Colloidal quantum dots and a method of making such colloidal quantum dots

Abstract

A method of synthesising AB colloidal quantum dots (CQDs) is provided, wherein A is Al, Ga, In or a mixture thereof, and B is P, As, Sb or a mixture thereof. The method comprises the steps of: a) reac

Description

[0001] COLLOIDAL QUANTUM DOTS AND A METHOD OF MAKING SUCH COLLOIDAL QUANTUM DOTS

[0002] Field

[0003] The present application relates to colloidal quantum dots and to a method of making such colloidal quantum dots.

[0004] Background

[0005] The short-wave infrared (SWIR), which typically covers a wavelength range 1100- 3000 nm, is a region of the electromagnetic spectrum in which critical technologies operate, such as optical communication, machine vision, medical imaging, biometric authentication, spectroscopy, and surveillance.1-4These key technologies are enabled by SWIR detectors, which depend on semiconductor absorbers with an appropriate bandgap in the SWIR.3The conventional absorbing layer in SWIR detectors is typically formed from vacuum-deposited, epitaxially-grown, group lll-V semiconductors.2 4Within this family of semiconductors, InSb stands out as a direct bandgap infrared absorber (0.17 eV, 7294 nm) possessing notable electronic properties, such as ultrafast bulk electron mobility of -77000 cm2 / (V S) at 300 K, low carrier effective masses and an ultralow thermal conductivity of 0.18 W / (cm °C).5-8Its large Bohr exciton radius (-61 nm) enables bandgap tunability of InSb nanostructures (via the quantum size effect) across the SWIR region.9Therefore, nanostructured InSb is considered a potential environmentally friendly alternative to other SWIR absorbers that include Pb, Cd, and Hg chalcogenides.10-14Although InSb nanostructures have been prepared by chemical vapor deposition and epitaxial growth, these techniques require expensive tools and large capital for industrial-scale fabrication.15-17

[0006] Colloidal quantum dots (CQDs) have emerged as potentially disruptive materials for SWIR detectors in terms of cost, scalability, and versatility.10’121 181 19CQDs are semiconductor nanocrystals whose bandgap can be tailored via the quantum size effect by controlling their size and shape through solution-based colloidal synthesis routes.9However, the most developed CQD systems for SWIR applications are Pb and Hg chalcogenides, which are practically limited due to their toxicity and regulated usage in devices.3’121 131 20This has spurred intense interest in lll-V semiconductors, e.g., InP, InAs, and InSb, as potential environmentally friendly alternatives to Pb and Hg chalogenides.21-34

[0007] Among lll-V semiconductors, InSb CQDs offer the widest potential for tunability across the SWIR region of the spectrum, due to their narrower bandgap and large Bohr exciton radius.9Unfortunately, InSb CQDs have hardly been exploited due to their complex synthesis, and uncommon, reactive precursors.22’24’26’27’29 35Compared to InP and InAs, the covalent bonding in InSb is stronger,36which necessitates highly reactive indium and antimony precursors being reacted at a high temperature. Moreover, the synthesis of InSb CQDs is normally carried out under reducing conditions because the most readily available antimony precursors are in the Sb3+state.24’26’29 35In previous studies, various approaches have been explored for the synthesis of InSb CQDs using the heat-up colloidal method.37For example, the colloidal synthesis of size-tunable InSb CQDs that exhibit sharp excitonic features in the absorption spectrum was reported by using an antimony silylamide complex.24However, the antimony silylamide complex is unstable, needs to be stored at -40°C in an inert atmosphere (like an N2 glove box), and also presents scalability issues. In another report, a commercial antimony source, Sb[NMe2]s, and a indium silylamide complex, which shares the same technical issues that are associated with antimony silylamide, were utilized to synthesize InSb CQDs.26Although the method yielded InSb CQDs with nearly monodisperse shape and size, the resulting CQDs exhibited broad excitonic transitions in their optical absorption spectrum. On the other hand, synthesis reports of InSb CQDs utilizing commercially available precursors, for example using a combination of InCh and Sb[NMe2]s or InCh and Sbh, yielded CQDs with broad excitonic transitions arising from non- stoichiometric nanocrystal compositions and less control over particle size.22’25’27’29’35 38Accordingly, it can be seen that the materials discussed above have significant potential for use in CQDs but further developments are generally necessary to help achieve this potential.

[0008] Summary

[0009] The invention is defined in the appended claims.

[0010] A method of synthesising AB colloidal quantum dots (CQDs) is provided, wherein A is Al, Ga, In or a mixture thereof, and B is P, As, Sb or a mixture thereof. The method comprises the steps of: a) reacting AX3 and BX3, where X is Cl, Br or I, with metal bis(trimethylsilyl)amides, wherein said metal is Li, Na, K or Cs, in at least one long chain amine at an elevated temperature to form A and B silylamides; b) further reacting the A and B silylamides with the at least one long chain amine to form triamides of A and B; c) injecting a reducing agent to balance the reactivity of the triamides of A and B to form nanocrystals of AB CQDs covered by organic ligands; and d) isolating and purifying the AB CQDs. Also provided are CQDs formed by such a method, and devices that incorporate such CQDs.

[0011] Also provided are colloidal quantum dots (CQDs) comprising InSb and covered by organic ligands comprising such as 1-dodecanthiol. The CQDs are configured to absorb infrared radiation within at least a portion of the wavelength range of > 1200 nm. The CQDs have a response time under UV-VIS-IR light excitation which is less than 1 ns. Brief Description of the Figures

[0012] Various implementations of the claimed invention will now be described by way of example only with reference to the following drawings.

[0013] Figure 1 comprises top section (a) and bottom section (b). The top section (a) depicts an example of a reaction scheme for the synthesis of InSb CQDs using commercially available precursors as described herein. The bottom section (b) shows a control experiment carried out in the absence of Li HMDS, while keeping all other reaction parameters constant.

[0014] Figure 1A presents an example of an X-ray diffraction pattern of InSb synthesized via the reaction scheme shown in Figure 1 , section (b).

[0015] Figure 2 comprises a top row of sections (a), (b) and (c) and a bottom row of sections (d), (e) and (f) and depicts a set of experimental results relating to the synthesis of Figure 1.

[0016] Figure 2A comprises a left section (a) showing an example of an XPS survey and a right section (b) showing examples of high-resolution XPS spectra.

[0017] Figure 3 comprises a section (panel) (a), together with a top row of sections (b), (c) and (d) and a bottom row of sections (e), (f) and (g). These sections present examples of experimental results which may be obtained for InSb CQDs such as synthesised according to the reaction scheme of Figure 1.

[0018] Figure 3A comprises sections (a)-(f), each section (plot) showing an example of a size distribution plot and associated statistics for the CQDs as described herein.

[0019] Figure 3B is formed of sections (a)-(c), where top section (a) provides a comparison of steady-state absorption of InSb CQDs stabilized with DDT and OA ligands, and sections (b)-(c) show TEM images of the InSb CQDs stabilized with (i) DDT ligands for section (b) and (ii) OA ligands for section (c).

[0020] Figure 4 comprises sections (plots) (a)-(f) including sections (a)+(d) (right), (b)+(e) (centre), and (c)+(f) (right). The plots represent (or are derived from) femtosecond transient absorption (fs-TA) spectra of DDT-stabilized InSb CQDs.

[0021] Figure 4A comprises sections (plots) (a)-(d) with sections (a)+(c) (left) and sections (b)+(d) (right). The plots represent (or are derived from) femtosecond transient absorption (fs-TA) spectra of DDT-stabilized and OA-stabilized InSb CQDs.

[0022] Figure 5 shows femtosecond transient absorption spectra of OA-stabilized InSb CQDs after exciting with a pump at 900 nm.

[0023] Figure 6 shows femtosecond transient absorption spectra of DDT-stabilized InSb CQDs after exciting with a pump at 700 nm at an early timescale (a) and at a longer timescale (b).

[0024] Figure 7 illustrates an example of the use of a ‘one-pot’ method for the synthesis of CQDs as described herein. Detailed Description

[0025] As discussed above, colloidal quantum dots (CQDs) are emerging materials for short-wave infrared (SWIR) photodetectors, which are technologically important for a broad array of applications. Unfortunately, the most developed SWIR CQD systems are Pb and Hg chalcogenides; their toxicity and regulated compositions limit their applications. An InSb CQD system is a potential environmentally friendly alternative whose bandgap may be tunable via quantum confinement across the SWIR spectrum. However, InSb CQDs are difficult to exploit, due to their complex syntheses and uncommon reactive precursors, which greatly hinder their application and study. A one-pot synthesis strategy is described herein using commercially available precursors to synthesize - under standard colloidal synthesis conditions - high-quality, size-tunable, monodisperse InSb CQDs. With this strategy, the large Bohr exciton radius of InSb can be exploited for tuning the bandgap of the CQDs over a wide range of wavelengths (such as 1250-1860 nm) within the SWIR region. Furthermore, by changing the surface ligands of the CQDs from oleic acid (OA) to 1-dodecanthiol (DDT), a ~50-fold enhancement was observed in the excited-state lifetime, efficient carrier multiplication, and slower carrier annihilation. This opens a wide range of SWIR applications to a promising class of Pb- and Hg-free CQDs.

[0026] As described herein, a convenient one-pot synthesis strategy is provided for obtaining high-quality InSb CQDs with well-defined excitonic transitions. This approach uses easily accessible reaction conditions and commercially available precursors. The synthesis mechanism has been investigated to uncover a new reaction pathway that appears to have escaped attention in the prior literature. The reaction proceeds via the in-situ formation of an indium- and antimony-silylamide complex, which upon further reacting with oleylamine forms triamides of indium and antimony. The present method balances the indium triamide and the antimony triamide reactivities with the reducing agent, which is important for obtaining a stoichiometric nanocrystal composition. The synthesis approach has produced monodisperse InSb CQDs with tunable sizes ranging from 4.2 ± 0.8 up to 8.5 ± 0.9 nm. The synthesized CQDs display a sharp excitonic transition in the SWIR region between 1250 and 1860 nm. Upon surface ligand modification, carrier multiplication was observed in the CQDs without requiring a complex core / shell structure - an indicator of their high optoelectronic surface quality.

[0027] The approach described herein therefore surmounts a major hurdle associated with the colloidal synthesis of high-quality InSb CQDs, thereby paving the way for their practical application in SWIR devices. In one example, a one-pot reaction was conducted to synthesize size-tunable, monodisperse InSb CQDs using commercially available precursors. The synthesis followed the scheme shown in Figure 1 (see in particular the top section (a)). Initially, lnCI3and Sbl3were reacted with lithium hextrimethylsilylamide (LiHMDS) in oleylamine at 120 °C to form antimony and indium silylamides ‘5’, which further react with oleylamine to form triamides of antimony and indium ‘o’. Unlike previous reports, the challenges associated with the storage and stability of metal silylamide were avoided because it was synthesized in situ.24 26Subsequently, lithium triethylborohydride (LiEt3BH) was injected at a lower temperature to initiate the reaction as described in more detail below. This method ensures a balanced reactivity for the triamides of indium and antimony, which is important for obtaining stoichiometric nanocrystals with the desired monodisperse sizes and shapes.26In the absence of LiHMDS (see the reaction scheme in Figure 1, in particular the bottom section (b)), the Sbl3and I nCI3precursors reacted to form a metal-OLA coordination complex ‘y’. This complex rapidly reacted with the LiEt3BH, leading to the formation of bulk InSb instead of CQDs.

[0028] Figure 1, top section (a), depicts an example reaction scheme of the present approach for the synthesis of InSb CQDs using commercially available precursors. The species labelled a, p, y and 5 are characterized and identified using1H NMR analysis. The formation of covalently bonded indium- and antimony-triamide species has an important role in the synthesis of InSb CQDs.

[0029] Figure 1, bottom section (b). shows a control experiment carried out in the absence of LiHMDS, while keeping all other reaction parameters constant. This led to the formation of bulk InSb due to the high reactivity of the metal-oleylamine coordination complex towards the reducing agent.

[0030] Figure 2 shows a set of experimental results relating to the above synthesis. These results are presented in three pairs of sections, namely a top pair of sections (a) and (b), a second pair of sections (c) and (d), and a third pair of sections (e) and (f). In overview, Figure 2, section (a), shows a PXRD pattern of InSb CQDs absorbing at 1560 nm, with an inset showing a HRTEM image of a single InSb QD displaying lattice fringes corresponding to (111) planes of zinc blende lattice of bulk InSb. Figure 2, section (b), shows 1H NMR spectra of free OLA in comparison with indium-triamide, and indium-oleylamine complex. The insets of Figure 2, section (b), highlight the chemical shift of amine protons peak of free oleylamine. Figure 2, section (c), shows 1H NMR spectra of free LiHMDS in deuterated chloroform in comparison with the by-product vapours formed after degassing a mixture of OLA + LiHMDS and lnCI3 + OLA + LiHMDS at 120 °C. Figure 2, section (d), shows a comparison of 1H NMR spectra of InSb CQDs stabilized with OA and DDT ligands. High- resolution XPS spectra of (i) Sb 3d core level and (ii) In 3d core level are shown in Figure 2, section (e), and Figure 2, section (f), respectively. The identification of the two plotted lines (QDs@OA and QDs@DDT) in section (e) applies also to section (f). Accordingly, Figure 2, section (a), shows the powder X-ray diffraction (PXRD) pattern of the product CQDs formed via the reaction scheme depicted in Figure 1, section (a). The PXRD is in good agreement with the cubic zinc blende reference of bulk InSb, consistent with previous reports on InSb CQDs.24The inset of Figure 2, section (a), presents a high- resolution transmission electron microscopy (HRTEM) image of InSb QDs revealing lattice fringes. An inter-plane distance of 3.71 A (0.371 nm) is assigned to the (111) planes of the zinc blende phase of InSb. In the absence of the LiHMDS complex, the reaction scheme in Figure 1 , section (b) results in the formation of bulk InSb along with an impurity phase.

[0031] Figure 1A presents an X-ray diffraction pattern of InSb synthesized via the reaction scheme shown in Figure 1 , section (b) to form bulk InSb along with a metallic In impurity phase. Therefore, it is important to uncover the mechanistic role of LiHMDS in controlling the size and shape of InSb CQDs.

[0032] Figure 2, section (b), shows the1H NMR spectra of oleylamine (OLA, C18H35NH2) in comparison with indium-triamide ‘a’ (obtained by reacting InCL with LiHMDS in OLA at 120 °C) and with the indium-oleylamine complex ‘y’ (obtained by reacting InCh and OLA at 120 °C). The three spectra differ in the position of amine protons. In the case of free oleylamine, the amine protons appear at 6 = 1.4 ppm (see Figure 2, section (b) and its inset). However, in the case of ‘o’, a slightly deshielded and broader peak at 5 = 1.67 ppm suggests indium is binding through covalent interactions. On the other hand, the amine peak is more deshielded in the coordination complex ‘y’ at 5 = 1.94 ppm, and is relatively intense, indicating the presence of the two amine protons in ‘y’ compared to one proton in ‘o’. It should be noted that the signal corresponding to free amine protons appears in ‘y’ (see the inset of Figure 2, section (b)), because an aliquot was taken for the1H NMR experiment. Unlike ‘y’, the indium-triamide ‘a’ is easily removable from OLA using acetonitrile as an antisolvent.

[0033] LiHMDS has a low boiling point of 80 °C, which prompted an investigation of the reaction byproducts that escape during degassing of the reaction at 120 °C. When LiHMDS reacted with OLA at 120 °C, the1H NMR analysis of the collected vapours showed a peak at 0.31 ppm, which is the typical chemical shift of pure LiHMDS in deuterated toluene (see Figure 2, section (c)). This observation shows that LiHMDS and OLA do not react when heated to 120 °C. However, upon mixing LiHMDS with InCh and OLA, and heating the reaction mixture to 120 °C in a vacuum, vigorous bubbling is observed at 80 °C due to the formation of volatile by-products. The vapors were collected in deuterated toluene, and the corresponding1H NMR spectra are shown in Figure 2, section (c). A new peak is observed at 4.9 ppm, in addition to the peak of silane-protons at 0.31 ppm, compared to free LiHMDS or OLA + LiHMDS. This new peak is attributed to the amide proton, due to its deshielded position and broad nature (see the inset of Figure 2, section (c)).39Therefore, InCh also acts as a catalyst by reacting with LiHMDS to form an indium silylamide complex, which further reacts with OLA to form indium triamide and the by-product bis(trimethylsilyl)amide (see the scheme in Figure 1, section (a)). It is noted that the molar stoichiometric ratio in Figure 1 , section (a), for LiHMDS:Sbl3:lnCh is 6:1:1 respectively for the synthesis of colloidal CQDs. However, because of the volatile nature of LiHMDS at 80 °C, slightly higher molar ratios of LiHMDS were used to ensure a complete reaction (as described in more detail below).

[0034] The synthesis of InSb CQDs was carried out in OLA as a solvent, which also acts as a weakly binding ligand. To enhance surface passivation and colloidal stability, two different ligands were used during the post-synthesis processing of CQDs. A comparison of optical properties for the two different ligand shells was also performed. Figure 2, section (d), illustrates a comparison of1H NMR spectra of InSb CQDs passivated with oleic acid (OA) and 1 -dodecanethiol (DDT) with the ligand structures and proton assignments shown in the insets. The results indicate that OLA significantly contributes to the ligand shell of the final product when the CQDs are stabilized with OA. However, a small contribution from OLA ligands was observed in DDT-stabilized CQDs, suggesting that DDT ligands have a stronger binding tendency to InSb CQDs than OA. The detailed photophysics of InSb CQDs stabilized with different ligands is described in further detail below.

[0035] We consider now the composition and local structure of InSb CQDs stabilized with OA and DDT ligands by employing X-ray photoelectron spectroscopy (XPS). Figure 2A comprises a left section (a) showing XPS survey spectra, and a right section (b) showing high-resolution XPS spectra of the Sb 4d core level of InSb CQDs stabilized with OA in comparison with CQDs stabilized with DDT ligands. The identification of the two plotted lines (lnSb@OA and lnSb@DDT) in section (a) applies also to section (b). The additional peaks in high-resolution XPS spectra (see section (b)) are indicative of antimony oxide formation in OA-stabilized CQDs. Regarding section (a) of Figure 2A showing the XPS survey spectra of InSb CQDs capped with OA and DDT ligands, it is noted that both ligands preserve the nearly ideal stoichiometric ratio of 1:1 for ln:Sb respectively.

[0036] High-resolution XPS spectra of the core levels of Sb 3d and In 3d are shown in Figure 2, sections (e) and (f), along with the corresponding peak assignments. InSb CQDs capped with DDT ligands exhibit no peaks corresponding to indium and antimony oxide formation, whereas antimony oxide peaks contribute significantly in the case of OA-stabilized CQDs (Figure 2, section (e)). The formation of metal-oxide in OA-stabilized CQDs is further confirmed by the broadening of In 3d core-level peaks compared to DDT-stabilized QDs (see Figure 2, section (f)), and by the high-resolution XPS of Sb 4d core level spectra in Figure 2A, section (b). It is noted that the presence of metal-oxide peaks in OA-stabilized CQDs is expected since the dangling bonds on the surface of CQDs are passivated by OA ligands via an RCOO' functional group. It is discussed below how this OA, due to the more ionic character of metal-oxide bonds compared to metal-sulfide bonds on the surface of CQDs, leads to the presence of numerous trap states in OA-stabilized InSb CQDs.

[0037] Figure 3 comprises a left section or panel (a) together with a top row of sections (b), (c) and (d) and a bottom row of sections (e), (f) and (g). These sections are used to present examples of experimental results obtained for InSb CQDs. Thus Figure 3, section (panel) (a), depicts steady-state optical absorption spectra of InSb CQDs exhibiting sharp excitonic features in the SWIR window 1250 - 1860 nm. The optical bandgap is tailored by varying the concentration of the reducing agent and by size-selective precipitation. Increasing the bandgap from 1250 nm to 1860 nm (via the intermediate values) moves the absorption peak to a longer wavelength and increases the absorbance at wavelengths beyond the absorption peak.

[0038] Sections (b)-(g) of Figure 3 present TEM images with insets (top right) showing corresponding HRTEM images of InSb CQDs for optical bandgaps shown in panel (a). The average size of CQDs falls in the range of 4.2 ± 0.8 to 8.5 ± 0.9 nm. For reference, the scale bars located in the HRTEM insets of Figure 3, sections (b)-(g), are all 5 nm length.

[0039] More particularly, Figure 3, section (a), shows an example of the steady-state optical absorption spectra of InSb CQDs with well-defined excitonic features in the SWIR region 1250 - 1860 nm (0.99 - 0.67 eV). The size tuning of InSb CQDs is achieved by varying the concentration of the reducing agent and by size-selective precipitation.24Typically, InSb CQDs with absorption in the wavelength window of 1250-1440 nm are obtained from one reaction by using a higher concentration of reducing agent, 6.4 equiv. of LetsBH, compared to InC , and then carrying out size-selective precipitation. On the other hand, larger sizes of InSb CQDs absorbing in the wavelength range of 1560-1860 nm were obtained by using a lesser amount of 4.8 equiv. of LiEtsBH, followed by size-selective precipitation. Previous studies have reported excitonic features with a longest wavelength of 1660 nm, corresponding to a size of 6.3 nm of InSb CQDs.24The present approach may be used to produce InSb CQDs having excitonic features at a longer wavelength of 1860 nm, corresponding to a size of 8.5 ± 0.9 nm.

[0040] The TEM images were captured to identify the shape and size of the particles. Figure 3, sections (b)-(g), shows TEM images for the optical bandgaps depicted in Figure 3, section (a). The insets in each TEM image display HRTEM images of individual quantum dots with clear lattice fringes suggesting high crystallinity observed for all sizes of CQDs. The particles exhibit a quasi-spherical shape, similar to prior reports on colloidal InSb CQDs.24’27Notably, a narrow size distribution was achieved for all sizes. Figure 3A comprises sections (a)-(f), with sections (a)-(c) on a top row and sections (d)-(f) on a bottom row; each section showing a size distribution plot and associated statistics for the CQDs. In particular, Figure 3A presents size distribution histograms of InSb CQDs absorbing at 1250 nm (a), 1310 nm (b), 1440 nm (c), 1560 nm (d), 1730 nm (e), and 1860 nm (f).

[0041] It was further observed that the particle shape, size, and optical absorption of InSb CQDs stabilized with either DDT or OA ligands, for CQDs obtained from the same batch of reaction, were unaffected by the type of ligand present on the surface (see Figure 3B below). Both OA and DDT-passivated InSb CQDs demonstrate excellent colloidal stability, with no precipitation observed for a few months, whether stored inside or outside the glove box.

[0042] Figure 3B is formed of three sections. Figure 3B, section (a) (top), provides a comparison of steady-state absorption of InSb CQDs stabilized with DDT and OA ligands obtained from the same batch of reaction. The lowest excitonic transition emerges at the same wavelength for both samples suggesting that the optical bandgap is not affected by post-synthesis ligand passivation. (In Figure 3B section (a), the separation at around 1600 nm has lnSb@DDT higher than INSb@OA).

[0043] Figure 3B, sections (b) (lower left) and (c) (lower right), show TEM images of InSb CQDs stabilized with (i) DDT ligands for section (b) and (ii) OA ligands for section (c). Both the samples, as depicted in sections (b) and (c), show similar sizes and shapes of CQDs, despite the difference in respect of the surface ligands at the surface of the CQDs.

[0044] Figure 4 comprises pairs of sections (a) top and (d) bottom; (b) top and (e) bottom; and (c) top, f (bottom). The plots represent (or are derived from) femtosecond transient absorption (fs-TA) spectra of DDT-stabilized InSb CQDs after exciting with a pump at 900 nm. Section (a) relates to an early or short timescale whereas section (b) relates to a long timescale. Section (c) provides a comparison of time-resolved bleach recovery profiles of QDs stabilized with OA and DDT ligands at 1380 nm in response to 900 nm optical excitation. Fs-TA spectra are shown at an early timescale in the case of (i) DDT-stabilized for section (d), and (ii) OA-stabilized for section (e), for InSb CQDs following 400 nm excitation. Section (f) depicts kinetics at 1400 nm at different pump wavelengths.

[0045] These pump-probe experiments were carried out to understand the charge carrier dynamics of InSb CQDs passivated with different surface ligands. Figure 4, sections (a) and (b) show the temporal evolution at early and late timescales respectively of DDT-stabilized InSb CQDs in response to 900 nm excitation. The transient spectra at each time delay comprise a negative absorption change known as ground state bleach (GSB) due to the depletion of the ground state population at band filling. The GSB peak maxima at 1380 nm are consistent with the steady-state absorption (see Figure 3, section (a), as discussed above). However, under the same conditions, the OA-stabilized CQDs showed a weak GSB and photoinduced absorption (PIA) which recovered within a sub-ten picosecond timescale. This is illustrated in Figure 5, which shows femtosecond transient absorption spectra of OA- stabilized InSb CQDs after exciting with pump at 900 nm. This diagram plots four lines, the first corresponding to -1 ps is generally flat across the diagram at level zero. The remaining three lines correspond to 200 fs, 2 ps and 10 ps. At shorter wavelengths, the ordering from zero upwards is -1 ps, 10 ps, 2 ps and 200fs; conversely at longer wavelengths (at around 1200 nm and longer) the ordering from zero downwards is -1 ps, 10 ps, 2 ps and 200fs. The plotted lines then return (approximately) to zero around 1500 nm.

[0046] A comparison of GSB for InSb CQDs stabilized with DDT and with OA is presented in Figure 4, section (c). A bi-exponential time constant is required to achieve the best fit, with time constants of 20 and 600 ps for DDT-stabilized CQDs, and 2 and 12 ps for OA-stabilized InSb CQDs. DDT-stabilized CQDs show an improved average lifetime of 194 ± 18 ps, significantly longer than the average lifetime of OA-stabilized CQDs, 4.5 ± 0.25 ps. Previous studies reported fast excited state decay dynamics with a lifetime of the order of a few picoseconds for InSb CQDs, similar to our OA-stabilized CQDs40. This ultrafast (-few ps) lifetime of the excited state may arise due to non-radiative relaxation through the defect states with OA-stabilization. The presence of oxide formation on the surface of CQDs has been confirmed through XPS measurements (see Figure 2, sections (e) and (f)), which is a susceptibility with OA-stabilization. On the other hand, the longer lifetime of the excited state in DDT-stabilized CQDs suggests less contribution to the excited state dynamics from the defect states (indicating an oxide-free surface).

[0047] To obtain a deeper understanding of carrier dynamics, carrier multiplication was studied in InSb CQDs after exciting them at different pump wavelengths. It is noted that 400 nm and 700 nm represent energy levels which are energetically greater than twice the bandgap ‘Eg’ (1440 nm, 0.86 eV) of InSb CQDs used for the carrier multiplication.41The time-resolved spectra at 400 nm (-3.6* Eg) and 700 nm (~2xEg) pump excitation are depicted in Figure 4, section (d), and Figure 6, respectively. Unlike the 900 nm excitation, the 400 nm pump excitation results in an instantaneous formation of PIA, instead of GSB, which quickly transforms to GSB that we attribute to multiple exciton generation. (Note that all measurements were performed at a constant pump fluence of 1.8 ± 0.2 pJcm-2). Carrier multiplication is highly efficient at 400 nm pump excitation; however, a weaker amplitude for carrier multiplication is observed at 700 nm excitation wavelength - see Figure 6, which shows femtosecond transient absorption spectra of DDT-stabilized InSb CQDs after exciting with a pump at 700 nm at an early timescale (a), and at a longer timescale (b). This corresponds to the threshold pump energy required for carrier multiplication in InSb CQDs absorbing at 1440 nm. Interestingly, OA-stabilized InSb CQDs also undergo carrier multiplication, Figure 4, section (e); however, they convert to GSB faster than DDT-stabilized InSb CQDs.

[0048] In Figure 6(a) the plot for -1 ps is flat at zero. The remaining plots correspond to 200 fs, 410 fs, 900 fs and 2 ps and respectively show an increasingly deep minimum at 1400 nm. The plots in Figure 6(b) also show a plot for 2 ps (which corresponds to the 2 ps line in Figure 6(a)). The remaining plots correspond to 5 ps, 10 ps, 35 ps, 225 ps, 1 ns which respectively show an increasingly shallow minimum at 1400 nm. In effect therefore the minimum at 1400 nm increases in depth for times up to 2 ps and then decreases the depth at higher times.

[0049] By way of further investigation, a comparison was performed in relation to the lifetime of the multiple carriers by probing the kinetics at 1400 nm following different pump excitations - see Figure 4, section (f). It can be observed that the GSB exhibits biexponential recovery when the DDT-stabilized InSb is excited close to the bandgap, i.e., at 900 nm. However, the same kinetics behave differently when pumped at 400 nm (3.5* Eg). The kinetics at 1400 nm, upon 400 nm pump excitation, show a positive signal that quickly decays to zero with a time constant of 0.41 ps and converts to GSB. The GSB then follows the same recovery pathway as observed with that of 900 nm pump excitation, and the best fit is achieved with the time constants of 20 ps and 495 ps. The fast time constant (0.41 ps) of the decay of the positive signal refers to the multiple carrier annihilation time. To understand better, a comparison was made of the kinetics of OA-stabilized InSb CQDs at 400 nm pump excitation in the same plot - see Figure 4, section (f). Similarly to DDT- stabilized InSb CQDs, the OA-stabilized CQDs show a positive signal that undergoes GSB; however, carrier annihilation time is faster (0.21 ps) compared to DDT-stabilized InSb CQDs. Moreover, there is a fast GSB recovery that relates to the presence of defect states. Efficient carrier multiplication and slower carrier annihilation in DDT-stabilized InSb CQDs without core / shell modification suggest that thiol ligand capped CQDs possess the quality required for optoelectronic applications.42’43

[0050] Note that in Figure 4(a) the ordering of the lines within the minimum (from top down) is -1 ps, 200fs, 380fs, 610fs, 1.5ps and 2.2ps (with the last two generally coincident with one another). In Figure 4(b) the ordering of the lines within the minimum (from top down) is -1ps, 2ns, 900ps, 205ps, 60ps, 20ps, 8ps, and 3ps. In Figure 4(d) the -1ps line is horizontal in the middle and the ordering of the remaining lines (from top down) is 200fs, 750fs, 1ps, 1.5ps and 2.8ps. In Figure 4(e) the -1ps line is horizontal in the middle and the ordering of the remaining lines (from top down) is 200fs, 450fs, 620fs, 1.2ps and 2ps. In Figure 4(f), the -1ps line is horizontal in the middle and the ordering of the other lines (from top down, for a time delay of 2-4ps) is QA_400nm, DDT 400nm and DDT 900.

[0051] Figure 4A comprises sections (a)-(d) with section (a) (top) presented together with section (c) (bottom) and section (b) (top) presented together with section (d) (bottom). The plots represent (or are derived from) femtosecond transient absorption (fs-TA) spectra of DDT-stabilized and OA-stabilized InSb CQDs after exciting with a pump at 900 nm. Section (a) relates to ground state bleach whereas section (b) provides a comparison of time- resolved bleach recovery profiles of QDs stabilized with OA and DDT ligands at 1380 nm in response to 900 nm optical excitation. Section (c) relates to GSB dynamics at varying pump excitation energies after maintaining the constant average number of photons that are absorbed per CQD per pulse excitation i.e. <N> = 0.26; the inset shows absorbance spectra and corresponding pump excitation energies used to study the dynamics. Section (d) relates to GSB dynamics at a shorter timescale for <N> = 1.05, and at varying pump excitation energies.

[0052] These pump-probe experiments were carried out to understand the charge carrier dynamics of InSb CQDs passivated with different surface ligands. Figure 4A(a) shows the femtosecond transient absorption 2D contour plot for DDT-stabilized InSb CQDs up to 1 ns time windows at various excitations. The average number of photons absorbed per CQD per pulse excitation, denoted as <N>, is fixed to be 0.26 for all excitation wavelengths.

[0053] Figure 4A(a) represents a contour plot (heat map) with an associated scale bar to the right. There is a minimum (negative) portion bottom centre-right of the drawing, largely surrounded by an intermediate region, and a maximum (positive) portion mainly to the top of the drawing. The 2D contour plot (heat map) depicts a negative absorption change, known as ground state bleach (GSB) due to the depletion of the ground state population at band filling. The GSB peak maxima at -1360 nm is consistent with the steady-state absorption. A comparison of GSB for InSb CQDs stabilized with DDT and OA ligands is presented in Figure 4A(b). A bi-exponential time constant is required to achieve the best fit, with time constants of 9.3 ± 0.6 (63%) and 193 ± 11.5 (37%) ps for DDT-stabilized CQDs, and 2 ± 0.1 (70%) and 12 ± 0.6 (30%) ps for OA-stabilized InSb CQDs. DDT-stabilized CQDs show an improved average lifetime of 77.3 ± 4.5 ps, significantly longer than the average lifetime of OA-stabilized CQDs, 5.0 ± 0.25 ps.

[0054] Previous studies reported fast excited state decay dynamics with a lifetime of the order of a few picoseconds for InSb CQDs, similar to the OA-stabilized CQDs described herein.40This ultrafast (-few ps) lifetime of the excited state may arise due to non-radiative relaxation through the defect states with OA-stabilization. The presence of oxide formation on the surface of CQDs has been confirmed through XPS measurements (see Figure 2, sections (e) and (f)), which is a susceptibility with OA-stabilization. On the other hand, the longer lifetime of the excited state in DDT-stabilized CQDs suggests less contribution to the excited state dynamics from the defect states (indicating an oxide-free surface).

[0055] These results motivated us to further investigate the carrier dynamics at high excitation energies and the potential for carrier multiplication in DDT-stabilized InSb CQDs. Figure 4A(c) presents GSB kinetics, normalized at 200 ps for various excitation energies, after maintaining a constant <N> value of 0.26 by varying the pump flux in correlation with the absorption (see the insert of Figure 4A(c)). The pump excitation energies are varied in the range of 1.49 - 3.46 eV, after normalizing the pump energy (Ehr) with the bandgap (Eg) of 0.92 eV (-1360 nm) for the probed InSb CQDs.

[0056] In each of Figure 4A(c) and 4A(d), there are plots of five different lines, each line having a respective value for EhV / Egof 1.49, 1.82, 2.14, 2.69 and 3.46. The lines are separated in order with the line corresponding to 1.49 being the lowest line and the line corresponding to 3.46 being the highest line. Figure 4A(c) further shows an inset graph having five lines representing the same five values for EhV / Eg, with a corresponding ordering of Ehv / Egfrom 1.49 (left) to 3.46 (right).

[0057] Interestingly, the magnitude of differential absorption corresponding to GSB and the recombination rate both increase at excitation energies > 2.14 eV, which is energetically greater than twice the Eg(0.92 eV) of the probed InSb CQDs for carrier multiplication. The fitted parameters for bleach dynamics reveal the decay time constants decrease significantly at high pump excitation energies > 2.14 eV for the same <N> = 0.26. This enhancement in the recombination rate at high excitation energy for the same <N> could be attributed to carrier multiplication. Upon further increasing the <N> value to 1 .05, the GSB dynamics, normalized at a shorter timescale of 3 ps for different pump excitation energies, are plotted in Figure 4A(d). Clearly, at pump excitation energies > 2.14 eV, the carrier dynamics result in an instantaneous formation of PIA instead of GSB, which quickly transforms to GSB. We ascribe this PIA to the carrier multiplication, which is observed above threshold excitation energies that are twice the bandgap.

[0058] Carrier multiplication is highly efficient at 3.46 eV pump excitation; however, a weaker amplitude for carrier multiplication is observed at 2.14 eV excitation wavelength. Consequently, we suggest 2.14 eV may correspond to the threshold pump energy required for carrier multiplication in InSb CQDs absorbing at -1360 nm. Interestingly, OA-stabilized InSb CQDs also undergo carrier multiplication (Figure 5); however, there is a weaker amplitude of PIA compared to the DDT-stabilized CQDs. Moreover, a faster carrier annihilation time of (0.21 ps) for PIA, along with a faster GSB recovery in OA-stabilized CQDs, relates to the interference from the defect states. In contrast, efficient carrier multiplication and slower carrier annihilation in DDT-stabilized InSb CQDs without core / shell modification suggest that thiol ligand-capped CQDs possess the quality required for optoelectronic applications.

[0059] As described herein, it is now known that the excited state dynamics of InSb CQDs are strongly affected by the type of organic ligands present on the surfaces. This is due to the fact that the bonding in InSb has a strong covalent character;29therefore, it is anticipated that the covalent interactions on the surface should lead to fewer surface traps. Interestingly, DDT ligands form metal-thiol-type bonds that are more covalent in nature compared to the metal-oxygen bonds formed with OA ligands. It is possible that bidentate sulfide ligands may further reduce the surface traps, potentially leading to an enhanced carrier lifetime. Further experiments may be performed to understand full details of the impact of covalently bonded ligands on the optical and electronic properties of InSb CQDs.

[0060] More information about the above experiments is provided below. It will be appreciated that this information is by way of illustration and example, and not by way of limitation. In particular, it will be understood that the information represents potential implementations, but other implementations may use different techniques, conditions, suppliers, concentrations, procedures and materials as appropriate.

[0061] Chemicals. Indium chloride (InCh, 99.999%), antimony iodide (Sbh, 98%), antimony chloride (SbCh, 99.99%), lithium bis(trimethylsilyl)amide (UN(Si(CH3)3)2, 97%), oleylamine (OLA, >98%), dioctyl ether (DOE, 99%), 1 -dodecanethiol (DDT, >98%), oleic acid (OA, 98%), acetonitrile (99.8%), toluene (99.5%), methanol (99%), tetrachloroethylene (TOE, 99%) were purchased from Sigma-Aldrich (part of Merck), see https: / / www.sigmaaldrich.com / , and used without purification. Lithium triethylborohydride (Li(C2Hs)3BH, 1.0 M in tetrahydrofuran) was purchased from Beantown Chemical (part of Thomas Scientific), see https: / / www.thomassci.com / scientific-supplies / Beantown-Chemical. Colloidal synthesis of InSb. Colloidal synthesis of InSb was carried out under inert conditions (N2 atmosphere) using commercially available precursors. Typically, 0.5 mmol InCh, 0.5 mmol Sb , 4.2 mmol LiHMDS, and 3 mL TOP were placed in a reaction flask inside a glove box, and mounted on a Schleck line without exposing the contents to air. Then 12 mL of OLA was injected at room temperature and the reaction mixture was slowly heated to 120 °C under a vacuum while stirring continuously. At around 80 °C, vigorous bubbling was observed in the vacuum due to the formation of volatile byproducts. The reaction was kept overnight under a vacuum at 120 °C for efficient degassing. After degassing, the appearance of the solution was transparent and greyish. The temperature was cooled down to 40 °C and LiEtsBH (2M in dioctyl ether, prepared by transferring commercial LiEtsBH in THF into dioctyl ether slowly at 80 °C under vacuum) 1.2 - 1.6 mL was injected depending upon the targeted sizes. The reaction was heated slowly to 240 °C (more generally to a temperature of 240 °C ± 20 °C, optionally ±10 °C). The total heating time was fixed to 40 minutes from 40 - 240 °C (more generally fixed to 40 minutes ± 15 minutes). The reaction was left undisturbed for 15 minutes before being cooled down with blowing cold air. The injection of 3 mL OA in 20 mL toluene at room temperature was followed by the purification of the OA-stabilized CQDs. For DDT-stabilized CQDs, 2 mL DDT in 15 mL toluene was injected followed by purification.

[0062] Size-selective washing. Purification for OA-stabilized CQDs was carried out in accordance with the following. First, the crude solution was centrifuged at 8000 rpm to remove any gel- like substances. The resulting supernatant was washed by adding acetonitrile in small amounts (1-2 mL at a time) to obtain size-selective precipitation of the CQDs. For the purification of DDT-stabilized CQDs, small amounts of methanol (2 - 3 mL) were directly added to a crude mixture, and a size-selective precipitate of CQDs was collected and further processed. Subsequent washing steps are similar for both cases. The precipitated CQDs were dispersed in a small volume (1-2 mL) of toluene and precipitated again using small amounts of acetonitrile (~1 mL). The collected precipitate was then dispersed in tetrachloroethylene for optical characterization.

[0063] An alternative washing procedure for OA-stabilized CQDs was carried out as follows. The crude solution was mixed with 20 mL methanol and centrifuged quickly. Methanol readily dissolved the gel and the precipitate obtained contained the InSb CQDs. This precipitate was then dispersed in fresh toluene (~20 mL) followed by the addition of half the volume (~10 mL) of a mixture of OA:OLA in the ratio of 2:1 respectively. The precipitation was then carried out by adding small amounts of acetonitrile (~1-2 mL) to obtain size selective CQDs.

[0064] TEM measurements. TEM images were obtained by using a Titan ST 300 KeV TEM instrument from Field Electron and Ion Company (FEI), part of Thermo Fisher Scientific, see https: / / corporate.thermofisher.com / us / en / index.html, to examine the crystallographic structure, size distribution, and morphology of InSb CQDs.

[0065] X-ray photoelectron spectroscopy (XPS). XPS studies were carried out using a Kratos Axis Ultra DLD spectrometer (see https: / / www.kratos.com / products) equipped with a monochromatic Al Ka X-ray source (hv = 1486.6 eV) operating at 150 W, a multi-channel plate and delay line detector under a vacuum of ~10-9mbar. All spectra were recorded using an aperture slot of 300 x 700 pm. Survey spectra were collected using a pass energy of 160 eV and a step size of 1 eV. A pass energy of 20 eV and a step size of 0.1 eV were used for the high-resolution spectra. For XPS measurements, the synthesis and washing of CQDs were carried out in an inert atmosphere without exposing CQDs to oxygen or moisture at any stage of processing. XPS sample preparation was carried out inside the glove box using an air-sensitive sample holder.

[0066] Powder X-ray diffraction. A PXRD pattern was recorded on a Bruker AXS D8 diffractometer using Cu Karadiation (see https: / / www.bruker.com / en.html).

[0067] Surface Characterization.1H NMR analysis was carried on a Bruker AVANCE HI-600 spectrometer and processed by Topspin 2.1 (also from Bruker).

[0068] Optical Characterization. The steady-state absorption spectra of CQDs dispersed in TCE solvent were obtained using a Cary 6000i UV-vis-NIR spectrophotometer (Cary is part of Agilent, see https: / / www.agilent.com / ). Femtosecond transient absorption (fs-TA) measurements. The fs-TA spectra on timescales of 0.1 ps to 6 ns were collected by employing a multipass amplified Trsapphire laser (800 nm laser pulses of 7 mJ / pulse with a -100 fs pulse width at a repetition rate of 1 kHz, Astrella from Coherent - see https: / / content.coherent.com / legacy- assets / pdf / COHR_Astrella_DS_0320_1.pdf), coupled with Helios spectrometers from Ultrafast systems, see https: / / ultrafast.systems / products / spectrometers-accessories / helios / .

[0069] Excitation pump pulses at 900, 700, and 400 nm (or at 900, 740, 600, 500 and 390 nm) were obtained by passing a fraction of an 800 nm beam into a spectrally tunable (240-2600 nm) optical parametric amplifier (Newport Spectra-Physics, see https: / / www.newport.com / ). A neutral density (ND) filter was used to adjust the pump fluence of the excitation laser source to avoid multiple charge carrier generation. The white light probe pulse was generated by passing another fraction of the 800 nm pulses through a sapphire crystal. Before white light generation, the amplified 800 nm pulses were passed through a motorized delay stage. Depending on the movement of the delay stage, the transient species were detected following excitation at different time scales. The white light was split into two beams (denoted “signal” and “reference”) and focused into two optical fibers to improve the signal-to-noise ratio. The excitation pump pulses were spatially overlapped with the probe pulses over the samples after passing through a synchronized mechanical chopper (having a frequency of 500 Hz), which blocked the alternative pump pulse. The absorption change (AA) was measured with respect to the time delay and wavelength (A). All spectra were averaged over a period of 2 s for each time delay.

[0070] Figure 7 illustrates an example of the use of a ‘one-pot’ method for the synthesis of CQDs as described herein, having regard also to Figure 1, section (a). The one-pot synthesis may be used, for example, to provide highly tunable short-wave infrared quantum dots exhibiting carrier multiplication.

[0071] The left-hand portion of Figure 7 shows a reaction chamber or flash for holding halides of Sb and In along with lithium bis(trimethylsilyl)amides (such as lithium hextrimethylsilylamide - LiHMDS) and at least one long chain amine such as OLA. The reaction chamber is heated to an elevated temperature (i.e. greater than room temperature), for example to a temperature of 120 °C or more generally a temperature in the range of 120±AT °C, where AT = 40 °C, optionally wherein AT = 20 °C. The halides of Sb and In are consumed in the reaction flask to form In and Sb silylamides, which further react with the oleylamine (OLA) to form triamides of Sb and In. Generating the In and Sb silylamide complexes in situ within the one-pot reaction chamber sidesteps the challenges generally associated with the storage and stability of metal silylamide. A reducing agent, such as lithium triethylborohydride, is then added into the reaction chamber to balance the reactivity of the triamides of Sb and In to form nanocrystals of InSb CQDs covered by organic ligands. The InSb CQDs may then be isolated and further processed as appropriate, for example, to encapsulate the CQCs into a protective film.

[0072] The synthesis shown in Figure 7 is well-suited to the formation of InSb colloidal quantum dots but may be employed for the formation of colloidal quantum dots formed from other materials. Accordingly, it will be understood that the above details shown in Figure 7 and further described herein are provided by way of illustration rather than limitation, and the skilled person will be able to adapt the approach of Figure 7 to other syntheses according to their particular circumstances.

[0073] The CQDs described herein have optical (infrared) properties such that they may be used to form a photoactive layer. Such a photoactive layer may, for example, have a thickness in the range from 50nm to 500nm. The photoactive layer may be incorporated into an infrared image sensor to convert incoming infrared light into an electrical signal representing an image and / or to convert incoming infrared light into an electrical signal representing a binary signal or a spectrum. The image sensor or photodetector may have a response time to incoming infrared radiation of below 100 ns, preferably below 10 ns, preferable below 1 ns. Such an image sensor or photodetector may include a stack of layers comprising (for example): a CMOS substrate, a bottom electrode, a hole transport layer, the photoactive layer, an electron transport layer, and a top electrode. The infrared image sensor may be used, for example, in a device comprising a machine vision camera, a camera for robotics, virtual reality goggles, mixed reality goggles, a smartphone, an automotive camera, an eye interactive camera, or a set of contact lenses. The infrared photodetector may be used, for example, in a device comprising a smartphone or an eye interactive camera.

[0074] Accordingly, a one-pot synthesis method is described herein for obtaining high- quality and monodisperse InSb CQDs using commercially available precursors and easily accessible reaction conditions. The synthesis may utilize LiHMDS to form in situ indium and antimony triamides which have similar reactivity with a LiEtsBH reducing agent, leading to monodisperse, stoichiometric InSb CQDs. The product InSb CQDs adopt a cubic zinc blende lattice structure, similar to bulk InSb. TEM studies suggest that the particles are highly crystalline and quasi-spherical in shape. The InSb CQDs steady-state absorption spectra show a well-defined excitonic transition in the SWIR region. The CQDs show a size- tunable bandgap falling within the absorption window of 1250-1860 nm for sizes ranging from 4.2 ± 0.8 to 8.5 ± 0.9 nm. Transient pump-probe studies suggest OA-stabilized CQDs have a short-lived excited state that lasts for a few picoseconds likely due to surface defects, which is consistent with previous reports. In contrast, DDT-stabilized CQDs showed a nearly 50-fold enhancement in the excited state lifetime. In addition, significantly more efficient carrier multiplication and slower carrier annihilation was observed in the case of DDT- stabilized CQDs. Unlike OA, the DDT ligands bind through stronger covalent interactions, making them the preferred choice for stabilizing InSb CQDs. Taken together, these findings indicate that such a one-pot synthesis strategy yields InSb CQDs of high optoelectronic quality, which is important for exploring their fundamental properties and exploiting their use for SWIR device applications.

[0075] In conclusion, while various implementations and examples have been described herein, they are provided by way of illustration, and many potential modifications will be apparent to the skilled person having regard to the specifics of any given implementation. Accordingly, the scope of the present case should be determined from the appended claims and their equivalents. Furthermore, unless the context clearly indicates to the contrary, it is specifically disclosed herein that the features of any independent claim and / or its associated dependent claims may be combined with the features of any other independent claim and / or its associated dependent claims (irrespective of whether such a combination is explicitly claimed, since the claims are used to determine the scope of protection, not the overall disclosure of the application).

[0076] REFERENCES

[0077] (1) Marc, P. H.; Douglas, S. M. Overview of SWIR Detectors, Cameras, and Applications. In Proc.SPIE, 2008; Vol. 6939, p 69390I.

[0078] (2) Karim, A.; Andersson, J. Y. Infrared Detectors: Advances, Challenges and New Technologies. IOP Conf. Ser.: Mater. Sci. Eng. 2013, 51 , 012001.

[0079] (3) Lu, H.; Carroll, G. M.; Neale, N. R.; Beard, M. C. Infrared Quantum Dots: Progress, Challenges, and Opportunities. ACS Nano 2019, 13, 939-953.

[0080] (4) Rogalski, A. Infrared Detectors: An Overview. Infrared Phys. Technol. 2002, 43, 187-210.

[0081] (5) Rode, D. L. Electron Mobility in Direct-Gap Polar Semiconductors. Phys. Rev. B 1970, 2, 1012- 1024.

[0082] (6) Del Alamo, J. A. Nanometre-Scale Electronics with 111— V Compound Semiconductors. Nature 2011 , 479, 317-323.

[0083] (7) Krauss, T. D.; Peterson, J. J. Electronic Structure and Optical Transitions in Colloidal Semiconductor Nanocrystals. In Colloidal Quantum Dot Optoelectronics and Photovoltaics, Sargent, E. H., Konstantatos, G. Eds.; Cambridge University Press, 2013; pp 59-86.

[0084] (8) Levinshtein, M.; Rumyantsev, S. L.; Shur, M.; Scientific, W. Handbook Series on Semiconductor Parameters: Ternary and quaternary lll-V compounds; World Scientific Publishing Company, 1999.

[0085] (9) Efros, A. L.; Rosen, M. Quantum Size Level Structure of Narrow-Gap Semiconductor Nanocrystals: Effect of Band Coupling. Phys. Rev. B 1998, 58, 7120-7135.

[0086] (10) Kagan, C. R.; Lifshitz, E.; Sargent, E. H.; Talapin, D. V. Building Devices from Colloidal Quantum Dots. Science 2016, 353, aac5523.

[0087] (11) He, M.; Xu, Z.; Zhang, S.-W.; Zhang, M.; Wu, C.; Li, B.; Li, J.; Wang, L.; Zhao, S.; Kang, F.; et al. Colloidal InSb Quantum Dots / Organic Integrated Bulk Heterojunction for Fast and Sensitive NearInfrared Photodetectors. Adv. Photon. Res. 2022, 3, 2100305.

[0088] (12) De lacovo, A.; Venettacci, C.; Colace, L.; Scopa, L.; Foglia, S. PbS Colloidal Quantum Dot Photodetectors Operating in the Near Infrared. Sci. Rep. 2016, 6, 37913.

[0089] (13) Chu, A.; Greboval, C.; Prado, Y.; Majjad, H.; Delerue, C.; Dayen, J.-F.; Vincent, G.; Lhuillier, E. Infrared Photoconduction at the Diffusion Length Limit in HgTe Nanocrystal Arrays. Nat. Commun. 2021 , 12, 1794.

[0090] (14) Mir, W. J.; Assouline, A.; Livache, C.; Martinez, B.; Goubet, N.; Xu, X. Z.; Patriarche, G.; Ithurria, S.; Aubin, H.; Lhuillier, E. Electronic Properties of (Sb;Bi)2Te3 Colloidal Heterostructured Nanoplates Down to the Single Particle Level. Sci. Rep. 2017, 7, 9647.

[0091] (15) Bennett, B. R.; Magno, R.; Shanabrook, B. V. Molecular Beam Epitaxial Growth of InSb, GaSb, and AlSb Nanometer-Scale Dots on GaAs. Appl. Phys. Lett. 1996, 68, 505-507.

[0092] (16) Barrigon, E.; Heurlin, M.; Bi, Z.; Monemar, B.; Samuelson, L. Synthesis and Applications of lll-V Nanowires. Chem. Rev. 2019, 119, 9170-9220.

[0093] (17) Badawy, G.; Gazibegovic, S.; Borsoi, F.; Heedt, S.; Wang, C.-A.; Koelling, S.; Verheijen, M. A.; Kouwenhoven, L. P.; Bakkers, E. P. A. M. High Mobility Stemless InSb Nanowires. Nano Lett. 2019, 19, 3575-3582. (18) Ackerman, M. M.; Tang, X.; Guyot-Sionnest, P. Fast and Sensitive Colloidal Quantum Dot MidWave Infrared Photodetectors. ACS Nano 2018, 12, 7264-7271.

[0094] (19) Tan, C. L.; Mohseni, H. Emerging Technologies for High Performance Infrared Detectors. Nanophotonics 2018, 7, 169-197.

[0095] (20) Tang, X.; Ackerman, M. M.; Chen, M.; Guyot-Sionnest, P. Dual-Band Infrared Imaging Using Stacked Colloidal Quantum Dot Photodiodes. Nat. Photon. 2019, 13, 277-282.

[0096] (21) Guzelian, A. A.; Banin, U.; Kadavanich, A. V.; Peng, X.; Alivisatos, A. P. Colloidal Chemical Synthesis And Characterization of InAs Nanocrystal Quantum Dots. Appl. Phys. Lett. 1996, 69, 1432- 1434.

[0097] (22) Evans, C. M.; Castro, S. L.; Worman, J. J.; Raffaelle, R. P. Synthesis and Use of Tris(trimethylsilyl)antimony for the Preparation of InSb Quantum Dots. Chem. Mater. 2008, 20, 5727- 5730.

[0098] (23) Li, L.; Reiss, P. One-pot Synthesis of Highly Luminescent InP / ZnS Nanocrystals without Precursor Injection. J. Am. Chem. Soc. 2008, 130, 11588-11589.

[0099] (24) Liu, W.; Chang, A. Y.; Schaller, R. D.; Talapin, D. V. Colloidal InSb Nanocrystals. J. Am. Chem. Soc. 2012, 134, 20258-20261 .

[0100] (25) Maurice, A.; Haro, M. L.; Hyot, B.; Reiss, P. Synthesis of Colloidal Indium Antimonide Nanocrystals Using Stibine. Part. Part. Syst. Charact. 2013, 30, 828-831.

[0101] (26) Yarema, M.; Kovalenko, M. V. Colloidal Synthesis of InSb Nanocrystals with Controlled Polymorphism Using Indium and Antimony Amides. Chem. Mater. 2013, 25, 1788-1792.

[0102] (27) Tamang, S.; Kim, K.; Choi, H.; Kim, Y.; Jeong, S. Synthesis of Colloidal InSb Nanocrystals via In Situ Activation of lnCI3. Dalton Trans. 2015, 44, 16923-16928.

[0103] (28) Srivastava, V.; Dunietz, E.; Kamysbayev, V.; Anderson, J. S.; Talapin, D. V. Monodisperse InAs Quantum Dots from Aminoarsine Precursors: Understanding the Role of Reducing Agent. Chem. Mater. 2018, 30, 3623-3627.

[0104] (29) Busatto, S.; de Ruiter, M.; Jastrzebski, J. T. B. H.; Albrecht, W.; Pinchetti, V.; Brovelli, S.; Bals, S.; Moret, M.-E.; de Mello Donega, C. Luminescent Colloidal InSb Quantum Dots from In Situ Generated Single-Source Precursor. ACS Nano 2020, 14, 13146-13160.

[0105] (30) Ginterseder, M.; Franke, D.; Perkinson, C. F.; Wang, L.; Hansen, E. C.; Bawendi, M. G. Scalable Synthesis of InAs Quantum Dots Mediated through Indium Redox Chemistry. J. Am. Chem. Soc. 2020, 142, 4088-4092.

[0106] (31) Zhu, D.; Bellato, F.; Bahmani Jalali, H.; Di Stasio, F.; Prato, M.; Ivanov, Y. P.; Divitini, G.; Infante, I.; De Trizio, L.; Manna, L. ZnCI2 Mediated Synthesis of InAs Nanocrystals with Aminoarsine. J. Am. Chem. Soc. 2022, 144, 10515-10523.

[0107] (32) Tessier, M. D.; De Nolf, K.; Dupont, D.; Sinnaeve, D.; De Roo, J.; Hens, Z. Aminophosphines: A Double Role in the Synthesis of Colloidal Indium Phosphide Quantum Dots. J. Am. Chem. Soc. 2016, 138, 5923-5929.

[0108] (33) Li, Y.; Hou, X.; Dai, X.; Yao, Z.; Lv, L.; Jin, Y.; Peng, X. Stoichiometry-Controlled InP-Based Quantum Dots: Synthesis, Photoluminescence, and Electroluminescence. J. Am. Chem. Soc. 2019, 141 , 6448-6452. (34) Leemans, J.; Dumbgen, K. C.; Minjauw, M. M.; Zhao, Q.; Vantomme, A.; Infante, I.; Detavernier, C.; Hens, Z. Acid-Base Mediated Ligand Exchange on Near-Infrared Absorbing, Indium-Based 111— V Colloidal Quantum Dots. J. Am. Chem. Soc. 2021 , 143, 4290-4301.

[0109] (35) Zhao, T.; Oh, N.; Jishkariani, D.; Zhang, M.; Wang, H.; Li, N.; Lee, J. D.; Zeng, C.; Muduli, M.; Choi, H.-J.; et al. General Synthetic Route to High-Quality Colloidal III— V Semiconductor Quantum Dots Based on Pnictogen Chlorides. J. Am. Chem. Soc. 2019, 141 , 15145-15152.

[0110] (36) Adachi, S. Physical properties of 111- V semiconductor compounds; John Wiley & Sons, 1992.

[0111] (37) Van Embden, J.; Chesman, A. S. R.; Jasieniak, J. J. The Heat-Up Synthesis of Colloidal Nanocrystals. Chem. Mater. 2015, 27, 2246-2285.

[0112] (38) Crisp, R. W.; Grimaldi, G.; De Trizio, L.; Evers, W. H.; Kirkwood, N.; Kinge, S.; Manna, L.; Siebbeles, L. D. A.; Houtepen, A. J. Selective Antimony Reduction Initiating the Nucleation and Growth of InSb Quantum Dots. Nanoscale 2018, 10, 1 1110-111 16.

[0113] (39) Friebolin, H. Basic one- and two-dimensional NMR spectroscopy, 5th, completely revised and updated edition, Wiley VCH, 2010.

[0114] (40) Chang, A. Y.; Liu, W.; Talapin, D. V.; Schaller, R. D. Carrier Dynamics in Highly Quantum- Confined, Colloidal Indium Antimonide Nanocrystals. ACS Nano 2014, 8, 8513-8519.

[0115] (41) Beard, M. C. Multiple Exciton Generation in Semiconductor Quantum Dots. J. Phys. Chem. Lett. 2011 , 2, 1282-1288.

[0116] (42) Chen, Y.; Yin, J.; Wei, Q.; Wang, C.; Wang, X.; Ren, H.; Yu, S. F.; Bakr, O. M.; Mohammed, O. F.; Li, M. Multiple Exciton Generation in Tin-Lead Halide Perovskite Nanocrystals for Photocurrent Quantum Efficiency Enhancement. Nat. Photon. 2022, 16, 485-490.

[0117] (43) Dursun, I.; Maity, P.; Yin, J.; Turedi, B.; Zhumekenov, A. A.; Lee, K. J.; Mohammed, O. F.; Bakr, O. M. Why are Hot Holes Easier to Extract than Hot Electrons from Methylammonium Lead Iodide Perovskite? Adv. Energy Mater. 2019, 9 (22), 1900084.

Claims

Claims1. A method of synthesising AB colloidal quantum dots (CQDs), wherein A is Al, Ga, In or a mixture thereof, and B is P, As, Sb or a mixture thereof, the method comprising the following steps: a) reacting AX3 and BX3, where X is Cl, Bror I, with metal bis(trimethylsilyl)amide, wherein said metal is Li, Na, K or Cs, in at least one long chain amine at an elevated temperature to form A and B silylamides; b) further reacting the A and B silylamides with the at least one long chain amine to form triamides of A and B; c) injecting a reducing agent to balance the reactivity of the triamides of A and B to form nanocrystals of AB CQDs covered by organic ligands; and d) isolating and purifying the AB CQDs.

2. The method of claim 1 , wherein synthesising the AB colloidal quantum dots is performed using a one-pot synthesis strategy.

3. The method of claim 1 or 2, wherein the at least one long chain amine comprises a mixture of multiple long chain amines and / or wherein the at least one long chain amine comprises primary or secondary amines.

4. The method of any preceding claim, wherein the at least one long chain amine comprises oleylamine and / or dodecylamine.

5. The method of any preceding claim, wherein the reducing agent further acts to initiate the reaction.

6. The method of any preceding claim, further comprising synthesising size-tunable colloidal quantum dots.

7. The method of claim 6, further comprising varying the concentration of the reducing agent to achieve size tuning of the CQDs.

8. The method of claim 6 or 7, further comprising tuning the size of the CQDs within the range from 1.5 nm to 20 nm in diameter.

9. The method of any preceding claim, wherein AB represents InAs, InAsP, InGaAs, InSb, InAsSb, InGaSb, and / or InSbP.

10. The method of any preceding claim, wherein the reducing agent comprises one or more of lithium triethylborohydride (LiEtsBH), borane tert-butylamine, alane N,N- dimethylethylamine complex, diisobutylaluminium hydride, and Tris(diethylamino)phosphine.

11. The method of any preceding claim, wherein the CQDs are covered with organic and / or inorganic ligands.

12. The method of any preceding claim, wherein the A and B silylamide complexes are formed in situ.

13. Colloidal quantum dots (CQDs) formed by the method of any preceding claim.

14. AB colloidal quantum dots (CQDs), wherein A is Al, Ga, In or mixture thereof, B is P, As, Sb or mixture thereof, synthesized by a reaction involving AX3 and BX3, where X is Cl, Br or I, with metal bis(trimethylsilyl)amides, wherein said metal is Li, Na, K or Cs, in at least one long chain amine at an elevated temperature to form A and B silylamides, followed by a further reaction with the at least one long chain amine to form triamides of A and B, and the introduction of a reducing agent to balance the reactivity of the triamides of A and B to form nanocrystals of AB CQDs covered by organic ligands, followed by isolation and purification of the AB CQDs.

15. The CQDs of claim 14, wherein the synthesis of the AB colloidal quantum dots is performed using a one-pot synthesis strategy.

16. The CQDs of claim 14 or 15, wherein the at least one long chain amine comprises a mixture of multiple long chain amines and / or wherein the at least one long chain amine comprises primary or secondary amines.

17. The CQDs of any of claims 14 to 16, wherein the at least one long chain amine comprises oleylamine and / or dodecylamine.

18. The CQDs of any of claims 14 to 17, wherein the reducing agent is active to initiate the reaction.

19. The CQDs of any of claims 14 to 18, wherein the synthesis forms size-tunable colloidal quantum dots.

20. The CQDs of claim 19, wherein the concentration of the reducing agent is adjusted to achieve a desired size tuning of the CQDs.

21. The CQDs of claim 19 or 20, wherein the size of the CQDs is tuned to a value within the range from 1.5 nm to 20 nm in diameter.

22. The CQDs of any of claims 14 to 21, wherein A and B are InAs, InAsP, InGaAs, InSb, InAsSb, InGaSb, and / or InSbP.

23. The CQDs of any of claims 14 to 22, wherein the reducing agent comprises one or more of lithium triethylborohydride (LiEtsBH), borane tert-butylamine, alane N,N- dimethylethylamine complex, diisobutylaluminium hydride, and Tris(diethylamino)phosphine.

24. The CQDs of any of claims 14 to 23, wherein the CQDs are covered with organic and / or inorganic ligands.

25. The CQDs of any of claims 14 to 24, wherein the A and B silylamide complexes are formed in situ.

26. The CQDs of any of claims 13 to 25, wherein the CQDs display a sharp excitonic transition in the short-wave infrared (SWIR) region or / and the middle-wave infrared (MWIR) region between 1100 nm and 5000 nm.

27. The CQD of any of claims 13 to 26, wherein the CQDs are configured to absorb infrared radiation within at least a portion of the short-wave infrared (SWIR) range between 1100 nm and 3000 nm and / or within at least a portion of the middle-wave infrared (MWIR) wavelength range between 3000 nm and 5000 nm.

28. The CQD of any of claims 13 to 26, wherein the CQDs are configured to absorb infrared radiation within at least a portion of the wavelength range of > 1200 nm.

29. The CQDs of any of claims 13 to 28, wherein CQDs have a response time under UV- VIS-IR light excitation which is less than 1 ns.

30. The CQDs of any of claims 13 to 29, wherein the CQDs form a photoactive layer having a thickness in the range from 50nm to 500nm.

31. An infrared image sensor including a photoactive layer formed from the CQDs of any of claims 13 to 30 to convert incoming infrared light into an electrical signal representing an image.

32. The infrared image sensor of claim 31 , wherein the sensor includes a stack of layers comprising: a CMOS substrate, a bottom electrode, a hole transport layer, the photoactive layer, an electron transport layer, and a top electrode.

33. The infrared image sensor of claim 31 or 32, wherein the infrared image sensor has a response time to incoming infrared radiation of below 100 ns, preferably below 10 ns, preferable below 1 ns.

34. A device comprising a machine vision camera, a camera for robotics, virtual reality goggles, mixed reality goggles, a smartphone, an automotive camera, an eye interactive camera, or a set of contact lenses, wherein said device incorporates an infrared image sensor of any of claims 31 to 33.

35. An infrared photodetector including a photoactive layer formed from the CQDs of any of claims 13 to 30 to convert incoming infrared light into an electrical signal representing a binary signal or a spectrum.

36. The infrared photodetector of claim 35, wherein the photodetector includes a stack of layers comprising: a CMOS substrate, a bottom electrode, a hole transport layer, the photoactive layer, an electron transport layer, and a top electrode.

37. The infrared photodetector of claim 35 or 36, wherein the infrared photodetector has a response time to incoming infrared radiation of below 100 ns, preferably below 10 ns, preferable below 1 ns.

38. A device comprising a smartphone or an eye interactive camera, wherein said device incorporates an infrared photodetector of any of claims 34 to 37.