Narrow band-edge photoluminescence from AgInS2 semiconductor nanoparticles by the formation of amorphous III–VI semiconductor shells | NPG Asia Materials

Choice of shell materials

Preparation of a type I core/shell structure, meaning that the bandgap of the shell covers the bandgap of the core, is a common strategy for obtaining strong and stable emission from semiconductor NPs18. Zinc sulfide (ZnS) has been widely used as a shell material for II–VI and III–V semiconductor NPs because of its chemical stability, high crystallinity, and wide bandgap (3.7 eV), which is crucial for type I band alignment. Coating with ZnS has also been attempted for CuInS2 ternary NPs by reacting them with a mixed solution of zinc and sulfur sources19. The treatment resulted in blueshifts in PL, in contrast to the redshift generally observed for semiconductor NPs after overcoating with larger-bandgap semiconductors. The expected redshift is usually explained by the leakage of excitons from the core into the shell, whose bandgap is essentially smaller than the HOMO-LUMO gap of insulating organic ligands on its surface20. The blueshift, in contrast, is often explained as the result of changes in surface defect levels by surface passivation with ZnS and/or the result of partial alloying by cation exchange from copper and indium into zinc21,22,23. In our experiments, we also observed a blueshift in PL when AgInS2 NPs were reacted with zinc acetate (Zn(OAc)2) and elemental sulfur at 200 °C, as shown in the optical spectra in Fig. 1a, b.

Fig. 1

UV-vis and PL spectra of AgInS2 NPs coated with different semiconductors. a UV-vis and b PL spectra of AgInS2 NPs synthesized with the heating up approach and the same NPs after the reaction with zinc acetate and elemental sulfur to form ZnS shells. c UV-vis and d PL spectra of AgInS2 NPs, InSx NPs, and AgInS2 NPs after the reaction with indium acetate and elemental sulfur to form indium sulfide shells. The excitation wavelength was 450 nm for all spectra. Samples were diluted with chloroform to become ca. 0.1 µM in terms of particle upon measurements

Full size image

According to the abovementioned considerations, the semiconductor used for the shell should be selected from materials that do not easily form alloys, or solid solutions, with the core NPs. However, the formation of defect levels between the core and shell must be avoided. Considering the band-edge criteria for the type I core/shell structure, indium sulfide, which shares two of the three elements composing the core AgInS2 and has a wider bandgap (2.2 eV) than that of the core (1.8 eV for bulk), emerged as a candidate material. The phase diagram for the Ag2S–In2S3 quasi-binary system shows that AgInS2 forms continuous solid solution series with In2S3 only within a limited region of high indium content (81–96% In2S3)24, which is less miscible than with ZnS25. Figure 1c, d shows UV–vis absorption and PL spectra obtained after the AgInS2 core NPs were reacted with indium acetate (In(OAc)3) and elemental sulfur. The indium sulfide (InSx) shell was synthesized by slowly increasing the temperature of the oleylamine solution of the AgInS2 core in the presence of these precursors (10 °C/min up to 200 °C) and maintaining this temperature for 1 h. The resulting PL spectrum was apparently different from that of the ZnS coating, with the emergence of a new narrow PL peak on the short-wavelength side of the original defect emission. In contrast, InSx NPs synthesized by a method similar to that used for the shell coating was completely non-luminescent, confirming that the narrow PL peak originated from the AgInS2/InSx core/shell structure. However, due to insufficient passivation, substantial broad emission persisted, with a blueshift of 70 nm, suggesting that part of the indium source participated in changing the core composition to an indium-rich form (e.g., AgIn5S8) during the reaction26.

The monochromaticity of the emission was further greatly improved by coating the AgInS2 core NPs with gallium sulfide (GaSx) instead of with InSx. First, a structure composed of several shoulders appeared in the UV–vis spectrum, resembling the absorption of size-quantized, well-stabilized semiconductor QDs (Fig. 2a)27. In Fig. 2b, PL spectra are shown for different elapsed times (0–60 min) after the reaction temperature reached 260 °C. The broad-band defect emission (>600 nm) of the uncoated AgInS2 NPs decreased with the formation of the gallium sulfide shell, followed by the appearance of a new narrow peak on the short-wavelength side of the defect emission, as demonstrated in the imaged obtained under UV illumination (Fig. 2c). Similarly to the indium sulfide of the same group, no single phase has been reported between AgInS2 and Ga2S3 according to the quasi-ternary phase diagram of Ag2S, In2S3, and Ga2S328. The great improvement of monochromaticity might be attributed to the wider bandgap of gallium sulfide (2.8 eV) compared with that of indium sulfide (2.2 eV) (Fig. 2d). The similarity in wavelength of the newly generated PL peaks between InSx and GaSx-coated AgInS2 NPs ruled out the possibility of cation exchange between indium in the core and gallium in the shell.

Fig. 2

Optical changes of AgInS2 NPs by the formation of gallium sulfide shell. a Absorption spectra of AgInS2 NPs and those reacted 60 min to form gallium sulfide shells. b Time changes of PL spectra of AgInS2 NPs when heated with gallium acetylacetonate and elemental sulfur in oleylamine to form gallium sulfide shells. Times indicate the periods after the solution temperature reached 260 °C. Both UV-vis and PL spectra were recorded by diluting samples with chloroform to become ca. 0.1 µM (in terms of particle) and PL spectra were normalized by the absorbance at the excitation wavelength (400 nm). c Photograph of AgInS2 NPs during coating with gallium sulfide under UV irradiation. d Energy diagrams of core/shell structures

Full size image

To the best of our knowledge, such drastic changes in the PL and UV–vis spectra only appeared when the AgInS2 NPs were reacted with indium (III) or gallium (III) species in the presence of sulfur and were not been observed when the mixture was heated in the absence of one or both precursor materials. When the core NPs were heated at 270 °C for 1 h in oleylamine without adding any precursor materials, the NPs partially agglomerated and the quantum yield decreased by approximately 80%, possibly due to Ostwald ripening (Figure S2, supplementary information). Therefore, the emergence of narrowband emission, which reflects a diminishing contribution of the defect sites to the PL, likely originated from the NPs’ surface modification, rather than from the removal of defect levels inside the crystals or the transformation of the crystalline phase by the applied heat. Several studies have investigated the mechanism of broad defect emission with a large Stokes shift13,14,16,26. The defect emission is often explained in relation to the existence of localized donor defect or acceptor defect states caused by vacancies, interstitials or anti-sites within the crystal structure or by structural defects on the surface. The defect states are often proposed to explain the fact that the actual composition of I–III–VI semiconductors often deviates from the stoichiometric composition. However, the present results indicate the prevailing importance of surface conditions in generating band-edge emission.

Origin of narrow-band emission

Figure 3a shows PL spectra of the original bare AgInS2 NPs and of those coated with gallium sulfide in chloroform after isolation of the NPs via the precipitation/dissolution method. The PL QY was 12.4% for the sharp band emission (530–630 nm) and 21.7% for the entire PL range (530–900 nm). The narrowband emission of the core/shell NPs had a FWHM as small as 28.6 nm (103 meV), much smaller than that of the uncoated AgInS2 NPs (220 nm, 388 meV). The corresponding photoluminescence excitation (PLE) spectra recorded at the respective peak wavelengths are shown in Fig. 3b, along with a UV–vis spectrum of the core/shell NPs. A distinct excitonic feature at 530 nm was observed for the excitation spectrum of the core/shell NPs only, whereas the uncoated AgInS2 NPs showed a monotonic increase in PL intensity with decreasing excitation wavelength. The PL decay curves in Fig. 3c show a significant decrease in the lifetime components for the core/shell NPs compared with those of the uncoated AgInS2 NPs. The decay curves for the core/shell NPs were fitted to a three-exponential equation:

$$I\left( t \right) = A_1\exp \left( { – \frac{t}{{\tau _1}}} \right) + A_2\exp \left( { – \frac{t}{{\tau _2}}} \right) + A_3\exp \left( { – \frac{t}{{\tau _3}}} \right),$$

where I(t) is the PL intensity at time t; A1, A2, and A3 are amplitudes; and τ1, τ2, and τ3 are lifetimes. In contrast, the decay curve of the uncoated AgInS2 NPs was sufficiently fit with a single-component exponential. The fitting results are summarized in Table 1. For the core/shell NPs, the second component (24.8 ns) contributed to 45% of the total emission intensity and the third component (96.6 ns) contributed to 50% of the total emission intensity compared with the components described in terms of products of amplitude and lifetime (A × τ). These lifetimes are comparable with the lifetime of the band-edge emission from CdSe QDs29, indicating that the new short-wavelength peak that appeared after coating with gallium sulfide may originate from the band-edge transition as well.

Fig. 3

Optical analyses of AgInS2 NPs and AgInS2/GaSx core/shell NPs. a PL spectra for the two types of NPs with 450 nm excitation, b PL excitation spectra for the corresponding samples recorded at each PL peak (585 nm and 800 nm), c PL decay curves recorded at 585 nm for the core/shell NPs and at 780 nm for the core-only NPs with 470 nm excitation. The inset shows the magnified curve for the core/shell NPs. Samples were diluted with chloroform to become ca. 0.1 µM upon measurements

Full size image

Table 1 PL decay components for AgInS2 NPs and AgInS2/GaSx core/shell NPs

Full size table

Plots of (αhν)2 as a function of hν, generally known as Tauc plots, support this interpretation (Figure S3, supplementary information). Flat regions in these plots, which appeared both for AgInS2 core NPs (Fig. S3a) and for those coated with ZnS (Fig. S3b), InSx (Fig. S3c), and GaSx (Fig. S3d) shells, are spectroscopic features of direct allowed transitions and indicate their presence in these samples. The estimated bandgap was very similar between these four samples, indicating that alloying was limited under the experimental conditions used for ZnS coating. However, differences between these bandgap values and the PL peak energies, i.e., Stokes shifts, were significantly larger for the AgInS2 core NPs (0.59 eV) and for those coated with ZnS (0.37 eV) and much smaller for those coated with InSx (0.05 eV) and GaSx (0.06 eV), indicating a band-edge or near band-edge emission for the latter two samples.

Structural characterization of core/shell nanoparticles

The synthesized core-only and core/shell NPs were characterized by TEM. Figure 4a, c shows TEM images of AgInS2 NPs before and after the formation of gallium sulfide shells, respectively, as well as size histograms of the two types of NPs (Fig. 4b, d). The AgInS2 NPs show irregular, angular shapes that transition to rounded ones after the formation of the shell with a slight increase in the average particle diameter (0.4 nm) corresponding to the formation of 0.2 nm thick shell. However, the actual effect of overcoating on the average NP diameter is difficult to quantify at this stage due to the polydispersity of the original uncoated AgInS2 NPs. HRTEM (Fig. 4e) and HAADF-STEM (Fig. 4f) images indicate the presence of an amorphous layer surrounding the crystalline AgInS2 NPs, suggesting that the GaSx shell is amorphous. Elemental analyses for the ensemble showed an increase in gallium and sulfur after shell formation. Compared with the composition of the cores, that of the amorphous GaSx was closer to that of Ga2S3 rather than to that of another possible candidate, GaS (Table S1 and Figure S4, supplementary information). The amorphous nature of the shell was also supported by X-ray diffraction (XRD) analysis; the XRD patterns showed no changes in the NP crystal structure before and after the formation of the Ga2S3 shells (Fig. 5a). In the X-ray photoelectron spectroscopy (XPS) spectra, the positions of the peaks for silver, indium, and sulfur coincided with those of a metal sulfide, and they remained nearly unchanged during the formation of Ga2S3 shells, indicating that the NP cores were chemically unaffected by the formation of the shells (Fig. 5b, c, e, f). Although we could not obtain reference data directly for gallium sulfide, the measured peaks for gallium (Fig. 5g, h) were singular, and their position was different from that of the two possible byproducts, metal gallium and gallium oxide30,31. In terms of chalcogens, no additional peaks were detected for the core/shell NPs, indicating that the sulfur in the shell occurred in the form of a metal sulfide (Fig. 5i).

Fig. 4

TEM images and the corresponding size histograms of AgInS2 and AgInS2/Ga2S3 core/shell NPs. a TEM image and b size histogram for AgInS2 NPs. c TEM image and d size histogram for AgInS2/Ga2S3 core/shell NPs. e HRTEM and f HAADF-STEM images of AgInS2/Ga2S3 core/shell NPs. Upon evaluating particle size, each particle was fit elliptically using Image-J software and evaluated as the square root of the product of major and minor axes. Size histograms, average diameter (dav), and standard deviation (σ) were calculated from a few sets of TEM images including >100 nanoparticles in total

Full size image

Fig. 5

XRD patterns and XPS spectra of core/shell NPs. a XRD patterns of AgInS2 (red) and AgInS2/Ga2S3 core/shell (blue) NPs. b–d XPS spectra of AgInS2 core NPs for Ag 3d, In 3d, and S 2p orbitals. e–i AgInS2/Ga2S3 core/shell NPs for Ag 3d, In 3d, Ga 2p, Ga 3d, In 4d, and S 2p orbitals. All the XPS spectra were drawn in red lines with blue fitting curves

Full size image

The concept of an amorphous shell is unusual compared with that of commonly used II–VI shells, such as ZnS and CdS, coated onto II–VI or III–V semiconductor NPs. Generally, II–VI shells have been preferred because of their good crystallinity and expected lattice matching at the interface between the core and shell materials used. Indium and gallium sulfides possess crystal structures that are different from those of both chalcopyrite and wurtzite AgInS2 and are known to contain many crystalline defects32,33,34. Therefore, these III–VI semiconductors have never been used as shell materials. However, we consider that the amorphous nature of the shell may sometimes be advantageous in stabilizing the complex surface elements on I–III–VI semiconductors, as this amorphous nature allows for small, continuous displacement of elements into energetically stable conformations. Similar effects have been observed for solar cells based on copper indium gallium sulfide (CIGS); in particular, the introduction of an In2S3 buffer layer between CIGS and n-type ZnO layers improved cell performance35,36. These buffer layers deposited by chemical vapor deposition (CVD) or atomic layer deposition (ALD) were reportedly amorphous in nature, although a definite correlation between the amorphous nature and improved photovoltaic performance has not been proven.

Single-particle analysis

We next used single-particle spectroscopy to gain insight into the extent of inhomogeneity in the band-edge emission. This method has been previously used to successfully uncover the multiple-defect nature of individual ZnS–AgInS2 solid solution NPs and to demonstrate suppression of blinking by two-color excitation37. In the context of the current study, the spectra of single core/shell NPs can directly answer the important question regarding whether the residual defect emission originates from the same NPs as the bend-edge emission or whether the two types of emission derive from physically distinct particles. Figure 6 shows nine examples of PL spectra from individual NPs immobilized on a quartz substrate. The single-particle nature of the emission was confirmed by measuring the concentration dependence of the sample and by observing the PL intermittency (blinking) phenomenon (Figure S1 in the supplementary information) from individual NPs. The examples in Fig. 6 were chosen from a set of more than 250 measured NPs, each analyzed for the first 10 s of its emission. As shown in Fig. 6, the top four NPs produced only band-edge type spectra, while the bottom five NPs produced only defect-type spectra. Statistically, 50.4% of the analyzed NPs showed band-edge emission, 36.8% showed defect emission and the remaining 12.8% NPs showed both types of emission simultaneously. The same sets of spectra as a function of photon energy are shown in Figure S5 in the supplementary information together with the spectrum of an NP ensemble used for the single-particle measurements. As shown in Fig. 6, the PL FWHMs from the individual NPs were smaller than the FWHM of the ensemble spectra. This observation indicates that the ensemble PL spectra include a distribution of peak wavelengths of individual particles and are thus inhomogeneously broadened, likely due to the size distribution of the core NPs. The PL peak energy and FWHM obtained from the observation of 250 individual NPs were analyzed and plotted as histograms, as shown in Figure S6 of the supplementary information. For the band-edge emission, the mean FWHM obtained from the histogram was as small as 80.0 meV (24.0 nm). Importantly, this value is close to the PL FWHM for single CdSe/CdS/ZnS core/shell/shell QDs embedded in polymer matrices (55–65 meV) recorded at room temperature38. Thus, the AgInS2/Ga2S3 core/shell NPs have considerable potential to serve as Cd-free alternatives in terms of emission properties.

Fig. 6

Single particle PL spectra of AgInS2/Ga2S3 core/shell NPs recorded for nine different NPs. Each spectrum was fitted with a single gaussian functions. Peak wavelengths and FWHM were displayed beside peaks in the form of “peak wavelength [nm] (FWHM [nm])”

Full size image

Improvement of quantum yield

The low QY (12.4% for band-edge) and the remaining defect emission are major issues that must be improved for potential practical applications. Both are considered to originate from the insufficient removal of defect levels in both the core and the shell. The aforementioned ‘mix everything and heating up method’ used to synthesize AgInS2 core NPs generates particles with large size polydispersity, as shown in the TEM image and corresponding histogram presented in Fig. 4a, b. The polydispersity originates from several types of reactions occurring simultaneously during heating to 200 °C. When the as-synthesized core NPs were size separated before the formation of gallium sulfide shells, the smaller core fractions tended to yield more intense band-edge emission, suggesting that smaller NPs have fewer defect levels and/or are more favorably shelled. Therefore, we made modifications to the core synthesis conditions to selectively obtain smaller core NPs.

According to previous reports, AgInS2 NPs are most commonly generated through Ag2S as an intermediate product because Ag(I) species react more favorably with sulfur sources than with In(III) complexes. The two main roles of the intermediate Ag2S are to act as seed particles on which cation exchange of Ag+ and In3+ occurs39 and to serve as resource hosts that operate as catalysts for the reaction of indium and sulfur sources40. Theoretically, the former reaction leads to an approximately 15% decrease in particle volume, which has actually been observed39. The latter reaction would result in a three-fold increase in particle volume, which is undesirable in the present case. In our reaction system using thiourea as a sulfur source, the color of the solution began to change at 120 °C from yellow to orange and to dark brown, indicating the formation of Ag2S NPs. The target AgInS2 NPs were then formed at higher temperatures (>130 °C) either through cation exchange with In(III) species (at 130 °C–150 °C) or via reaction with indium sulfide source materials generally recognized as clusters before nucleation (above 150 °C). Therefore, in this study, the reaction temperature was kept low at 135 °C for 10 min to allow for the composition of the entire NP ensemble to change to AgInS2. This modified procedure successfully generated AgInS2 core NPs with a smaller particle diameter and a narrower size distribution than those of particles generated previously (Fig. 7a, b). The PL QY of the core NPs was 64% (Fig. 8a), which is a record high value for AgInS2 core-only NPs, even though the spectral shape is that of broad defect emission.

Fig. 7

TEM images and size histograms of NPs prepared using the modified procedure. a TEM image and b size histogram of AgInS2 core NPs. c TEM image and d size histogram of AgInS2/GaSx core/shell NPs. e HRTEM image of AgInS2/GaSx core/shell NPs. Particle size was evaluated by fitting each particle elliptically and calculated as the square root of the product of two axes. Size histograms, average diameter (dav), and standard deviation (σ) were calculated from a few sets of TEM images including >200 nanoparticles in total

Full size image

Fig. 8

Optical properties of quantum dots synthesized using the modified procedures. UV-vis and PL spectra for a AgInS2 core NPs, b AgInS2/GaSx core/shell NPs, and c AgInS2/GaSx core/shell NPs with TOP treatment. d Photographs of the three samples under room light (top) and UV irradiation (bottom). Samples were diluted with chloroform to be ca. 0.05 µM and PL spectra were recorded with the excitation wavelength of 450 nm

Full size image

For the shell formation reaction, the sulfur source was changed from elemental sulfur to 1,3-dimethylthiourea, which reacts with Ga(acac)3 at higher temperatures (>230 °C) to generate gallium sulfide. Although elemental sulfur is a common sulfur source for many chalcogenide nanoparticles, we observed that it has a strong quenching effect on the PL of AgInS2 core NPs. Undesired contact with elemental sulfur in the early stage of the shell formation reaction could thus be avoided by the use of the sulfur compound. In addition, the change in atomic ratio between gallium and sulfur (from Ga:S = 1:1.5 to 1:1) led to gallium-rich conditions. The temperature gradient was also modified in accordance with the decomposition of 1,3-dimethylthiourea to avoid the nucleation of gallium sulfide as much as possible. These modifications to the synthesis protocol improved the PL QY of the AgInS2/GaSx core/shell NPs to 15–20% for the band-edge emission and to 25−30% for the entire spectral region (Fig. 8b). For the band-edge emission, this increase represents an improvement by a factor of 1.5 over the PL QY obtained using the original synthesis conditions. The thickness of the shell, calculated from the average diameter change before and after shell formation, was increased from 0.2 nm for the previous method to 0.6 nm (Fig. 7b, d), although the shell was still thinner than the expected value (1 nm) calculated based on the amount of materials. An HRTEM image clearly shows the existence of 0.5–1.0 nm amorphous shells around crystalline cores measuring 3–5 nm in diameter, depending on location (Fig. 7e). The difference between the expected bandgap and PL peak energy became even smaller (0.02 eV), indicating better passivation (Figure S3e, supplementary information). Further improvement of the PL QY was achieved by post-treatment of the AgInS2/GaSx core/shell NPs with TOP. The post-treatment was carried out after removing the NPs from the solution used for shell formation. The intensity of the band-edge emission nearly doubled within a few hours in the presence of TOP, with a slight decrease in the remaining defect emission, keeping the UV–vis absorption profile nearly unchanged (Fig. 8c). As a result, nearly single-band emission with a QY as high as 56% was obtained with high reproducibility, yielding a brilliant yellow emission bright enough to observe under ambient light (Fig. 8d and Figure S7, supplementary information). The crystal structures of the AgInS2 core and AgInS2/GaSx core/shell NPs were similar to those prepared by the original method, but a purer tetragonal phase was observed (Figure S8, supplementary information). These findings are well consistent with a previous report by Doh et al.39 that showed that 4 nm Ag2S NPs were converted to tetragonal AgInS2 NPs, while larger (8 nm) NPs were transformed into orthorhombic AgInS2 NPs, upon cation exchange with In3+ species. According to that report, Ag2S seed particles are prone to conversion to orthorhombic AgInS2, which is structurally more favorable, whereas the thermodynamically more stable tetragonal phase can be generated when the particle size is small enough (<4.5 nm for seed Ag2S) to invoke a rearrangement.

The remarkable improvement in PL QY by the TOP treatment was surprising, and we offer the following hypothesis. The gallium sulfide shell passivates the defect levels on the surface of the AgInS2 core NPs, which leads to the generation of the band-edge emission. However, excitons could still be leaking outside of the core due to an insufficient potential barrier between the core and the thin amorphous gallium sulfide shells. The major traps for electrons were, potentially, zero-valent sulfur sites existing on the surface of the gallium sulfide shell. These traps operate as acceptors for photogenerated electrons. TOP is well known to coordinate with chalcogens; therefore, the TOP treatment has the potential to passivate the electron traps on the surface of the gallium sulfide shells, either by filling electrons in the LUMO level of the surface sulfur sites or by removing the sulfur sites via conversion to a more stable tri-n-octylphosphine sulfide. Elemental analyses of the core/shell NPs synthesized with the modified procedure showed a decrease in the ratio of sulfur in the shell (to Ga:S = 1:0.84) from that observed for the shells synthesized using elemental sulfur with a 1.5 equivalent of Ga(acac)3 (Table S2 and Figure S9, supplementary information). The abovementioned observations suggest that the amount and valency of sulfur may be key factors controlling the PL QY of core/shell NPs.

Unexpectedly, however, the PL decay curves obtained before and after the TOP treatment did not show such drastic changes (Fig. 9). The PL lifetimes, which nearly doubled those of the core/shell NPs synthesized using >6 nm cores and elemental sulfur for the shell source material (Table 1), remained unchanged after the TOP treatment (Table 2). These results suggest suppression of blinking, which itself is generally caused by NP charging. We indeed observed strong blinking with long off-times for the AgInS2/GaSx core/shell NPs synthesized by the first method (Figure S1, supplementary information), similar to the blinking behavior of uncoated core NPs37. However, Figure S10 shows the time traces of PL intensity of individual core/shell NPs synthesized with the modified method before and after TOP treatment. The core/shell NPs without TOP treatment (Figure S10, bottom, red traces) showed short ‘on’ durations with an on/off frequency often higher than the recording frame rate (20 Hz). This behavior was substantially altered by the addition of TOP, which resulted in longer and more stable ‘on’ states (Figure S10, top, blue traces). Therefore, the doubling of the PL QY by TOP treatment is mostly attributed to the increase in ‘on’ duration and was likely caused by the effective blocking of charges from escaping from the core. In addition, it is worth mentioning that the TOP-treated core/shell NPs exhibited a PL QY of 56.0%, which is comparable to that of the core NPs (64%), suggesting that we were successful in altering the pathway of photogenerated carriers without causing a significant efficiency loss. Studies aimed at further increasing the PL QY and identifying the quantitative aspects of the corresponding excitonic characteristics are underway.

Fig. 9

Time dependent PL analyses for AgInS2/GaSx core/shell NPs synthesized using the modified procedures. PL decay curves a before and b after TOP treatment. Both samples were diluted with chloroform to be 0.1 µM and decay curves were recorded with excitation and emission wavelengths of 470 and 570 nm, respectively

Full size image

Table 2 PL decay components for AgInS2/GaSx core/shell NPs synthesized using the modified procedures before and after TOP treatment

Full size table