The ability to tune thin metal oxide coatings by wet-chemistry is desirable for many applications, yet it remains a key synthetic challenge. In this work, we introduce a general colloidal atomic layer deposition (c-ALD) synthesis to grow metal oxide shells (AlOx, ZnOx, TiOx) of tunable thickness (1 to 15 nm) around nanocrystalline cores of different compositions, including quantum dots (perovskites-PeQDs, CdSe, C-dots, InP).¹

We compare the c-ALD with the previously developed gas-phase ALD in film to highlight its advantages which comprise the preserved colloidal dispersability, the improved optical properties and the stability.²

Finally, we illustrate the importance of such a finely tuned metal oxide shell thickness to study nanoscale phenomena such as energy transfer between PeQDs and CdSe nanoplates, between PeQDs and metal nanoparticles and the anion exchange reaction in PeQDs.^1,3-5

Lead-halide perovskite APbX₃ (A=Cs or organic cation; X=Cl, Br, I) quantum dots (QDs) are subject of intense research due to their exceptional properties as both classical¹ and quantum light sources.^2-4 Many challenges often faced with this material class concern the long-term optical stability, a serious intrinsic issue connected with the labile and polar crystal structure of APbX₃ compounds. When conducting spectroscopy at a single particle level, due to the highly enhanced contaminants (e.g., water molecules, oxygen) over NC ratio, deterioration of NC optical properties occurs within tens of seconds, with typically used excitation power densities (1-100 W/cm²), and in ambient conditions. By using a suitable polymer matrix, these detrimental effects can be suppressed, and intrinsic exciton and multi-exciton dynamics can be explored at the single particle level. 

Here, we report a comprehensive investigation of the room temperature single QD optical properties. The results reveal the origin of the QD homogeneous PL linewidths, and the peculiar size-dependent exciton and multi-excitons recombination dynamics.  

Such findings guide the further design of robust single photon sources operating at room temperature.  

References:

[1] Akkerman et al., Genesis, challenges and opportunities for colloidal lead halide perovskite nanocrystals. Nat. Mater. 17, 394–405 (2018).

[2] Becker et al., Bright triplet excitons in caesium lead halide perovskites. Nature 553, 189–193 (2018).

[3] Rainò et al., Superfluorescence from lead halide perovskite quantum dot superlattices. Nature 563, 671–675 (2018).

[4] Utzat et al., Coherent single-photon emission from colloidal lead halide perovskite quantum dots. Science 363, 1068–1072 (2019).

[5] Rainò et al., Underestimated Effect of a Polymer Matrix on the Light Emission of Single CsPbBr₃ Nanocrystals. Nano Lett. 19, 3648-3653, (2019).

At the core of the heated debate around 0D lead bromide perovskites is the origin of the green emission, which was observed in bulk powders and single crystals, but not in nanocrystals. Early speculations of an intrinsic emission of the material gradually sedimented into two main lines of thought: some groups point at deep band levels induced by point defects, most likely Br vacancies, while others claim the green emission stems from embedded CsPbBr₃ (3D) nanocrystals or some other lower dimensional perovskite structures, whose formation within the bulk 0D material is challenging to be discerned from XRD patterns.  Although the latest research seems to lean toward the 3D contamination induced emission, no final consensus has been achieved so far.

In this work we present ab initio molecular dynamics (MD) simulations to study Cs₄PbBr₆ in a wide range of temperatures from 34 K to room temperature, and we analyze the electronic structure of the system at every timestep. We also extended the study to the 0D crystal with Br vacancies and to the 3D perovskite crystal. By comparison of the three systems, we demonstrate that point defects cannot be responsible of the green emission because we observe a fast phonon-mediated quenching mechanism of the intrinsic emission of the material already at low temperatures. This is idea is corroborated by variable temperature photoluminescence studies on 200 nm size non-emissive 0D NC, which lack the impurities present in powders.

The relaxation of above-gap (‘hot’) carriers is responsible for major efficiency losses in present-day solar cells, and involves a complex interplay between carrier-carrier and carrier-phonon coupling. Unravelling the mechanisms of cooling is therefore an essential step for both understanding and developing emerging photovoltaic materials. Perovskite nanomaterials are an exciting class of compounds because they offer facile and broad optoelectronic tunability by size, dimensionality and composition. Here, we aim to elucidate the effects of these properties on carrier cooling by employing ultrafast pump-push-probe spectroscopy. This three-pulse technique allows cooling to be isolated from a melee of other excited state processes, while also allowing independent control over the hot and cold (band-edge) carrier subpopulations. These experiments show that while carrier cooling is generally indifferent to nanocrystal size in moderately confined systems, intriguing results are obtained upon altering the shape of the nanocrystal, and are also influenced greatly by material composition.

Lead halide perovskite nanocrystals (NCs) have emerged as a potential material for LED and solar cell applications [1]. However, despite of their promising performance, the band gap of lead halide perovskite NCs remains large that limit the absorption in infrared (IR) part of spectrum [2]. In this work, we explore the possibility to harvest the IR spectrum using formamidinium lead iodine (FAPI) perovskite nanocrystals by doping of PbS NCs. As a short wave IR absorber, PbS NCs has attracted much attention yet it suffer with high dark current eventually that limit the device performance. By mixing the two compounds with an optimum ratio, it is possible to preserve most of the IR absorption while the transport driven by the wider band gap of the perovskite, this enabling a dark current reduction. In addition to understand the electronic structure of FAPI/PbS hybrid, we fabricated an FET using high capacitance solid state gating. Using this strategy, we show that the hybrid material has an n-type nature with a charge carrier mobility of 2 x 10^-3 cm² V^-1s^-1.

However, as FAPI is introduced into the PbS NCs array, the benefit of the reduced dark current is partly mitigated by a reduced IR absorption. This problem is address by introducing a plasmonic resonator. The latter relies on a grating that generate a multi passes of the light into the absorbing layer thus enhancing the IR absorption [3]. The resonant electrode enhances the light-matter coupling within the NCs film that enhance the IR absorption up to 3 times [4]. In addition, the reduction of the interelectrode spacing enable photoconduction gain leading to an improved responsivity and detectivity by two order of magnitude in comparison to pristine PbS.

Colloidal nanoplatelets (NPLs) have recently emerged as a novel and exciting class of materials. While several established procedures are available for highly luminescent 4.5 monolayer and thicker NPLs, with emission spanning the green and red spectral region, typical synthesis protocols to prepare blue-emitting CdSe NPLs (λ ≈ 460 nm) yields particles with large surface areas and poor photoluminescence quantum efficiency, often accompanied by a strong emission peak from intragap trap states. Further applications may therefore be hampered by their high surface-to-volume ratio.

Here we present our work on the development of a synthesis protocol that achieves improved control over the lateral size, by exploiting a series of long-chained carboxylate precursors, varying from cadmium octanoate (C₈) to cadmium stearate (C₁₈). The length of the metallic precursor is key to tune the width and aspect ratio of the final NPLs (from 3:1 to 15:1), as well as the overall reaction yield, which increases for shorter chain length. The reduced NPL lateral dimensions lead to enhanced photoluminescence quantum efficiencies, reaching up to 30%, and good colloidal stability. As the width can be tuned down to 3.7 nm, we were able to construct also a convenient sizing curve, relating the NPL absorption position and width, and careful comparison with 4.5 monolayer NPLs demonstrate that the blue-emitting NPLs are characterized by faster emission lifetimes and a higher absorption coefficient. Via a slight adjustment, we also obtained 2.5 monolayers NPLs, with near-UV emission (λ ≈ 400 nm), and a quantum efficiency up to 11%. Our results contribute to achieving stable and efficient sources for applications such as blue and UV light emitting devices or lasers, or fast quantum light sources.

Core/shell nanocrystals in which the materials change gradually from core to shell are very very promising structures to optimise the opto-electronic properties and quantum efficiencies of nanoscale semiconductors. Gradients are able to minimise crystal defects, lattice mismatch, and can be used to engineer the envelope wave function of excitons in order to suppress non-radiative Auger processes. However, due to the small size of the particles, so far no reliable method exists to quantify the extent of such a gradient.

In this work we have measured the material gradient of ZnSe/CdS core/shell nanocrystals, which were synthesised at elevated temperatures (260 and 290 °C), which controls the rate of radial ion migration [1]. We used EXAFS spectroscopy to determine the average coordination of selenium ions, which were fitted to a continuum model for the radial distribution of cations and anions [2].

It could be shown that for the 260 °C sample the data shows strong cation migration, which transports significant amounts (> 50%) of cadmium into the core, while the anion gradient is consistent with negligible ion migration beyond the interfacial monolayer. This is significant, because many shell growth protocols that are assumed to produce sharp interfaces are performed at similar temperatures. At higher temperatures of 290 °C the data deviates strongly from the model, with effectively less cation migration. This is explained by the formation of an ordered Zn_0.5Cd_0.5Se superlattice in the core in order to mitigate the lattice mismatch die to the increasing CdSe content of the core [3]. Raman spectroscopy shows selective resonant enhancement of the core LO phonon overtones, which indicates that the exciton is primarily localized in the core and at interfacial traps, and that the electronic structure flips from a type-II to a type-I system.

Hence, the combination of X-ray and Raman spectroscopy is able to identify both the chemical and electronic structure of core/shell particles and produces an accurate gradient model that can be employed in more precise and predictive structural calculations. The high-temperature product sheds light on why some highly emissive nanocrystals still blink and struggle to reach unity quantum yield [4].

Vacancy-ordered triple perovskites have recently come under the scientific spotlight as promising materials for high-performance next-generation optoelectronic technologies.[1,2] Their A3B2X9 stoichiometry facilitates the replacement of the toxic Pb2+ cation with a benign isolectronic B3+ cation (e.g. Bi3+ or Sb3+), while preserving the perovskite crystal structure. Unfortunately, however, these materials tend to exhibit large bandgaps (> 2 eV), impeding their application in many photo-catalytic/voltaic devices.[3,4]

In this work, we demonstrate a drastic shift of over 1 eV in the optical absorption onset of Cs3Bi2Br9 (from 2.58 eV to 1.39 eV), upon doping with tin. Through a combination of detailed theoretical and experimental characterisation of this novel material, we elucidate the origin of broadband absorption. Sn is found to disproportionate in the doped material, inducing a strong intervalence charge transfer (IVCT) transition, whilst preserving the structural integrity of the perovskite framework.

Our work provides valuable insight regarding the effects of mixed-valency and structure-property relationships in perovskite-inspired materials, guiding design strategies and expanding the compositional space of candidate materials. Moreover, we anticipate that this massive reduction in absorption onset could aid charge transport and/or photo-catalytic performance, opening the door to unexplored applications of this material class.

Photon upconversion, where two or more low energy photons are converted into one high energy photon, shows great potential in bioimaging, catalysis and solar energy conversion. Photon upconversion has traditionally been realized with lanthanide-doped nanoparticles, or organic dye sensitized triplet-triplet annihilation (TTA) based upconversion platforms. In recent years, QD sensitized triplet-triplet annihilation based upconversion systems have achieved impressive upconversion quantum efficiency and demonstrated many unique advantages, including high photostability, large extinction coefficient, high spectral coverage and tunability, and low singlet-triplet energy gap. In this talk, we discuss our recent work in developing and understanding QD/mediator interface for efficient QD-sensitized photon upconversion. We summarize the main results of time-resolved spectroscopic studies of various factors affecting the rate of triplet energy transfer (TET) from the QD to the surface attached mediator (TET1) and from the mediator to the emitter in solution (TET2). To identify the key design rules, we compare three PbS sensitized upconversion systems using three mediator molecules with the same tetracene triplet acceptor at different distances from the QD. Our results show that the mediator triplet state is mostly formed by direct TET from quantum dot. With increasing distance between the mediator and PbS QD, the efficiency of the TET1 from the QD to the mediator decreases due to a decrease in the rate of this triplet energy transfer step, while the efficiency of the TET2 from the mediator to emitter increases due to a reduction in the QD induced mediator triplet state decay (via the external heavy atom effect). The rate constant of TET2 is three orders of magnitude slower than the diffusion limited value. We show that the effect of QD/mediator on the total unconversion efficiency measured under CW illumination conditions can be well accounted for by the independently determined efficiencies of TET1 and TET2 steps, providing important insight on the design and rational improvement of efficient photon upconversion systems.

Singlet exciton fission is a process that occurs in select organic materials wherein a spin-singlet exciton redistributes its energy to form a spin-correlated triplet exciton pair. Incorporating singlet fission materials into light harvesting platforms offers potential to enhance their performance by 40% while singlet fission’s ability to create entangled spin states at room temperature makes it attractive for quantum computing devices. Likewise, singlet fission’s inverse process, triplet fusion, has been used to create photon upconversion systems that produce high-energy excitons from pairs of low-energy photons. Such systems can enable new near-infrared sensors and photocatalysts driven by low-energy light. Intrinsic to the design of any singlet fission or triplet fusion-based device, however, is the exchange of energy, typically in the form of a spin-triplet exciton, between an organic material from an inorganic semiconductor. Hybrid materials consisting of semiconductor quantum dots functionalized with organic molecules are a premier platform for study of this energy transfer process. The high surface to volume ratio of these materials effectively means they consist entirely of interfacial molecules and the energy level tunability of quantum dots allows exploration of how the redox properties of the interface impact energy transfer.

Here, we report results on both PbS and Si quantum dots interfaced with a range of acene and rylene energy acceptors. We find that by tuning the energy level alignment of PbS quantum dots to that of rylene acceptors, the transfer of charge carriers across the interface can be varied by an order of magnitude. Interestingly, electronic structure calculations suggest this rate variation stems from electrostatic effects that both alter interfacial energy level alignment and shift the average orientation of molecules tethered to PbS. For Si quantum dots, energy transfer to acene and rylene acceptors is decidedly slow, unfolding on nanosecond to microsecond timescales due to weak coupling. Nevertheless, this process is highly efficient due to a lack of competing deactivation pathways. Interestingly, we find subtle changes in the structure of the triplet exciton energy acceptor lead to large changes in the energy transfer rate. These rate changes are unexpected on the basis of electronic structure calculations performed on molecules tethered to Si(111) surfaces, suggesting more exotic surface structures may play a key role in facilitating triplet energy transfer from silicon to organic molecules.

Although colloidal semiconductor nanocrystals have been widely studied for three decades, the understanding of their excited-state dynamics continues to evolve. Charge-carrier trap states on nanocrystal surfaces play an essential role in processes such as electron–hole recombination and charge transfer but their dynamics are challenging to probe spectroscopically. Photogenerated holes in CdS and CdSe nanocrystals trap to the orbitals of undercoordinated S and Se atoms on the particle surface on a picosecond timescale. We recently presented evidence that trapped holes on the surfaces of CdS and CdSe nanocrystals are not stationary but instead undergo a diffusive random walk at room temperature. The initial evidence came from interpretation of electron-hole recombination dynamics in transient absorption (TA) spectroscopy data of non-uniform CdS and CdSe nanorods. More recently, temperature-dependent TA data provided insights into the mechanisms of trap-to-trap hole hopping. The experimental data, together with theoretical insights from our collaborators Joel Eaves and coworkers, builds an increasingly precise description of trapped-hole diffusion in nanocrystals. This presentation will feature our most recent progress in this area.

Cation exchange, a chemical transformation used to modify a crystal whereby a cation from solution replaces a host cation, has recently become a highly effective tool for enabling the synthesis of nanoparticles with novel chemical compositions. In particular, aliovalent doping of CdSe nanocrystals (NCs) via cation exchange of cadmium ions for silver ions has become quite popular for manipulating the optical and electronic properties of the doped NCs, such as for producing n- or p-type NCs. However, despite over a decade of study, the relationship between optical properties of the NC and the aliovalent dopants has largely gone unexplained, partially due to an inability to precisely characterize the physical properties of the doped NC.

We will discuss how electrostatic force microscopy (EFM) with single electron sensitivity can be used to determine the charges of individual, cation-doped CdSe NCs in order to investigate their net charge as a function of added cations. While there was no direct trend relating the NC charge to the relative amount of cation per NC, there was a remarkable and unexpected correlation between the average NC charge and ensemble exciton photoluminescence (PL) intensity, for all dopant cations introduced [1]. We use an effective mass theoretical model to conclude that the changes in PL intensity, as tracked also by changes in NC charge, are likely a consequence of changes in the NC radiative rate caused by symmetry breaking of the electronic states of the nominally spherical NC due to the Columbic potential introduced by ionized cations. Further, we show through energy loss spectroscopy and PL spectroscopy on individual NCs that the cation exchange process is highly heterogeneous, which has profound implications for possible future applications of doped NCs.

Delocalization of excitons within semiconductor quantum dots (QDs) into states at the interface of the inorganic core and organic ligand shell by so-called “exciton-delocalizing ligands (EDLs)” is a promising strategy to enhance coupling of QD excitons with proximate molecules, ions or other QDs. EDLs thereby enable enhanced rates of charge carrier extraction from, and transport among, QDs and dynamic colorimetric sensing. The application of reported EDLs – which bind to the QD through thiolates or dithiocarbamates – is however limited by the irreversibility of their binding and their low oxidation potentials, which lead to a high yield of photoluminescence-quenching hole trapping on the EDL. Here we discuss a new class of EDLs for QDs, 1,3-dimethyl-4,5-disubstituted imidazolylidene N-heterocyclic carbenes (NHCs), where the 4,5 substituents are either Me, H, or Cl, and 1,3-dimesitylnapthoquinimidazolylidene NHC (nqNHC). Post-synthetic ligand exchange of native oleate capping ligands for NHCs results in a bathochromic shift of the optical band gap of CdSe QDs (R = 1.17 nm) of up to 111 meV while maintaining colloidal stability of the QDs. This shift is reversible for the MeNHC-capped, HNHC-capped, and nqNHC-capped QDs upon protonation of the NHC. The magnitude of exciton delocalization induced by the NHC (after scaling for surface coverage) increases with increasing acidity of its pi-system, which depends on the substituent in the 4,5 positions of the imidazolylidene. The NHC-capped QDs maintain photoluminescence quantum yields of up to 4.2 ± 1.8 % for shifts of the optical band gap as large as 106 meV. Spectroelectrochemistry shows that the reduction of the napthoquinone moiety of QD-bound nqNHC ligandsto their radical anions results in an additional magnitude of bathochromic shift, ΔΔR, relative to the QDs capped with nqNHC ligands in their neutral state, a redox-sensitive exciton delocalizing system.

Colloidal halide perovskite nanocrystals (NCs) have the possibility of easy scale-up due to their batch synthesis and have demonstrated excellent optoelectronic properties. In particular, perovskite NCs have remarkably high photoluminescence quantum yields in solution and as thin films and impressive open circuit voltages in photovoltaic devices. Despite these promising results, little work has been done to understand the stability of CsPbI₃ NCs for optoelectronic device applications. It has been previously shown that the ligands impart tensile surface strain, which stabilizes the black three-dimensional (3D) perovskite phase against phase degradation, making CsPbI₃ NCs some of the most structurally robust inorganic halide perovskites to date. However, understanding exactly how CsPbI₃ NCs degrade under ambient conditions is critical. We demonstrate that the degradation mechanism of NCs is unique from, and 2 orders of magnitude slower than, their polycrystalline thin-film counterparts. Under specific conditions, CsPbI₃ NC films show a compositional instability instead of the phase instability seen in large grain CsPbI₃. This is mediated through reactions with superoxide and other reactive oxygen species, which are initiated from surface defect states, O₂ and light. We then use this mechanistic insight to identify multiple strategies to prolong the lifetimes of CsPbI₃ NC films, by going beyond surface strain to mitigate key surface chemistries. We demonstrate that (1) minimizing the number of surface defects (2) using an alkylammonium bromide ligand surface treatment and (3) encapsulation with an oxygen scavenging layer all increase NC film lifetimes by inhibiting various steps in the photo-oxidation degradation reaction.

Solution-processed quantum dots (QDs) are finding applications in a wide-range of technologies from displays and lighting to photovoltaics and photodetectors. Advances in real-world technologies have been enabled by an increasing ability to fine-tune opto-electronic properties with strategies including quantum confinement effects, advanced heterostructuring (band-structure engineering at the nanoscale), and chemical manipulation of interfaces and surfaces. Taken together, these strategies have yielded numerous breakthroughs and insights into key fundamental excited-state processes in semiconductor nanocrystals. In our lab, we have focused on developing heterostructures that lead to suppression of non-radiative processes, including blinking, photobleaching, and Auger recombination.[1-8] Despite realizing novel properties that support a wide range of applications,[9-13] we cannot claim to have found the perfect QD. Even the most robust nanocrystal will fail in its most fundamental of property – the ability to emit light – in the face of specific stressors, such as high photon flux, temperature, and exposure to atmospheric oxygen/water.

Previously, we developed a “single-QD stress test” that was used to evaluate degradation-photophysics in two types of ultrastable, “giant” core/thick-shell QDs (gQDs).[14] Here, I will describe the latest in our efforts to elucidate QD failure mechanisms, with the aim to pinpoint structural and chemical features leading to the “killing” of a QD.[15] Specifically, we developed a method based on solid-state spectroscopy to obtain kinetic and thermodynamic parameters of photo-thermal degradation in single QDs, systematically varying ambient temperature and photon-pump fluence. We described the resulting degradation in emission with a modified form of the Arrhenius equation and showed that this reaction proceeds via pseudo zero-order reaction kinetics by a surface-assisted process with an activation energy of 60 kJ/mol. We note that the rate of degradation is ~12 orders of magnitude slower than the rate of excitonic processes, indicating that the reaction rate is not determined by ultrafast electron/hole trapping. We further determined that full-power LED-like excitation can add 90-140 K of heat to the nanocrystals due to high excitation rates and associated nonradiative relaxation. The specific reactions that are responsible for photo-oxidative degradation in gQDs are as yet unknown. However, at least one of the primary degradation reactions is now known to be a zero-order process with a reaction activation energy that is independent of photon flux or wavelength. In the search for new robust light-emitting nanocrystals, the new analysis method will enable direct comparisons between differently engineered nanomaterials or different organic/inorganic surface treatments.