529.
83. D. Bozyigit, M. Jakob, O. Yarema and V. Wood, ACS Appl. Mater. Interfaces, 2013, 5, 2915–2919.
84. C. B. Murray, C. R. Kagan and M. G. Bawendi, Annu. Rev. Mater. Sci., 2000, 30, 545–610.
85. O. Ouellette, N. Hossain, B. R. Sutherland, A. Kiani, F. P. García De Arquer, H. Tan, M. Chaker, S. Hoogland and E. H. Sargent, ACS Energy Lett., 2016, 1, 852–857.
86. S. A. Maier, Plasmonics: Fundamentals and Applications, Springer US, New York, NY, 2007.
87. C. Heine and R. H. Morf, Appl. Opt., 1995, 34, 2476.
88. M. Agrawal and P. Peumans, Opt. Express, 2008, 16, 5385.
89. G. I. Koleilat, I. J. Kramer, C. T. O. Wong, S. M. Thon, A. J. Labelle, S. Hoogland and E. H. Sargent, Sci. Rep., 2013, 3, 2166.
90. X. Zhang and E. M. J. Johansson, Nano Energy, 2016, 28, 71–77.
91. H. Kim, S.-Y. Ahn, S. Kim, G. Ryu, J. H. Kyhm, K. W. Lee, J. H. Park and W. J. Choi, Opt. Express, 2017, 25, 17562.
92. X. Tang, M. M. Ackerman, G. Shen and P. Guyot-Sionnest, Small, 2019, 15, 1804920.
93. J. H. Kim, J. Lee, J. Lim, J. Roh, S. Baek, W. Kim, M. C. Suh and H. Yu, Adv. Funct. Mater., 2023, 33, 2214530.
94. L. Gu, K. Wen, Q. Peng, W. Huang and J. Wang, Small, 2020, 16, 2001861.
95. L. De Sio, T. Placido, R. Comparelli, M. L. Curri, M. Striccoli, N. Tabiryan and T. J. Bunning, Prog. Quantum Electron., 2015, 41, 23–70.
96. S. A. Maier, in Plasmonics: Fundamentals and Applications, Springer US, New York, NY, 2007, pp. 21–37.
97. F. Enrichi, A. Quandt and G. C. Righini, Renewable Sustainable Energy Rev., 2018, 82, 2433–2439.
98. J. He, K. Qiao, L. Gao, H. Song, L. Hu, S. Jiang, J. Zhong and J. Tang, ACS Photonics, 2014, 1, 936–943.
99. J. Hong, B.-S. Kim, B. Hou, Y. Cho, S. H. Lee, S. Pak, S. M. Morris, J. I. Sohn and S. Cha, Plasmonics, 2020, 15, 1007–1013.
100. T. Guan, W. Chen, H. Tang, D. Li, X. Wang, C. L. Weindl, Y. Wang, Z. Liang, S. Liang, T. Xiao, S. Tu, S. V. Roth, L. Jiang and P. Müller-Buschbaum, ACS Nano, 2023, 17, 23010–23019.
101. Y. Qiu, N. Yan, H. Yao and M. Chen, Infrared Phys. Technol., 2023, 135, 104980.
102. J. Tong, F. Suo, J. Ma, L. Y. M. Tobing, L. Qian and D. Zhang, Opto-Electronic Advances, School of Electrical and Electronic Engineering, Nanyang Technological University, Nanyang Avenue, 639798, Singapore, 2019, vol. 2, pp. 18002601–18002610.
103. M. Chen, L. Lu, H. Yu, C. Li and N. Zhao, Adv. Sci., 2021, 8, 2101560.
104. C.-C. Chang, Y. D. Sharma, Y.-S. Kim, J. A. Bur, R. V. Shenoi, S. Krishna, D. Huang and S.-Y. Lin, Nano Lett., 2010, 10, 1704–1709.
105. F. J. Beck, A. Stavrinadis, S. L. Diedenhofen, T. Lasanta and G. Konstantatos, ACS Photonics, 2014, 1, 1197–1205.
106. X. Tang, G. F. Wu and K. W. C. Lai, J. Mater. Chem. C, 2017, 5, 362–369.
107. X. Tang, M. M. Ackerman and P. Guyot-Sionnest, ACS Nano, 2018, 12, 7362–7370.
108. Q. Qiu and Z. Huang, Adv. Mater., 2021, 33, 2008126.
109. B. Zhu, M. Chen, Q. Zhu, G. Zhou, N. M. Abdelazim, W. Zhou, S. V. Kershaw, A. L. Rogach, N. Zhao and H. K. Tsang, Adv. Mater. Technol., 2019, 4, 1900354.
110. Q. Song, Y. Xu, Z. Zhou, H. Liang, M. Zhang, G. Zhu, J. Yang and P. Yan, ACS Photonics, 2022, 9, 2520–2527.
111. T. K. Gaylord, W. E. Baird and M. G. Moharam, Appl. Opt., 1986, 25, 4562.
112. F. J. Beck, A. Stavrinadis, T. Lasanta, J.-P. Szczepanick and G. Konstantatos, Opt. Express, 2016, 24, 759.
113. A. Chu, C. Gréboval, N. Goubet, B. Martinez, C. Livache, J. Qu, P. Rastogi, F. A. Bresciani, Y. Prado, S. Suffit, S. Ithurria, G. Vincent and E. Lhuillier, ACS Photonics, 2019, 6, 2553–2561.
114. M. Kopytko and A. Rogalski, Prog. Quantum Electron., 2016, 47, 1–18.
115. I. Yang, X. Zhang, C. Zheng, Q. Gao, Z. Li, L. Li, M. N. Lockrey, H. Nguyen, P. Caroff, J. Etheridge, H. H. Tan, C. Jagadish, J. Wong-Leung and L. Fu, ACS Nano, 2018, 12, 10374–10382.
116. M. A. Iqbal, M. Malik, T. K. Le, N. Anwar, S. Bakhsh, W. Shahid, S. Shahid, S. Irfan, M. Al-Bahrani, K. Morsy, H.-B. Do, V. K. Ponnusamy and P. V. Pham, ACS Mater. Lett., 2023, 5, 1027–1060.
117. M. Takase, Y. Miyake, T. Yamada, T. Tamaki, M. Murakami and Y. Inoue, in 2015 IEEE International Electron Devices Meeting (IEDM), IEEE, Washington, DC, USA, 2015, p. 30.2.1–30.2.4.
118. C. Dong, S. Liu, N. Barange, J. Lee, T. Pardue, X. Yi, S. Yin and F. So, ACS Appl. Mater. Interfaces, 2019, 11, 44451–44457.
119. M. Chen, X. Lan, X. Tang, Y. Wang, M. H. Hudson, D. V. Talapin and P. Guyot-Sionnest, ACS Photonics, 2019, 6, 2358–2365.
120. N. Goubet, A. Jagtap, C. Livache, B. Martinez, H. Portalès, X. Z. Xu, R. P. S. M. Lobo, B. Dubertret and E. Lhuillier, J. Am. Chem. Soc., 2018, 140, 5033–5036.
121. S. Zhang, C. Bi, T. Qin, Y. Liu, J. Cao, J. Song, Y. Huo, M. Chen, Q. Hao and X. Tang, ACS Photonics, 2023, 10, 673–682.
122. P. Malinowski, E. Georgitzikis, J. Maes, I. Vamvaka, F. Frazzica, J. Van Olmen, P. De Moor, P. Heremans, Z. Hens and D. Cheyns, Sensors, 2017, 17, 2867.
123. C. Gréboval, D. Darson, V. Parahyba, R. Alchaar, C. Abadie, V. Noguier, S. Ferré, E. Izquierdo, A. Khalili, Y. Prado, P. Potet and E. Lhuillier, Nanoscale, 2022, 14, 9359–9368.
124. G. Mercier, E. O. Polat, S. Shi, S. Gupta, G. Konstantatos, S. Goossens and F. H. L. Koppens, ACS Photonics, 2023, 10, 2994–3000.
125. E. Heves, C. Ozturk, V. Ozguz and Y. Gurbuz, IEEE Electron Device Lett., 2013, 34, 662–664.
126. P. E. Malinowski, J. Lee, Y. Li, E. Georgitzikis, V. Pejovic, I. Lieberman, T. Verschooten, S. Thijs, O. Furxhi, P. Heremans and D. Cheyns, Proc. of SPIE, 2021, 11765, 117650V.
127. V. Pejovic, J. Lee, E. Georgitzikis, Y. Li, J. H. Kim, I. Lieberman, P. E. Malinowski, P. Heremans and D. Cheyns, IEEE Electron Device Lett., 2021, 42, 1196–1199.
128. X. Tang, M. M. Ackerman, M. Chen and P. Guyot-Sionnest, Nat. Photonics, 2019, 13, 277–282.
129. S. Goossens, G. Navickaite, C. Monasterio, S. Gupta, J. J. Piqueras, R. Pérez, G. Burwell, I. Nikitskiy, T. Lasanta, T. Galán, E. Puma, A. Centeno, A. Pesquera, A. Zurutuza, G. Konstantatos and F. Koppens, Nat. Photonics, 2017, 11, 366–371.
130. A. Shultz, B. Liu, M. Gong, M. Alamri, M. Walsh, R. C. Schmitz and J. Z. Wu, ACS Appl. Nano Mater., 2022, 5, 16896–16905.
131. D. K. Hwang, Y. T. Lee, H. S. Lee, Y. J. Lee, S. H. Shokouh, J. Kyhm, J. Lee, H. H. Kim, T.-H. Yoo, S. H. Nam, D. I. Son, B.-K. Ju, M.-C. Park, J. D. Song, W. K. Choi and S. Im, NPG Asia Mater., 2016, 8, e233–e233.
132. H. T. Choi, J.-H. Kang, J. Ahn, J. Jin, J. Kim, S. Park, Y.-H. Kim, H. Kim, J. D. Song, G. W. Hwang, S. Im, W. Shim, Y. T. Lee, M.-C. Park and D. K. Hwang, ACS Photonics, 2020, 7, 1932–1941.
133. X. Hu, G. Xiao, Y. Li, S. Wu, Q. Chen, N. Li and X. Sui, ACS Appl. Electron. Mater., 2023, 5, 5378–5385.
134. Y. Zhang, Z. He, X. Du, J. Han, H. Lin, C. Zheng, J. Wang, G. Yang and S. Tao, Opt. Express, 2022, 30, 16644.
135. X. Du, J. Han, Z. He, C. Han, X. Wang, J. Wang, Y. Jiang and S. Tao, Adv. Mater., 2021, 33, 2102812.
136. D. Y. Kim, K. R. Choudhury, J. W. Lee, D. W. Song, G. Sarasqueta and F. So, Nano Lett., 2011, 11, 2109–2113.
137. N. Zhang, H. Tang, K. Shi, W. Wang, W. Deng, B. Xu, K. Wang and X. W. Sun, Appl. Phys. Lett., 2019, 115, 221103.
138. H. Yu, D. Kim, J. Lee, S. Baek, J. Lee, R. Singh and F. So, Nat. Photonics, 2016, 10, 129–134.
139. W. Chen, H. Tang, N. Li, M. A. Scheel, Y. Xie, D. Li, V. Körstgens, M. Schwartzkopf, S. V. Roth, K. Wang, X. W. Sun and P. Müller-Buschbaum, Nanoscale Horiz., 2020, 5, 880–885.
140. D. M. Balazs, N. Rizkia, H.-H. Fang, D. N. Dirin, J. Momand, B. J. Kooi, M. V. Kovalenko and M. A. Loi, ACS Appl. Mater. Interfaces, 2018, 10, 5626–5632.
141. G. Du, Z. Wang, T. Zhai, Y. Li, K. Chang, B. Yu, X. Zhao and W. Deng, ACS Appl. Mater. Interfaces, 2022, 14, 13572–13583.
142. R. Sliz, M. Lejay, J. Z. Fan, M.-J. Choi, S. Kinge, S. Hoogland, T. Fabritius, F. P. García De Arquer and E. H. Sargent, ACS Nano, 2019, 13, 11988–11995.
143. A. De Iacovo, C. Venettacci, C. Giansante and L. Colace, Nanoscale, 2020, 12, 10044–10050.
144. M. Böberl, M. V. Kovalenko, S. Gamerith, E. J. W. List and W. Heiss, Adv. Mater., 2007, 19, 3574–3578.
145. W. Chen, H. Tang, Y. Chen, J. E. Heger, N. Li, L. P. Kreuzer, Y. Xie, D. Li, C. Anthony, Z. Pikramenou, K. W. Ng, X. W. Sun, K. Wang and P. Müller-Buschbaum, Nano Energy, 2020, 78, 105254.
146. H. Aqoma and S.-Y. Jang, Energy Environ. Sci., 2018, 11, 1603–1609.
147. J. Z. Fan, M. Vafaie, K. Bertens, M. Sytnyk, J. M. Pina, L. K. Sagar, O. Ouellette, A. H. Proppe, A. S. Rasouli, Y. Gao, S.-W. Baek, B. Chen, F. Laquai, S. Hoogland, F. P. G. D. Arquer, W. Heiss and E. H. Sargent, Nano Lett., 2020, 20, 5284–5291.
148. D. Bederak, N. Sukharevska, S. Kahmann, M. Abdu-Aguye, H. Duim, D. N. Dirin, M. V. Kovalenko, G. Portale and M. A. Loi, ACS Appl. Mater. Interfaces, 2020, 12, 52959–52966.
149. X. Zhao, M. Li, T. Ma, J. Yan, G. M. G. Khalaf, C. Chen, H.-Y. Hsu, H. Song and J. Tang, Front. Optoelectron., 2023, 16, 27.
150. H. Wang, J. Pinna, D. G. Romero, L. Di Mario, R. M. Koushki, M. Kot, G. Portale and M. A. Loi, Adv. Mater., 2024, 2311526.
151. L. Liu, B. Yu, L. Kang, W. Deng and X. Zhao, Adv. Funct. Mater., 2023, 33, 2214781.
152. D. Wu, G. Du, H. Liu, W. Chen, X. Li, Z. Wang, H. Tang, B. Liu, C. Liu, Y. Chen, Z. Song, W. Deng, H. Yuan, K. Wang and X. Zhao, Adv. Opt. Mater., 2023, 11, 2300945.
153. P. Yang, L. Zhang, D. J. Kang, R. Strahl and T. Kraus, Adv. Opt. Mater., 2020, 8, 1901429.
154. G. Kara, S. Bolat, K. Sharma, M. J. Grotevent, D. N. Dirin, D. Bachmann, R. Furrer, L. F. Boesel, Y. E. Romanyuk, R. M. Rossi, M. V. Kovalenko, M. Calame and I. Shorubalko, Adv. Mater. Technol., 2023, 8, 2201922.
155. J. E. Bishop, T. J. Routledge and D. G. Lidzey, J. Phys. Chem. Lett., 2018, 9, 1977–1984.
156. M. Chen, H. Yu, S. V. Kershaw, H. Xu, S. Gupta, F. Hetsch, A. L. Rogach and N. Zhao, Adv. Funct. Mater., 2014, 24, 53–59.
157. I. J. Kramer, J. C. Minor, G. Moreno-Bautista, L. Rollny, P. Kanjanaboos, D. Kopilovic, S. M. Thon, G. H. Carey, K. W. Chou, D. Zhitomirsky, A. Amassian and E. H. Sargent, Adv. Mater., 2015, 27, 116–121.
158. I. J. Kramer, G. Moreno-Bautista, J. C. Minor, D. Kopilovic and E. H. Sargent, Appl. Phys. Lett., 2014, 105, 163902.
159. C. Girotto, D. Moia, B. P. Rand and P. Heremans, Adv. Funct. Mater., 2011, 21, 64–72.
160. X. Lan, O. Voznyy, A. Kiani, F. P. García De Arquer, A. S. Abbas, G. Kim, M. Liu, Z. Yang, G. Walters, J. Xu, M. Yuan, Z. Ning, F. Fan, P. Kanjanaboos, I. Kramer, D. Zhitomirsky, P. Lee, A. Perelgut, S. Hoogland and E. H. Sargent, Adv. Mater., 2016, 28, 299–304.
161. J. Yang, M. Kim, S. Lee, J. W. Yoon, S. Shome, K. Bertens, H. Song, S. G. Lim, J. T. Oh, S. Y. Bae, B. R. Lee, W. Yi, E. H. Sargent and H. Choi, ACS Appl. Mater. Interfaces, 2021, 13, 36992–37003.
162. S. Zhang, M. Chen, G. Mu, J. Li, Q. Hao and X. Tang, Adv. Mater. Technol., 2022, 7, 2101132.
163. F. Li, J. J. Liu, Q. Xu, R. Chang, L. Wang, Z. Wu, H. Shen and Z. Du, J. Phys. Chem. Lett., 2023, 14, 4252–4258.
164. J. Yang, S. Lu, B. Xia, P. Liu, Y. Yang, Z. Xiao, J. Zhang, L. Gao and J. Tang, Phys. Rev. Appl., 2023, 19, 014021.
165. H. Xia, H. Hu, Y. Wang, M. Yu, M. Yuan, J. Yang, L. Gao, J. Zhang, J. Tang and X. Lan, J. Mater. Chem. C, 2024, 12, 10919–10928.
166. P. Liu, S. Lu, J. Liu, B. Xia, G. Yang, M. Ke, X. Zhao, J. Yang, Y. Liu, C. Ge, G. Liang, W. Chen, X. Lan, J. Zhang, L. Gao and J. Tang, InfoMat, 2024, 6, e12497.
167. M.-J. Choi, F. P. García De Arquer, A. H. Proppe, A. Seifitokaldani, J. Choi, J. Kim, S.-W. Baek, M. Liu, B. Sun, M. Biondi, B. Scheffel, G. Walters, D.-H. Nam, J. W. Jo, O. Ouellette, O. Voznyy, S. Hoogland, S. O. Kelley, Y. S. Jung and E. H. Sargent, Nat. Commun., 2020, 11, 103.
168. M. Albaladejo-Siguan, D. Becker-Koch, E. C. Baird, Y. J. Hofstetter, B. P. Carwithen, A. Kirch, S. Reineke, A. A. Bakulin, F. Paulus and Y. Vaynzof, Adv. Energy Mater., 2022, 12, 2202994.
169. W. Chen, R. Guo, H. Tang, K. S. Wienhold, N. Li, Z. Jiang, J. Tang, X. Jiang, L. P. Kreuzer, H. Liu, M. Schwartzkopf, X. W. Sun, S. V. Roth, K. Wang, B. Xu and P. Müller-Buschbaum, Energy Environ. Sci., 2021, 14, 3420–3429.
170. L. Zheng, W. Zhou, Z. Ning, G. Wang, X. Cheng, W. Hu, W. Zhou, Z. Liu, S. Yang, K. Xu, M. Luo and Y. Yu, Adv. Opt. Mater., 2018, 6, 1800985.
171. Q. Nian, L. Gao, Y. Hu, B. Deng, J. Tang and G. J. Cheng, ACS Appl. Mater. Interfaces, 2017, 9, 44715–44723.
172. S. Ahn, C. Ingrosso, A. Panniello, M. Striccoli, G. V. Bianco, A. Agostiano, G. Bruno, M. L. Curri and O. Vazquez-Mena, Adv. Electrode Mater., 2022, 8, 2100672.
173. Y. Dong, D. Parobek and D. H. Son, J. Korean Ceram. Soc., 2018, 55, 515–526.
174. Y. Wang, X. Li, J. Song, L. Xiao, H. Zeng and H. Sun, Adv. Mater., 2015, 27, 7101–7108.
175. C. De Weerd, L. Gomez, A. Capretti, D. M. Lebrun, E. Matsubara, J. Lin, M. Ashida, F. C. M. Spoor, L. D. A. Siebbeles, A. J. Houtepen, K. Suenaga, Y. Fujiwara and T. Gregorkiewicz, Nat. Commun., 2018, 9, 4199.
176. H. Chen, J. M. Pina, Y. Hou and E. H. Sargent, Adv. Energy Mater., 2022, 12, 2100774.
177. M. Ahmadi, T. Wu and B. Hu, Adv. Mater., 2017, 29, 1605242.
178. M. Kim, G. Bae, K. N. Kim, H. Jo, D. S. Song, S. Ji, D. Jeon, S. Ko, S. J. Lee, S. Choi, S. Yim, W. Song, S. Myung, D. H. Yoon, K.-S. An and S. S. Lee, NPG Asia Mater., 2022, 14, 89.
179. X. Yang, J. Phys.: Conf. Ser., 2021, 1759, 012027.
180. P. Cheng, Z. Liu, R. Kang, J. Zhou, X. Wang, J. Zhao and Z. Zuo, ACS Omega, 2023, 8, 26351–26358.
181. B. Sun, A. M. Najarian, L. K. Sagar, M. Biondi, M. Choi, X. Li, L. Levina, S. Baek, C. Zheng, S. Lee, A. R. Kirmani, R. Sabatini, J. Abed, M. Liu, M. Vafaie, P. Li, L. J. Richter, O. Voznyy, M. Chekini, Z. Lu, F. P. García De Arquer and E. H. Sargent, Adv. Mater., 2022, 34, 2203039.
182. F. Pelayo García De Arquer, F. J. Beck, M. Bernechea and G. Konstantatos, Appl. Phys. Lett., 2012, 100, 043101.
183. J. He, M. Luo, L. Hu, Y. Zhou, S. Jiang, H. Song, R. Ye, J. Chen, L. Gao and J. Tang, J. Alloys Compd., 2014, 596, 73–78.
184. V. Adinolfi, I. J. Kramer, A. J. Labelle, B. R. Sutherland, S. Hoogland and E. H. Sargent, ACS Nano, 2015, 9, 356–362.
185. F. P. García De Arquer, T. Lasanta, M. Bernechea and G. Konstantatos, Small, 2015, 11, 2636–2641.
186. S. Masala, V. Adinolfi, J. Sun, S. D. Gobbo, O. Voznyy, I. J. Kramer, I. G. Hill and E. H. Sargent, Adv. Mater., 2015, 27, 7445–7450.
187. L. Gao, C. Chen, K. Zeng, C. Ge, D. Yang, H. Song and J. Tang, Light:Sci. Appl., 2016, 5, e16126–e16126.
188. I. Nikitskiy, S. Goossens, D. Kufer, T. Lasanta, G. Navickaite, F. H. L. Koppens and G. Konstantatos, Nat. Commun., 2016, 7, 11954.
189. Y. Kim, K. Bicanic, H. Tan, O. Ouellette, B. R. Sutherland, F. P. García De Arquer, J. W. Jo, M. Liu, B. Sun, M. Liu, S. Hoogland and E. H. Sargent, Nano Lett., 2017, 17, 2349–2353.
190. V. Adinolfi and E. H. Sargent, Nature, 2017, 542, 324–327.
191. E. Georgitzikis, P. E. Malinowski, J. Maes, A. Hadipour, Z. Hens, P. Heremans and D. Cheyns, Adv. Funct. Mater., 2018, 28, 1804502.
192. A. Jagtap, B. Martinez, N. Goubet, A. Chu, C. Livache, C. Gréboval, J. Ramade, D. Amelot, P. Trousset, A. Triboulin, S. Ithurria, M. G. Silly, B. Dubertret and E. Lhuillier, ACS Photonics, 2018, 5, 4569–4576.
193. Y. Tang, F. Wu, F. Chen, Y. Zhou, P. Wang, M. Long, W. Zhou, Z. Ning, J. He, F. Gong, Z. Zhu, S. Qin and W. Hu, Small, 2018, 14, 1803158.
194. J.-Y. Zhang, J.-L. Xu, T. Chen, X. Gao and S.-D. Wang, ACS Appl. Mater. Interfaces, 2019, 11, 44430–44437.
195. O. Özdemir, I. Ramiro, S. Gupta and G. Konstantatos, ACS Photonics, 2019, 6, 2381–2386.
196. K. Xu, X. Xiao, W. Zhou, X. Jiang, Q. Wei, H. Chen, Z. Deng, J. Huang, B. Chen and Z. Ning, ACS Appl. Mater. Interfaces, 2020, 12, 15414–15421.
197. X. Tang, M. Chen, A. Kamath, M. M. Ackerman and P. Guyot-Sionnest, ACS Photonics, 2020, 7, 1117–1121.
198. B. K. Jung, H. K. Woo, C. Shin, T. Park, N. Li, K. J. Lee, W. Kim, J. H. Bae, J. Ahn, T. N. Ng and S. J. Oh, Adv. Opt. Mater., 2022, 10, 2101611.
199. J. Jin, J. Kyhm, D. K. Hwang, K.-S. Lee, T.-Y. Seong and G. W. Hwang, ACS Appl. Nano Mater., 2021, 4, 1–6.
200. Y. Shi, Z. Wu, X. Dong, P. Chen, J. Wang, J. Yang, Z. Xiang, M. Shen, Y. Zhuang, J. Gou, J. Wang and Y. Jiang, Nanoscale, 2021, 13, 12306–12313.
201. L. Sheng, C. Yi, L. Zheng, Y. Liu, J. Zheng and X. Gong, J. Mater. Chem. C, 2022, 10, 2783–2791.
202. K. Xu, L. Ke, H. Dou, R. Xu, W. Zhou, Q. Wei, X. Sun, H. Wang, H. Wu, L. Li, J. Xue, B. Chen, T.-C. Weng, L. Zheng, Y. Yu and Z. Ning, ACS Appl. Mater. Interfaces, 2022, 14, 14783–14790.
203. J. Leemans, V. Pejović, E. Georgitzikis, M. Minjauw, A. B. Siddik, Y. Deng, Y. Kuang, G. Roelkens, C. Detavernier, I. Lieberman, P. E. Malinowski, D. Cheyns and Z. Hens, Adv. Sci., 2022, 9, 2200844.
204. B. Wang, H. Hu, M. Yuan, J. Yang, J. Liu, L. Gao, J. Zhang, J. Tang and X. Lan, Small Methods, 2024, 2301557.
205. L. Peng, Y. Wang, Y. Ren, Z. Wang, P. Cao and G. Konstantatos, ACS Nano, 2024, 18, 5113–5121.

---

Footnote

† These authors contributed equally to this work.

---