Gouranga H. Debnath,
Arijita Chakraborty and
Prasun Mukherjee*
Centre for Research in Nanoscience and Nanotechnology, University of Calcutta, JD-2, Sector-III, Salt Lake, Kolkata-700106, West Bengal, India. E-mail: pmukherjee12@gmail.com
First published on 31st August 2016
The spatial location of a foreign species (dopant) in a semiconductor nanoparticle matrix is probed by monitoring the infrared absorption and photoluminescence spectroscopy of the capping ligand and dopant moieties respectively at room temperature. The results have been rationalized within the domain of observables in each experiment and argued to provide a comprehensive picture in tracking the core and surface localized terbium cations that are incorporated in zinc sulfide nanoparticles. The progression of luminescence quantum yield of the core and surface localized terbium cations with doping extent induced variation has been found to be different. The generality of the experimental observations has been demonstrated with the corresponding europium incorporated nanoparticles.
In the context of doping Ln3+ in a semiconductor nanoparticle matrix, charge and size mismatch of the host and guest cation pair needs consideration. For example, Bol and co-workers20 discussed about the difficulties associated to dope Ln3+ in semiconductor nanoparticle matrices with charge and size mismatched pairs and argued that in such cases the dopants might only populate the surface sites of the nanoparticles. However, it is noteworthy that various reports in the literature demonstrate the feasibility of doping charge and size mismatched pairs of cationic host-guest (dopant) ingredients, indicating that with proper synthetic methodologies it is indeed possible to dope cations that differ in charge and size with respect to the original host cation.6,21–24 Specifically, Chen and co-workers21,22 demonstrated doping of Ln3+ in TiO2 nanoparticles (ionic radii for six co-ordinated Ln3+ 100–117 pm, Ti4+ 75 pm).25 In a case study, these authors further discussed that such a doping of Eu3+ in anatase TiO2 nanoparticles result in a change of symmetry of Ti4+ from D2d to either D2 or C2v for internal sites and C1 for disordered external surface related sites.21 Banin and co-workers23 discussed successful doping of Cu2+, Ag+ and Au3+ separately in InAs nanoparticles, (ionic radii for six co-ordinated In3+ 94 pm, Cu2+ 87 pm, Ag+ 129 pm, Au3+ 99 pm),25 with further demonstration of doping Cu2+ in the interstitial sites of InAs nanoparticles.24 The question on charge mismatch between the cationic ingredients might be visualized by a charge compensation that is mediated by the presence of appropriate defects in the nanoparticle or by the nanoparticle's surface capping ligands. The size mismatch between the two cations could even result in lattice distortion as has been discussed by Banin and co-workers, with the ability to dope InAs nanoparticles with Cu2+, Ag+ and Au3+ separately23 and Chen and co-workers in the context of doping Eu3+ in TiO2 nanoparticles.21
Based on electron microscopy, photoluminescence (steady-state, time-gated and time-resolved) and infrared absorption spectroscopy experiments, we have been arguing to successfully incorporate Ln3+ in the II–VI semiconductor nanoparticles.2,26–29 This rationalization is based on the following experimental observations; (i) identification of Tb3+ in the energy dispersive X-ray spectrum (EDS), which has also been observed in the corresponding spectrum from scanning transmission electron microscopy (STEM) mode covering only a few particles,27 (ii) increase in the Tb3+ and Eu3+ emission in presence of the Ln3+ incorporated nanoparticles compared to that in the corresponding free Ln(III) precursor in bulk solvent,26 (iii) appearance of a broad excitation profile upon monitoring the Ln3+ emission bands which closely resembles to that of the zinc sulfide centered emission, without significant contribution from direct intra-configurational 4f–4f excitation bands (an optical antenna effect),2,26,27 (iv) observation of a shorter and longer Ln3+ luminescence lifetime in the Zn(Ln)S [Ln = Tb, Eu] nanoparticles, suggesting a lesser (presumably surface localized) and better (presumably core localized) protected Ln3+ co-ordination environment in the nanoparticles,2,26,27 with the longer lifetime component values even longer than the corresponding values reported for well protected Ln3+ molecular complexes, being 1.3 and 0.78 ms respectively comprising with Tb3+ and Eu3+ based complexes,30 (v) near constancy of the infrared absorption characteristics of the symmetric and asymmetric stretching frequencies originating from the longer alkyl chain of the capping ligand in absence (undoped) and presence of dopants (lanthanide incorporated) in the nanoparticles argue in favour of insignificant role of Ln3+ being intercalated in the long alkyl chain of the capping ligands28 and (vi) difference in IR absorption signatures of Zn(Ln)S [Ln = Sm, Dy] nanoparticles with the corresponding Ln(III) stearate salts indicate negligible contribution from Ln(III) stearate salts in modulating the signature in the Zn(Ln)S nanoparticles (for representative spectral comparison see ESI†).28 Various other researchers have also presented experimental evidence supporting the incorporation of Ln3+ in II–VI semiconductor nanostructures.1,4–6,9,10,31,32
A well-orchestrated doped semiconductor nanoparticle would require knowledge of dopant spatial location (core versus surface) in the host matrix. Deciphering the dopant location helps understanding the physical and chemical properties of the material under investigation and has important implications towards the development of novel doped nanomaterials. Significant difference in dopant properties might be expected from a core localized and surface related dopant moiety. Few researchers addressed this important aspect of understanding the dopant location in semiconductor nanoparticles.24,26,33–36 A majority of the previous studies on this front focus on monitoring the luminescence of the species under interest26,33–36 and in some cases involves low temperature measurements for site selective spectroscopy.21,35,36 From a structural viewpoint, in a case study with Cu2+ doping in InAs nanoparticles, Banin and co-workers24 provided insight from X-ray spectroscopy and correlated the experimental observations with density functional theory calculations. The approach discussed in this study differs in that it considers monitoring different spectral properties from two different species; namely the infrared (IR) absorption of the capping ligand moieties and the dopant photoluminescence. The surface localized dopant cations are potentially eligible to alter the IR absorption characteristics of the capping ligand molecules located on or near the surface of the nanoparticles. Moreover, the luminescence quantum yield and the luminescence lifetime of surface localized dopant cations would be expected to be lesser than the corresponding core localized counterparts. In short, this account provides an unprecedented opportunity to compare easily accessible different spectral properties from different moieties in a given system at room temperature, in order to address the same question of identifying the dopant location in the semiconductor nanoparticle matrix.
In a recent work we discussed the modulation of capping ligand [strearate and trioctylphosphine oxide (TOPO)] IR absorption by Ln3+ cations in the lanthanide cation incorporated ZnS [Zn(Ln)S] [Ln = Sm, Eu, Tb, Dy] nanoparticles.28 A comparison of the IR absorption characteristics in the Zn(Ln)S nanoparticles to that in the undoped ZnS and zinc stearate (ZnSt) reveals that lanthanide cation incorporation (a) decreases the intensity of the carboxylate asymmetric (as) and symmetric (s) stretching bands (νas[COO−] = 1538 cm−1 and νs[COO−] = 1398 cm−1) and (b) induces the emergence of a new carboxylate symmetric stretching band (νs[COO−] = 1372 cm−1) in the Zn(Ln)S nanoparticles.
The present study discusses the doping extent induced change in the trend of IR absorption of capping ligand moiety (stearate) in the Tb3+ incorporated ZnS [Zn(Tb)S] nanoparticles and the corresponding variation in their luminescence properties, in the context of assessing the spatial location of dopant cations in the semiconductor nanoparticle matrix and the corresponding luminescence properties. A doping percent of 1%, 5%, 10%, 15%, 20% and 30% have been considered. The choice of the nanoparticle system for the present work has been considered based on our previous study,2 where a relative energy level schematic identifies ZnS nanoparticles as the most promising host candidate to sensitize the Tb3+ luminescence among the II–VI sulfide and selenide materials. The experiments follow an analysis to monitor and characterize the photophysical properties of core and surface localized lanthanide cations. In order to examine the generality of the trend in experimental observations, corresponding Zn(Eu)S nanoparticles with 5%, 15% and 30% doping extent have been considered.
The primary objective of this study which is to decipher the possible location of the Ln3+ dopant and understanding the luminescence properties of the dopants those are spatially placed in different locations in the zinc sulfide nanoparticle matrix with the combined use of capping ligand FTIR and dopant photoluminescence spectroscopy, maximizing the Ln3+ luminescence in the semiconductor nanoparticle matrix is desirable. A previous study by our group on the various sized Zn(Tb)S nanoparticles27 identifies a Zn(Tb)S particle size of 2.0 ± 0.3 nm in diameter as an optimum dimension in which maximum Tb3+ luminescence could be realized, with an increase in particle size resulting decreased Tb3+ luminescence from the Zn(Tb)S nanoparticles. These results have been rationalized with the relative energy level positions of the Tb3+ ground and excited states with respect to the valence and conduction band of the host material, with a smaller sized particle providing the favorable energetic alignment, with concomitant minimization of competitive pathways like autoionization of the excited electrons. This optimization of Zn(Tb)S nanoparticle size to achieve maximum Tb3+ luminescence from the nanoparticles lead us to select the Zn(Tb)S nanoparticles with smaller dimension for the current work. A smaller particle size as has been studied in the present work implies the importance of surface atoms, due to the larger surface to volume ratio. The possibility of co-localization of dopant moieties at the core (internal) and surface (external) sites in the nanoparticles connects with the objective of this study to evaluate the spatial location of the Ln3+ (dopant) in the interior and exterior sites of the ZnS (host) nanoparticles, both sites being populated. Such an indication of simultaneous population of Tb3+ in the Zn(Tb)S nanoparticles at the core and surface sites of the nanoparticles has been inferred from a Stern–Volmer (SV) quenching experiment, in which using methanol as the quencher for the Tb3+ luminescence resulted multiple slopes in the SV plot.28 The uniqueness of the analyses presented in the current study lies in the fact that while capping ligand infrared absorption selectively probes the surface localized Ln3+,28 dopant photoluminescence may arise from both the core and surface sites.
![]() | (1) |
![]() | ||
Fig. 1 The TEM and HRTEM images of the Zn(Tb)S particles with 10% doping are shown in panels (a) and (b) respectively. The corresponding energy dispersive X-ray spectrum (EDS) is shown in panel (c), confirming the presence of terbium in the nanoparticles. The presence of copper in the EDS spectrum is most likely associated with its presence in the TEM grid. The size distribution histograms of the Zn(Tb)S nanoparticles with 1%, 5%, 10%, 15%, 20% and 30% doping extent respectively are shown in panels (d–i). The data for the Zn(Tb)S nanoparticles with 15% doping extent has been reported in our earlier work.27 |
It is important to appreciate that while the identification of the presence of lanthanide cations in the purified Zn(Ln)S nanoparticles by either EDS spectroscopy or inductively coupled plasma atomic emission spectroscopy (ICPAES) (an alternate but similar approach to determine the elemental composition of a species) confirms the presence of a particular element in the nanoparticles, these techniques do not differentiate between the core and surface localized dopant moieties. A detailed insight on structural perspective; dopants being in the substitutional, interstitial or surface related sites of the lanthanide cation incorporated II–VI semiconductor nanoparticles are not yet fully characterized. These structural aspects will be reported comprehensively in a future study.
![]() | (2) |
![]() | ||
Fig. 2 FTIR spectra of the Zn(Tb)S nanoparticles in the cabroxylate stretching region is shown in the left panel. An enlarged part of the two IR absorption bands of consideration is shown in the right panel, in the absorbance scale. The corresponding spectra of ZnS nanoparticles and zinc stearate (ZnSt) are also shown. The spectra for the ZnS and Zn(Tb)S nanoparticles with 15% doping extent has been reported first in our earlier work.28 |
System | Capping ligand infrared absorption | Dopant photoluminescence | |
---|---|---|---|
R | ΦTb | a1c | |
a The values reported are the average and standard deviation from multiple measurements. For the photoluminescence lifetime decay fitting, the adjusted R2 values were found to be ≥0.994.b ΦT refers to the Tb3+ luminescence quantum yield.c a1 refers to the relative amplitude of the shorter luminescence lifetime component, with the shorter (τ1) and longer (τ2) lifetime components fixed at 1.0 and 3.7 ms respectively. The sum of relative amplitudes (a1 and a2) has been scaled to unity. | |||
Zn(Tb)S 1% | 0.37 ± 0.05 | (0.008 ± 0.002) × 10−3 | 0.40 ± 0.03 |
Zn(Tb)S 5% | 0.89 ± 0.18 | (0.24 ± 0.07) × 10−3 | 0.34 ± 0.02 |
Zn(Tb)S 10% | 0.65 ± 0.13 | (0.50 ± 0.02) × 10−3 | 0.32 ± 0.03 |
Zn(Tb)S 15% | 1.56 ± 0.12 | (2.0 ± 0.3) × 10−3 | 0.36 ± 0.02 |
Zn(Tb)S 20% | 3.10 ± 0.14 | (2.2 ± 0.4) × 10−3 | 0.41 ± 0.01 |
Zn(Tb)S 30% | 3.20 ± 0.14 | (2.3 ± 0.5) × 10−3 | 0.42 ± 0.01 |
ZnS | 0.15 ± 0.01 | — | — |
It is noteworthy that while McQuillan and co-workers40 reported the carboxylate symmetric stretching band at 1407 cm−1 in the oleate salt, the corresponding band appears at 1416 cm−1 in the oleate capped CdS quantum dots with an emergence of an additional IR absorption band at 1378 cm−1 (which was originally not assigned to specific bond vibration in their study). This band at 1378 cm−1 in their study is consistent with the emergence of an additional IR absorption band at 1372 cm−1 in the stearate capped zinc sulfide nanoparticles studied in the current work. Also, our independent effort to make oleate capped CdSe nanoparticles (Fig. S3 and S4†) resulted R values of 1.06 ± 0.02 and 2.45 ± 0.30 for the CdSe and Cd(Tb)Se nanoparticles respectively, suggesting the dopant Tb3+ induced increase of the 1376 cm−1 IR absorption band with respect to the intrinsic carboxylate symmetric stretching band located at 1410 cm−1. Moreover, the attenuation of carboxylate asymmetric and symmetric stretching bands at 1538 and 1398 cm−1 respectively with the formation of Zn(Tb)S nanoparticles, compared to the corresponding signals in the unbound stearate moiety observed in the current work is also consistent with the report by McQuillan and co-workers.40 These results collectively indicate that the attenuation of carboxylate asymmetric and symmetric stretching bands (as judged by the relative intensity ratio of these bands to the long chain methylene IR absorption band) in the IR absorption spectrum with an emergence of an additional band at lower energy with the formation of nanoparticles could be considered as general observation for carboxylate ligand capped II–VI semiconductor nanoparticles.
Additionally, we note that an analysis of nanoparticle formation induced changes in the IR absorption characteristics of capping ligand and its possible tuning in presence of dopant moieties is theoretically possible for nanoparticles capped with other ligands having different functional groups. However, such an analysis first requires an understanding of the spectral modulation of capping ligand IR absorption characteristics in nanoparticles, compared to the unbound capping ligand molecule. We anticipate that the nanoparticle formation induced changes and its potential tuning of the capping ligand IR absorption characteristics could be dependent on the identity of the ligand. For example, our study on 1-thioglycerol capped zinc sulfide nanoparticles have shown an attenuation of S–H stretching band intensity at 2556 cm−1 in both the undoped and Ln3+ [Ln = Sm, Eu, Tb, Dy] incorporated nanoparticles (with respect to that in the 1-thioglycerol molecule) and an Ln3+ induced change in IR absorption characteristics have been observed in the region of 615–635 cm−1, which was assigned to a C–S stretching band.29 Such a detailed analysis of capping ligand IR absorption characteristics with various functional groups in the ligand moiety is beyond the scope of the current work.
In the perspective of populating the sites by Tb3+ in the Zn(Tb)S nanoparticles with increase in dopant extent, at least two possibilities are relevant, (i) sequential gradient population of surface from core sites; that is, at lower dopant extent the Tb3+ locate at the core sites of the Zn(Tb)S nanoparticles and with an increase in dopant concentration the surface sites gradually gets populated and (ii) random population in both core and surface sites; that is, the core and surface sites of the nanoparticles gets populated simultaneously in a range of dopant concentration. The localization of the capping ligands being on the surface of the nanoparticles, it is reasonable to assume that the capping ligand IR absorption is not sensitive to the Tb3+ cations that are located at the core of the Zn(Tb)S nanoparticles and the observed trend is a reflection of the steady-state population of Tb3+ cations that are located on or near the surface of the Zn(Tb)S nanoparticles. Accordingly the capping ligand IR absorption selectively probes the concentration of surface located dopants and hence is unable to differentiate the two cases mentioned above. Dopant luminescence, on the other hand, is sensitive to both the core and surface located Tb3+ cations and hence would be able to differentiate between the two possibilities. The Tb3+ cations that are located on or near the surface of the Zn(Tb)S nanoparticles being more exposed to the immediate ligand and solvent molecules would be more prone to the non-radiative decay mechanisms and hence are less bright compared to the corresponding core located counterpart.
![]() | ||
Fig. 3 The steady state luminescence spectra of the Zn(Tb)S nanoparticles are shown in the top (left, right) and bottom (left) panels. The spectra in the bottom (left) panel have been normalized with respect to the relative quantum yield value, that is the areas under the spectra correspond to the quantum yield values reported in Table 1. The corresponding Tb3+ luminescence decay profiles are shown in the bottom (right) panel. |
The luminescence excitation spectrum while monitoring the Tb3+ emission at 545 nm provides an avenue to evaluate the excitation pathways in order to sensitize the Tb3+ luminescence in a given system. It has been argued previously that the luminescence sensitization by semiconductor nanoparticles (optical antenna effect) is an important path for Tb3+ luminescence sensitization in the Zn(Tb)S nanoparticles, while the contribution from direct excitation of intra-configurational 4f–4f excitation bands are negligible, see Fig. S5 in the ESI.†2,27 The corresponding excitation spectra in all the Zn(Tb)S nanoparticles as a function of Tb3+ doping extent corroborate with the general observation from the previous studies. It is remarkable to note that the normalized excitation spectra reveal a nearly identical profile for doping extent upto 15%, which has been attributed to the size independence of the Zn(Tb)S nanoparticles with varying dopant extent. This rules out any variation in luminescence of different Zn(Tb)S nanoparticles that may originate from the difference in size of the nanoparticles27 and suggests that the observed luminescence is a major reflection of spatial location of Tb3+ in the Zn(Tb)S nanoparticles. The corresponding excitation spectra for the Zn(Tb)S nanoparticles with doping extent 20% and 30% respectively reveal significant difference in the spectral profile compared to the corresponding cases upto 15% doping extent. The increase in dopant extent increases the relative contribution of direct excitation bands, as judged by the increase in relative excitation intensity in the wavelength range of 350–380 nm. The increase in the proportion of direct excitation bands in the Zn(Tb)S nanoparticles with 20% and 30% doping extent corroborates with more surface related Tb3+ and is consistent with near constancy of the Tb3+ emission quantum yield values for the Zn(Tb)S nanoparticles with 15%, 20% and 30% doping extent respectively. Moreover, a comparison of spectral profiles (Fig. S6†) reveals the full width at half maximum (FWHM) values in the range of 348 ± 13 cm−1 for the Tb3+ emission at 545 nm for the Zn(Tb)S nanoparticles with varying doping extent. This at least grossly suggests that the observed Tb3+ emission most likely monitors similar chemical species throughout the doping extent studied and is not having significant contribution from aggregated Tb3+ moiety.
The Tb3+ luminescence lifetime decay profiles were found to be bi-exponential. The decay profiles, I(t), were fitted by a sum of two decaying exponentials with time constant values fixed at 1.0 ± 0.05 and 3.7 ± 0.1 ms respectively according to eqn (3), where the shorter and longer lifetime components have been attributed to originate from the surface (lesser protected) and core (more protected) localized dopant cations in the nanoparticles and is in corollary with our previous reports.2,26,27 The lifetime components were fixed in the fitting process in order to better compare the relative amplitudes. Table S1† summarizes a representative detailed comparison of fitting parameters from fitting with all parameters varying and only relative amplitudes varying in the fitting process. The decrease in average luminescence lifetime with increase in dopant extent likely reflects an increase in non-radiative processes that is induced by the concentration quenching effect resulted by high local dopant concentration.
![]() | (3) |
It is remarkable to note that in a site-selective spectroscopic analysis at 10 K, Chen and co-workers36 have observed two lifetime components of 0.69 and 1.16 ms respectively in Eu3+ doped ZnO nanocrystals, which has been correlated to surface and inner lattice sites respectively. The observation of two lifetime components even at 10 K further emphasizes the importance of understanding the properties associated with the core and surface related sites. The trend observed from Tb3+ luminescence lifetime analysis also corroborate with a simultaneous population of core and surface sites in the Zn(Tb)S nanoparticles, in which similar shorter lifetime amplitude has been expected. The amplitude of the shorter lifetime component (a1), as summarized in Table 1, was found to be similar for the doping extent upto 15%, without a noticeable systematic trend with respect to the doping extent. A case where the surface sites preferentially and sequentially gets populated with an increase in dopant extent would generate an increase in the amplitude of the shorter lifetime component, due to the higher non-radiative decay processes originating from the surface related sites of the nanoparticles. The additional non-radiative decay pathways originate from the interaction of surface localized Tb3+ with the vibrational overtones of the high frequency vibrations at the nearby ligand and solvent molecules. This has indeed been observed for the nanoparticles with 20% and 30% doping extent respectively, where a distinct increase in the non-radiative rate is evident, as judged by the increase in the amplitude of shorter lifetime component and is consistent with the appearance of some direct 4f–4f intra-configurational excitation bands observed in the steady-state excitation spectra while monitoring the Tb3+ luminescence. We anticipate that the quenching effect by the surface defects of the Zn(Tb)S nanoparticles would be similar, as the surface to volume ratio is same that is guided by the near constancy of the Zn(Tb)S nanoparticle's size with varying Tb3+ extent (an average particle diameter of 2.2 ± 0.5, 2.0 ± 0.4, 1.9 ± 0.4, 2.0 ± 0.3, 2.1 ± 0.3 and 2.2 ± 0.3 nm for the 1%, 5%, 10%, 15%, 20% and 30% doping extent respectively).
For spherical ZnS particles (Fig. 1) with wurtzite phase,28 the diameter (d) is given by
![]() | (4) |
Volume of the nanoparticle (VNP) is given by
VNP = NTVZnS | (5) |
Simple mathematical calculation reduces eqn (5) to
![]() | (6) |
The surface area (SNP) of the nanoparticles is represented by
SNP = 4πrNP2 | (7) |
The number of surface localized atoms (NS) is given by
![]() | (8) |
The NT and NS values for a nanoparticle with 2 nm diameter are tabulated in Table 2. Importance of significant contribution from surface localized atoms is obvious and relates to small size of the nanoparticles. The total number of cationic and anionic ingredients in the nanoparticles has been estimated assuming a simple 1:
1 stoichiometry. The total number of Tb3+ [N(Tb3+)T] has been calculated according to the doping percentage. The similar luminescence lifetime values observed in a range of doping extent of the Zn(Tb)S nanoparticles (Table 1) indicates that the core and surface sites of the nanoparticles are randomly populated by the dopants. Based on this both the core (C) and surface (S) localized Tb3+ cations [N(Tb3+)C/S] have been calculated according to the NS and NT proportion for the given particle diameter.
System | Number of atoms (N) NT = 367, NS = 209 | Capping ligand infrared absorption | [Tb3+](x)C | Dopant photoluminescence | ||||
---|---|---|---|---|---|---|---|---|
N[Ln3+]T | N[Ln3+]C | N[Ln3+]S | [Tb3+](x)S | ΦS (%) | ΦS | ΦC | ||
Zn(Tb)S 1% | 2 | 1 | 1 | 0.22 ± 0.05 | 0.22 ± 0.05 | 15 ± 2 | (0.0012 ± 0.0003) × 10−3 | (0.0068 ± 0.0017) × 10−3 |
Zn(Tb)S 5% | 9 | 4 | 5 | 0.74 ± 0.18 | 0.59 ± 0.18 | 12 ± 1 | (0.03 ± 0.01) × 10−3 | (0.21 ± 0.06) × 10−3 |
Zn(Tb)S 10% | 18 | 8 | 10 | 0.50 ± 0.13 | 0.40 ± 0.13 | 11 ± 1 | (0.06 ± 0.01) × 10−3 | (0.44 ± 0.02) × 10−3 |
Zn(Tb)S 15% | 28 | 12 | 16 | 1.41 ± 0.12 | 1.06 ± 0.12 | 13 ± 1 | (0.26 ± 0.04) × 10−3 | (1.74 ± 0.26) × 10−3 |
Zn(Tb)S 20% | 37 | 16 | 21 | 2.95 ± 0.14 | 2.25 ± 0.14 | 16 ± 1 | (0.35 ± 0.07) × 10−3 | (1.85 ± 0.34) × 10−3 |
Zn(Tb)S 30% | 55 | 24 | 31 | 3.05 ± 0.14 | 2.36 ± 0.14 | 16 ± 1 | (0.37 ± 0.08) × 10−3 | (1.93 ± 0.42) × 10−3 |
The presence of capping ligands on the surface of the nanoparticles infers that the infrared absorbance of the capping ligand probes the concentration of the surface localized Ln3+ in the Zn(Ln)S nanoparticles. As shown in Fig. 2 and Table 1, the infrared absorption signal at 1372 cm−1 emerges with an increase in the Tb3+ doping extent. Accordingly, this analysis hypothesizes that the trend in R value (eqn (2)) from the Zn(Tb)S and ZnS nanoparticles may be used as a gauge to monitor the concentration of surface localized Tb3+ ions, [Tb3+](x)S, in the Zn(Tb)S nanoparticles, mathematically this has been expressed by
[Tb3+](x)S ∝ (R[Zn(Tb)S] − R[ZnS]) ∝ N(Tb3+)S | (9) |
The construction of eqn (9) considers the molar extinction coefficient of Zn(Tb)S nanoparticles as being independent of doping extent and that the pathlength of the samples remain same.
From luminescence spectroscopy the steady-state emission intensity at 545 nm (5D4 → 7F5 of Tb3+) [ISS(545 nm)] is given by,
![]() | (10) |
Correspondingly, the contribution from the surface localized Tb3+ [ΦS (%)] in the overall quantum yield (ΦT) is expressed by
![]() | (11) |
It is to note that the ΦT [as computed considering the 5D4 → 7Fn (n = 3–6) Tb3+ emission bands] would be proportional to ISS(545 nm). Accordingly, from the ΦT values the corresponding ΦS and ΦC have been calculated. It has been inferred that [Tb3+](x)S is responsible for the ΦS. The corresponding core localized Tb3+ correlate with the ΦC.
In order to evaluate the trend in Tb3+ luminescence quantum yield values with the relative concentrations of core (more protected) and surface (lesser protected) localized Tb3+, the ΦS/C are plotted as a function of [Tb3+](x)S/C. The corresponding plot from such an analysis has been shown in Fig. 4. In a case where the effect of Tb3+ localization in either the core or surface of the Zn(Tb)S nanoparticles merely monotonously increases; the Tb3+ emission, a plot as shown in Fig. 4, would reveal a linear dependence. However, significant deviation from linear behavior has been observed for both the core and surface localized Tb3+. Moreover, the extent of increase in the Tb3+ luminescence quantum yield for the surface localized dopants was found to be more flattened compared to that for the corresponding core localized counterpart. This implies that an increase in concentration of dopant (Tb3+) moieties does not result the core and surface localized Tb3+ equally brighter in the Zn(Tb)S nanoparticles. The less dramatic concentration dependence on the surface localized dopants is most likely associated with the environmental quenching effect from the capping ligand and solvent molecules in the immediate local environment.
The second order polynomial fitting while represents a satisfactory mathematical representation of the data with qualitative indication of deviation from linear increase of the quantum yield values (which most likely results from Tb3+ that interacts in a confined spatial location thereby introducing non-radiative decay mechanisms); we do not attribute any specific physical model associated with the second order polynomial fitting. While the real picture may well have contributions from higher order polynomials, we intend not to complicate the physical interpretation without prior knowledge of the exact dependence. In short, the second order polynomial fitting broadly captures the experimental trend that while the initial increase in Tb3+ incorporation content increases the Tb3+ luminescence quantum yield, it deviates from linearity beyond a certain Tb3+ incorporation extent. Moreover, the parameters of the fitting indicates that with the progression of the relative amount of core and surface localized Tb3+ moiety, the quantum yield values for the core localized components have been found to be at least 10 times more than the corresponding changes of the surface localized counterparts. The lesser sensitivity of the surface localized components most likely originates from the quenching of Tb3+ by adjacent ligand and solvent molecules that induces additional non-radiative decay pathways. The parameters from the fitting provides an idea of the relative changes of the core and surface localized Tb3+ in the Zn(Tb)S nanoparticles studied, however these numbers are not intended to have a quantitative meaning in an absolute scale due to the proportionality of Tb3+ concentration extracted from IR spectra with the corresponding numbers of Tb3+ (eqn (9)) (vide infra).
The analysis to probe the core (more protected) and surface (lesser protected) localized Tb3+ in the Zn(Tb)S nanoparticles studied with the corresponding luminescence properties presented in the current work relies on the following assumptions, the validity of which is justified by the following arguments; (a) the infrared absorption of the capping ligand probes the steady-state population of the surface localized Tb3+ which is guided by the confinement of the capping ligand moiety on the surface of the nanoparticles. Various other researchers have investigated the surface chemistry of nanoparticles by monitoring the IR spectral characteristics of the nanoparticles;40,45–58 (b) the two luminescence lifetime components can be correlated with the core (more protected) and surface (lesser protected) Tb3+ moieties, which is guided by the fact the surface sites being less ordered compared to the core sites induces additional non-radiative decay pathways due to the vibrational overtones of the adjacent ligand and solvent molecules. It is emphasized that the co-ordination environment of Tb3+ in the core and surface sites would be different which results the two luminescence lifetime components. Various works relate multiple Ln3+ luminescence lifetime components to multiple co-ordination environments.5,59–61 To this end, we further stress that while a fitting with two lifetime components might present a simplistic representation of the luminescence lifetime decay profiles, most likely there exists a distribution of lifetime components resulting from a gradient population of Tb3+ from the core towards the surface of the nanoparticles;26,28 (c) the proportionality of the contribution from the surface localized components extracted from the IR spectra to the corresponding number of Tb3+ extracted from a simple analysis from the diameter of the nanoparticles (eqn (9)). This proportionality imposes an important implication of the data presented in Fig. 4 being a relative trend of the luminescence characteristics of the core and surface localized components without necessarily emphasizing on the absolute values of the fitting parameters (see Fig. S7† for such a representative analysis).
System | Capping ligand infrared absorption | Dopant Photoluminescence | |
---|---|---|---|
R | ΦT | a1b | |
a The values reported are the average and standard deviation from multiple measurements. For the photoluminescence lifetime decay fitting, the adjusted R2 values were found to be ≥0.994.b The lifetime values were kept fixed at 0.80 ± 0.05 and 3.7 ± 0.1 ms respectively during the fitting process. | |||
Zn(Eu)S 5% | 1.24 ± 0.23 | (0.23 ± 0.04) × 10−4 | 0.66 ± 0.04 |
Zn(Eu)S 15% | 1.56 ± 0.11 | (2.6 ± 0.5) × 10−4 | 0.70 ± 0.02 |
Zn(Eu)S 30% | 2.00 ± 0.20 | (5.1 ± 0.2) × 10−4 | 0.75 ± 0.03 |
Footnote |
† Electronic supplementary information (ESI) available: TEM images, representative FTIR and luminescence spectra, luminescence lifetime decay fitting parameters have been presented in the electronic supplementary information. See DOI: 10.1039/c6ra19823f |
This journal is © The Royal Society of Chemistry 2016 |