Mn-Doped Quinary Ag-In-Ga-Zn-S Quantum Dots for Dual-Modal Imaging.

Doping of transition metals within a semiconductor quantum dot (QD) has a high impact on the optical and magnetic properties of the QD. In this study, we report the synthesis of Mn2+-doped Ag-In-Ga-Zn-S (Mn:AIGZS) QDs via thermolysis of a dithiocarbamate complex of Ag+, In3+, Ga3+, and Zn2+ and of Mn(stearate)2 in oleylamine. The influence of the Mn2+ loading on the photoluminescence (PL) and magnetic properties of the dots are investigated. Mn:AIGZS QDs exhibit a diameter of ca. 2 nm, a high PL quantum yield (up to 41.3% for a 2.5% doping in Mn2+), and robust photo- and colloidal stabilities. The optical properties of Mn:AIGZS QDs are preserved upon transfer into water using the glutathione tetramethylammonium ligand. At the same time, Mn:AIGZS QDs exhibit high relaxivity (r 1 = 0.15 mM-1 s-1 and r 2 = 0.57 mM-1 s-1 at 298 K and 2.34 T), which shows their potential applicability for bimodal PL/magnetic resonance imaging (MRI) probes.

Introduction

Single-modal imaging techniques such as photoluminescence (PL) imaging and magnetic resonance imaging (MRI) are widely used as diagnostic tools and for the monitoring of biological processes in living organisms. However, each imaging modality has its inherent advantages and limitations. PL imaging is relatively easy to use and exhibits a high sensitivity but its resolution is weak to get three-dimensional (3D) anatomical information.[1,2] On the contrary, MRI suffers from its low sensitivity but allows the 3D imaging of opaque and soft tissues with high spatial resolution and deep tissue penetration.[3,4] Thus, the development of bimodal probes combining sensitivity and spatial resolution is of high interest for whole body imaging.

In recent years, the development of nanoparticles with both PL and MR functionalities has been the object of tremendous efforts.[5−8] For more than 20 years, semiconductor quantum dots (QDs) have received high attention as optical contrast agents. Compared to conventional organic fluorophores, QDs exhibit unique optical properties including broad absorption bands, size- and composition-tunable PL emission, and high brightness and photostability.[9−11] Numerous strategies like the assembly of QDs with paramagnetic Fe3O4 nanoparticles,[12−14] the deposition of a paramagnetic coating at the surface of QDs, or the reverse option consisting in the deposition of QDs at the surface of Fe3O4 nanoparticles,[15−17] or finally, the conjugation of QDs with Gd3+ chelates[18−20] have been developed to prepare probes for bimodal imaging. However, some of these nanoparticles have shortcomings such as the decrease of the PL intensity after association with the magnetic center, a high hydrodynamic diameter, or tedious synthesis methods. Another strategy is the doping of QDs with paramagnetic cations, usually, Gd3+,[21,22] but the use of Fe3+ has also been reported.[23] Although Gd3+ complexes are usually considered safe for clinical applications,[24] Gd3+ may exhibit toxicity and its administration may conduct to the development of nephrogenic systemic fibrosis for patients suffering from severe renal impairment.[25] Mn2+ has lower toxicity, exhibits a relaxivity similar to that of Gd3+ and has been demonstrated to be a valuable alternative as a dopant to Gd3+ for engineering nanocrystals for bimodal imaging. Binary CdS,[26] PbS,[27] ZnS,[28,29] and ZnSe,[30−32] ternary In–Zn–S,[33] and quaternary Cu–In–Zn–S (CIZS) and Cu–In–Zn–Se (CIZSe) QDs[34−36] have successfully been doped with Mn2+ ions for combined PL and MR imaging.

We recently reported the synthesis of highly luminescent quinary Ag–In–Ga–Zn–S (AIGZS) QDs via thermal decomposition in oleylamine (OAm) of a single AgInGa2–Zn2(S2CN(C2H5)2)4 dithiocarbamate precursor.[37] In this study, we used the heavy-metal-free AIGZS QDs as host nanocrystals to incorporate Mn2+ dopants, and the dual functional (PL/MRI) properties of the doped QDs were investigated. Our results showed that Mn-doped AIGZS (Mn:AIGZS) QDs exhibited the Mn2+-related 4T1 → 6A1 PL with a PL quantum yield (PL QY) of up to 41.3% and a high PL lifetime of up to 887.9 μs. The dependence of the PL QY and the PL lifetime of Mn:AIGZS QDs on the Mn2+ loading was demonstrated. Mn:AIGZS QDs could easily be transferred into aqueous solution using the hydrophilic glutathione tetramethylammonium (GTMA) ligand without alteration of their optical properties and were demonstrated to show high relaxivity (r1 = 0.15 mM–1 s–1 and r2 = 0.57 mM–1 s–1 at 2.34 T). Due to their long PL lifetime that allows to improve the temporal discrimination of the PL signal from the autofluorescence background, to their photostability and to their high relaxivity, Mn:AIGZS QDs should have attractive applications in bioimaging.

Results and Discussion

Synthesis of Mn-Doped AIGZS QDs and Optical Properties

Mn-doped Ag–In–Ga–Zn–S (Mn:AIGZS) QDs were synthesized in a two-stage process.[37] First, a Ag, In, Ga, and Zn dithiocarbamate complex (AgInGa2–Zn2(S2CN(C2H5)2)4) used as a precursor was prepared using a Ag/In/Ga/Zn ratio of 0.25:0.875:0.875:2 to obtain AIGZS QDs that did not emit light in the same spectral window as Mn-doped QDs. The loading in the MnSt2 dopant was calculated relative to the total amount of metal cations and varied from 1 to 10%. Next, the AgInGa2–Zn2(S2CN(C2H5)2)4 precursor and MnSt2 were thermally decomposed at 220 °C in OAm used as a solvent and as a capping ligand (Scheme ). After 10 min of heating for completion of QD growth, the reaction mixture was centrifuged to remove large-sized particles. The surface defects were repaired by reheating Mn:AIGZS QDs of the colloidal solution at 220 °C for 30 min.

Scheme 1

Schematic Representation of Mn-Doped AIGZS QD Synthesis

In preliminary experiments, we also evaluated a synthetic process involving a dithiocarbamate complex containing Ag+, In3+, Ga3+, Zn2+, and Mn2+ metal cations. As can be seen in Figure S1, similar optical properties were obtained for the Mn(5) and Mn(10):AIGZS QDs using the two synthetic approaches. The loading in Mn2+ was found to be much more convenient to control when Mn(St)2 was added to the AgInGa2–Zn2(S2CN(C2H5)2)4 precursor, and the synthetic process described in Scheme is kept for further studies.

The accurate compositions of OAm-capped AIGZS and Mn:AIGZS QDs are determined by inductively coupled plasma-optical emission spectrometry (ICP-OES) (Table ). Using a Ag/In/Ga/Zn precursor ratio of 1:3.5:3.5:8 results in Ag1In3.07Ga3.33Zn7.40 QDs, which indicates that all metal cations are well incorporated in AIGZS QDs. The Ag/In/Ga ratio is weakly affected by the increase of the Mn loading. A slight decrease in the In content and a slight increase in the Ga content are observed when increasing the loading in Mn from 1 to 10%. The influence of the doping on the Zn content is more pronounced. The decrease of the Zn content with the increase of the Mn doping suggests that Mn2+ substitutes Zn2+ in the AIGZS crystal lattice. Finally, the amount of Mn2+ ions loaded in the nanocrystals is lower than the amount used for the synthesis (0.81, 1.96, 4.07, and 5.01% of Mn2+ are detected when using 1, 2.5, 5, and 10% of Mn2+, respectively).

Table 1

Compositions of OAm-Capped AIGZS and Mn:AIGZS QDs

Mn loading (%)	experimental composition
0	Ag1In3.07Ga3.33Zn7.40
1	Ag1In3.06Ga3.33Zn7.37Mn0.12
2.5	Ag1In3.03Ga3.40Zn7.11Mn0.26
5	Ag1In3.03Ga3.44Zn6.81Mn0.55
10	Ag1In2.96Ga3.48Zn6.62Mn0.7

Figure a shows the UV–vis absorption and the PL emission spectra of AIGZS and Mn-doped AIGZS QDs with loadings in Mn2+ of 1, 2.5, 5, and 10%. The absence of distinctive absorption is typical for multinary QDs and originates from heterogeneous composition distributions among the dots and thus due to the distribution of vibrational states in AIGZS QDs.[37−39] The absorption spectral profiles of AIGZS and Mn:AIGZS QDs are almost similar, indicating that both the diameter and the composition of the dots are not significantly modified by the doping. The band gap values of the dots are determined from the Tauc plots of (αhν)2 vs hν, where α is the absorption coefficient and hν is the photon energy, and are found to decrease from 2.94 for AIGZS QDs to 2.58 eV for Mn(10):AIGZS QDs (Figure b). This decrease in the band gap energy with the increase in Mn2+ content confirms the substitution of Zn2+ ions by Mn2+ in the AIGZS crystal lattice previously observed by ICP as the band gap energy of MnS (3.1–3.2 eV) is lower than that of ZnS (3.6 eV). These results also suggest modifications of the electronic states of the AIGZS QDs by the Mn dopant.

Figure 1

(a) UV–visible and normalized PL emission spectra of AIGZS and Mn:AIGZS QDs (the inset is a digital photograph of the QDs taken under UV light illumination) and (b) determination of optical band gaps of Mn:AIGZS QDs.

Multinary QDs such as CIZS and AIGZS differ from conventional binary QDs as they do not exhibit the excitonic recombination. The relatively broad PL emission with a large Stokes shift originates from surface states (vacancies and dangling bonds) and interstitial states (interstitial atoms and vacancies) that may act as a donor (D) and acceptor (A) for the electron–hole recombination.[37−40] For undoped AIGZS QDs, the D–A PL emission peak is located at 515 nm and a surface defect-related emission centered at ca. 650 nm can also be observed. The PL QY of AIGZS QDs is of 14.3% in toluene. With the introduction and then increase of the Mn2+ doping, the D–A emission is gradually quenched and an intense orange-red fluorescence appears at ca. 612 nm regardless of the Mn2+ doping percentage (Figure a). A marked increase of the PL QY is observed for Mn:AIGZS QDs (37.9, 41.3, 39.1, and 24.9% for the nanocrystals doped with 1, 2.5, 5, and 10% Mn, respectively). The signal at ca. 612 nm can be attributed to Mn d-state emission, as demonstrated below, by the analysis of the QD PL lifetime. The full width at half-maximum of the Mn emission is ca. 85 nm, a value consistent with previous reports.[41]

Mn-doped quaternary CIZS and Ag–In–Zn–S (AIZS) QDs have been the subject of several studies in recent years and their optical emission is particularly found to be dependent on the energy band gap, on the position of the valence and conduction bands of the host, and on the radial position of the dopant in the nanocrystals. Usually, the Mn-related emission is attributed to the radiative transition from the low-spin excited state of Mn2+ (4T1) to its high-spin ground state (6A1). This emission is observed for CIZS or AIZS QDs with an energy band gap larger than that of this Mn2+-related transition energy (dual D–A and Mn emissions can also be observed).[34,36,41−43] On the contrary, only the D–A emission is observed when the host QDs exhibit a band gap smaller than the Mn emission.[35,38,44] These results will be discussed below in the case of AIGZS QDs.

We also investigated the influence of the Mn doping on the PL lifetime of AIGZS QDs. Time-resolved PL emission was recorded at the PL emission maxima of AIGZS and Mn:AIGZS QDs (Figure a), and the results are shown in Figure and Table . For all of the samples, the PL decay curves are best fitted using a biexponential function I(t) = A1 exp(−t/τ1) + A2 exp(−t/τ2), where τ1 and τ2 are the time constants of the PL and A1 and A2 are the relative weights of the decay components. The average PL lifetimes (τav) are determined using τav = (A1τ1 + A2τ2)/(A1 + A2). The PL lifetimes are associated with different electron–hole recombination pathways and can be used to study the PL mechanisms of the dots. Characteristic decay times in the range of 100 ns, here considered as a fast component, and in the range of 100 μs to 1 ms, here considered as a slow component, can be extracted from the experimental curves.

Figure 2

Time-resolved PL spectra of AIGZS and Mn:AIGZS QDs at the PL maximum wavelength.

Table 2

PL QYs, τ1 and τ2 Time Constants, and the Contributions of Decays A1 and A2 of Undoped and Mn:AIGZS QDs

sample	PL QYs (%)	A1	τ1 (μs)	A2	τ2 (μs)	τav (μs)
AIGZS	14.3	0.829	66.3 × 10–3	0.171	262.3 × 10–3	99.9 × 10–3
Mn(1):AIGZS	37.9	0.482	232.5	0.518	1497.4	887.9
Mn(2.5):AIGZS	41.3	0.577	192.4	0.423	1212.3	624.3
Mn(5):AIGZS	39.1	0.643	127.5	0.357	946.3	419.9
Mn(10):AIGZS	24.9	0.702	105.7	0.298	734.5	293.2

For undoped QDs, the PL decay is characterized by fast components, giving rise to a τav value close to 100 ns, which can be associated to surface trap-state recombination or to electron–hole recombination from donor–acceptor states (D–A recombination). A marked increase of τav is observed upon Mn doping (up to 887.9 μs for Mn(1):AIGZS QDs), indicating that the PL emission is not dominated by the D–A recombination but by Mn dopants. The millisecond lifetime measured is indeed related to the 4T1 → 6A1 transition of Mn2+ ions[45] and further confirms their doping in AIGZS nanocrystals. A decrease of τav from 887.9 to 293.2 μs is observed when increasing the Mn loading from 1 to 10% likely due to Mn–Mn interactions and/or spin coupling.[44]

The red shift in the PL emission observed compared to conventional binary Mn-doped QDs (ca. 585–590 nm) originates from the complex environment of Mn2+ ions in AIGZS QDs and from Mn–Mn coupling that conducts to the splitting between the 4T1 and 6A1 energy levels as previously observed for AIZS and CIZS QDs.[44,46] These Mn–Mn interactions and the defects generated in the AIGZS crystal lattice when increasing the Mn2+ loading are also responsible for the decrease of the PL QY and of the PL lifetime of the dots.

Based on the previously described results, possible mechanisms for AIGZS and Mn:AIGZS QDs are depicted in Figure . The broad PL emission with a large Stokes shift and the PL lifetime of ca. 100 ns indicate that the PL emission of AIGZS QDs originates from a recombination between intragap donor and acceptor levels, as commonly observed for ternary and quaternary semiconductor nanocrystals (Figure a). Due to the relatively wide band gap of AIGZS QDs (2.58 eV), 4T1 and 6A1 Mn states are, respectively, located below and above the donor and acceptor defect states of the AIGZS host (Figure b). Upon light excitation, the exciton confined within the donor and acceptor states transfers the energy to the Mn d-states and the Mn-related emission at ca. 612 nm is observed. This emission will not have been observed if we have used the AIGZS host with a lower band gap as described in our previous report.[38]

Figure 3

Schematic representation of the PL emission mechanism for (a) AIGZS and (b) Mn:AIGZS QDs.

Structure and Morphology

X-ray diffraction (XRD) patterns of OAm-capped AIGZS and Mn:AIGZS QDs are shown in Figure a. The broad diffraction peaks can be attributed to the nanocrystalline size of the dots. The peaks located at 28.63, 47.63, and 56.52° correspond to the (112), (204/220), and (312) planes of the tetragonal chalcopyrite phase, respectively, and no extra peak or impurities can be detected. The average diameter of AIGZS and Mn:AIGZS QDs was determined to be ca. 2.0 nm based on the Debye–Scherrer equation using the (112) diffraction peak. As can be seen from the figure, the peaks are shifted toward lower angles when the doping in Mn increases. Based on ICP-OES results, this likely originates from the diffusion of Mn2+ ions into the QD core, which leads to a lattice expansion since the ionic radius of Mn2+ (0.066 nm) is higher than that of Zn2+ (0.06 nm).

Figure 4

(a) XRD patterns of AIGZS and Mn:AIGZS QDs and (b) electron paramagnetic resonance (EPR) spectra of Mn:AIGZS QDs.

Electron paramagnetic resonance (EPR) spectra of Mn:AIGZS QDs further confirm the doping of Mn2+ ions into the AIGZS nanocrystals (Figure b). Though detectable, the six hyperfine splitting related to the coupling of the electronic spin with the 5/2 nuclear spin of Mn2+ are ill resolved, probably due to a broadening of the signal originating either from Mn–Mn dipolar interactions[47] and/or to the complex environment of Mn2+ surrounded by Ag+, Ga3+, In3+, and Zn2+ ions.[41] Indeed, it has been proposed that structural defects in multinary QDs such as AIGZS generate random perturbations to the Mn2+ electron–nuclear hyperfine coupling.[34] However, an estimation of the hyperfine constant A of 67.5 G can still be extracted from the partially resolved 6-line pattern, suggesting that at least part of the Mn2+ cations are doped into the AIGZS tetragonal lattice and not located at their surface.[41,48]

Mn:AIGZS QDs were further characterized by transmission electron microscopy (TEM). Figure a shows a representative TEM of Mn(2.5):AIGZS QDs, indicating that the nanocrystals are nearly spherical in shape with a fairly monodisperse size distribution (1.9 ± 0.5 nm by analyzing ca 100 nanocrystals in the TEM image). Similar results are observed for all Mn-doped AIGZS QDs, indicating that the average diameter of the dots is independent of the Mn2+ loading. The high crystallinity of the dots is confirmed from the selected area electron diffraction (SAED) patterns (inset of Figure a). A few larger particles composed of 2–3 nanocrystals and with sizes around 7–9 nm could also be observed on the TEM image. The lattice spacing measured is of ca. 0.31 nm and corresponds to the (112) diffraction plane of the tetragonal chalcopyrite phase (Figure b). Finally, the elemental composition of the sample is analyzed by energy-dispersive X-ray spectroscopy (EDX), which shows the presence of Ag, In, Ga, Mn, and S elements and further confirms that Mn2+ is successfully doped into AIGZS QDs (Figure c).

Figure 5

(a) TEM image of Mn(2.5):AIGZS QDs (the insets are the size distribution and the SAED pattern), (b) high-resolution TEM (HRTEM) image of Mn(2.5):AIGZS QDs, and (c) EDX spectrum of the nanocrystals.

The presence of Ag, In, Ga, Zn, Mn, and S elements is further confirmed by X-ray photoelectron spectroscopy (XPS) (see Figure S2 for the overview spectrum). The peaks at 367.67, 445.75, 452.28, 1118.26, 1021.79, and around 641.2 eV can be assigned to Ag 3d5/2, In 3d5/2, In 3d3/2, Ga 2p3/2, Zn 2p3/2, and Mn 2p3/2, respectively, and confirm that the elements are in their expected valence states (Ag+, In3+, Ga3+, Zn2+, Mn2+, and S2–) (Figure S3). The Mn 3p3/2 signal appears at 641.2 eV, which is consistent with values reported for Mn2+ doped into ternary or quaternary nanocrystals.[49]

Magnetic Properties

In the first set of experiments, magnetic measurements using vibrating sample magnetometry (VSM) were carried out for undoped and Mn:AIGZS QDs at 10 and 310 K (Figure a,b). As expected, undoped AIGZS QDs exhibit a diamagnetic behavior (blue curve). For Mn-doped QDs, the magnetization curves show that the samples exhibit superparamagnetic behavior as evidenced by values of coercivity and remanence close to zero, thus confirming the onset of the superparamagnetic regime at 10 K (Figure a). In contrast, the samples display a clear paramagnetic behavior at 310 K (Figure b). The relatively low Ms values (i.e the maximum value is 1.13 emu/g for the 10% doped sample at 310 K) are not surprising owing to the low amount of incorporated Mn compared to the other element. The increase of the magnetization with the Mn loading is nevertheless witnessed at both temperatures (Figure a,b). Besides, the influence of the temperature is clearly demonstrated as the magnetic properties are enhanced by decreasing the temperature regardless of the loading amount of Mn (see Figure c for Mn(10):AIGZS QDs and Figure S4 for the other doped nanocrystals). These results confirm the efficiency of the Mn doping to produce QDs combining fluorescence and superparamagnetic behavior.

Figure 6

M(H) loop curves of Mn:AIGZS QDs at (a) 10 K and (b) 310 K and (c) M(H) loop curves of Mn(10):AIGZS QDs at different temperatures.

Aqueous Phase Transfer

Mn:AIGZS QDs were transferred into an aqueous phase using glutathione tetramethylammonium (GTMA) that displaces the native hydrophobic OAm ligand and bounds to the surface of the dots via metal–S bonds (Figure a).[50] The resultant dispersion of AIGZS QDs is transparent, indicating that the nanocrystals are well dispersed in aqueous solution (Figure b). AIGZS@GTMA QDs can be stored for months in water without alteration of their optical properties, which indicates high colloidal stability.

Figure 7

(a) Schematic representation of the water transfer of AIGZS QDs using the GTMA ligand and (b) digital photograph taken under ambient light of the dots before and after the ligand exchange.

A representative TEM image of Mn(2.5):AIGZS@GTMA QDs shows that the nanocrystals remain well-dispersed after the ligand exchange and that their average diameter is similar to that of native OAm-capped QDs (2.0 ± 0.6 nm) (Figure a). Their tetragonal crystalline structure also remains unaffected as can be seen in Figure b. The UV–visible and the PL emission spectra of Mn(2.5):AIGZS QDs selected as a representative indicate that the optical properties of Mn(2.5):AIGZS QDs are well preserved after transferring to water (Figure c). Upon ligand exchange, no shift of PL emission wavelength could be detected but a slight drop of PL QYs was observed (35.3, 38.4, 36.8, and 21.7% for AIGZS doped with 1, 2.5, 5, and 10% Mn, respectively).

Figure 8

(a) TEM image of Mn(2.5):AIGZS@GTMA QDs (the inset is the size distribution), (b) XRD patterns, (c) UV–visible absorption and PL emission spectra of Mn(2.5):AIGZS QDs before and after the ligand exchange.

The ζ-potential of Mn(2.5):AIGZS QDs is −58.5 ± 1.2 mV due to the negatively charged carboxylate functions of the GTMA ligand (Figure a). Their hydrodynamic diameter is of 13.7 ± 1.9 nm with a polydispersity index (PDI) of ca. 0.11, further confirming that the QDs are well dispersed in aqueous solution (Figure b). AIGZS@GTMA QDs were also found to be stable between pH 4 and 12 (Figure c). No significant shift of the PL emission wavelength is observed in this pH range but a marked decrease of the PL intensity occurs below pH 4 due to the protonation of the carboxylate functions, leading to the aggregation of the dots. A similar decrease is observed above pH 12 due to the decomposition of the QD surface into metal hydroxides. The dangling bonds formed act as traps for photogenerated electrons and quench the PL.[51] We also investigated the colloidal stability of Mn(2.5):AIGZS QDs by dispersing the dots in NaCl solutions with concentrations varying from 0.2 to 1 M (Figure S5). After 48 h, no aggregation was observed and the dots retained almost the same PL intensity regardless of the ionic strength. Finally, Mn(2.5):AIGZS QDs dispersed in water exhibit high photostability under the continuous irradiation of a Hg–Xe lamp (intensity of 50 mW/cm2). The PL intensity remained stable during the 60 min of irradiation while that of Rhodamine 6G used as a reference decreased by ca. 45% (Figure d). The good colloidal stability associated with the high photostability of Mn:AIGZS QDs demonstrate their potential as photoluminescent probes for real-time monitoring.

Figure 9

(a) ζ-Potential and (b) hydrodynamic diameter of Mn(2.5):AIGZS@GTMA QDs, (c) evolution of the PL emission wavelength and of the PL intensity with the pH from 3 to 13, and (d) photostability of Mn(2.5):AIGZS QDs and Rhodamine 6G under the continuous irradiation of a Hg–Xe lamp (intensity of 50 mW/cm2).

Mn:AIGZS QDs as MRI Probes

Mn:AIGZS QDs were in vitro investigated as contrast agents (CAs) for MRI. Surprisingly, similar relaxation rates were obtained for AIGZS QDs doped with 2.5, 5, and 10% Mn. ICP-OES measurements show that the amount of Mn2+ ions doped in AIGZS decreases for the QDs doped with 5 and 10% Mn2+ (from 4.67 and 5.01% Mn for the native OAm capped to 2.76 and 2.97% after the GTMA ligand exchange, respectively). These results suggest that ca. 2% of the Mn2+ ions are present at the surface of AIGZS doped with 5 and 10% Mn2+ and that these ions are driven by the OAm ligand during the ligand exchange. The loading in Mn2+ only slightly decreased for Mn(1) and Mn(2.5):AIGZS QDs and was found to be 0.77 and 1.86 atom % after the capping exchange.

The performance of GTMA-coated Mn(1) and Mn(2.5):AIGZS QDs as MRI contrast agents was investigated at room temperature on 1.4 and 2.34 T NMR analyzers. The variations of the longitudinal (R1) and transverse (R2) relaxation rates vs the molar concentration in Mn are shown in Figure a,b. The concentration-independent relaxivities (r1 and r2) were determined from the slopes of the concentration-dependent R1 and R2, and the values are r1 = 0.28 mM–1 s–1 and r2 = 0.75 mM–1 s–1 at 1.4 T and r1 = 0.15 mM–1 s–1 and r2 = 0.57 mM–1 s–1 at 2.34 T, respectively.

Figure 10

Longitudinal (R1) and transverse relaxation rates (R2) of Mn:AIGZS QDs as a function of Mn2+ concentration in aqueous solution (a) at 1.4 T, 60 MHz, and 298 K and (b) at 2.34 T, 100 MHz, and 298 K, and (c) T1-weighted and T2-weighted images of undoped Mn(1) and Mn(2.5):AIGZS QDs at 2.34 T.

To further evaluate the potential of Mn:AIGZS QDs in aqueous solution as contrast agents in MRI, images are recorded at 2.34 T on the same samples (Figure c). With the increase of the Mn concentration, the T1-weighted signal becomes much brighter, while the T2-weighted one becomes darker, thus confirming that Mn:AIGZS QDs can be used for T1- and T2-weighted MRI. Finally, the r1 values determined for Mn:AIGZS QDs are lower than those of Gd complexes (3.1 mM–1 s–1) and of Mn-doped Cu–In–Zn–S QDs.[34,52] This likely originates from the deeper localization of the Mn2+ dopant in AIGZS nanocrystals as a doping in the shell is usually used to increase the relaxivity.

Conclusions

In summary, Mn:AIGZS QDs with an average diameter of ca. 2 nm were prepared via a one-pot synthesis involving the thermal decomposition of AgInGa2–Zn2(S2CN(C2H5)2)4 precursors and Mn(St)2 in oleylamine. The obtained QDs show the Mn2+ 4T1 → 6A1 related PL emission, characterized by high PL QYs (up to 41.3%) and long PL lifetimes (up to 887.9 μs for AIGZS QDs doped with 2.5% Mn). The optical properties were maintained after the oil-soluble Mn:AIGZS QDs were transferred into aqueous solution via ligand exchange, indicating that the dots might be used as photoluminescent probes for bioimaging. Mn:AIGZS QDs can also be used as a contrast agent in MR imaging. Longitudinal relaxivity value (r1) and transverse relaxivity value (r2) of 0.15 and 0.57 mM–1 s–1, respectively, were determined using a 2.34 T MR imager. These results clearly indicate that Mn:AIGZS QDs have a high potential for dual-modal photoluminescent/MR imaging.

Experimental Section

Materials

Silver nitrate AgNO3 (ReagentPlus, >99%, Sigma), gallium nitrate hydrate Ga(NO3)3·xH2O (>99.9%, Sigma), indium nitrate hydrate In(NO3)3·xH2O (>99.9%, Alfa Aesar), zinc diethyldithiocarbamate Zn(DDTC)2 (97%, Sigma), manganese chloride tetrahydrate MnCl2·4H2O (99.99%, Sigma), oleylamine (OAm, 80–90%, Acros Organics), stearic acid (>98.5%, Sigma), glutathione (GSH, 99%, Sigma), and tetramethylammonium hydroxide pentahydrate (TMAH, 97%, Sigma) were used as received without any purification.

Preparation of AgInGa2–Zn2(S2CN(C2H5)2)4 Precursors

The single precursor used for the synthesis of AIGZS QDs was prepared according to our previous report.[37] Briefly, AgNO3 (0.25 mmol), In(NO3)3 (0.875 mmol), Ga(NO3)3 (0.875 mmol), and Zn(DDTC)2 (2 mmol) (a Ag/In/Ga/Zn molar ratio of 0.25:0.875:0.875:2) were mixed in 10 mL of a 1:1 water–MeOH mixture. The solution was stirred for 2 h at room temperature to form AgInGa2–Zn2(S2CN(C2H5)2)4 precursors. The solid obtained was collected by centrifugation (1700g for 15 min), dried under vacuum, and stored at 4 °C until use.

Synthesis of Mn(+2) Stearate (MnSt2)

Under an Ar flow, stearic acid (10 mmol) in 15 mL of MeOH was heated to 50–60 °C until getting a clear solution. After cooling to room temperature, TMAH (10 mmol) dissolved in 20 mL of MeOH was slowly added, and the mixture was stirred for 15 min. Next, MnCl2·4H2O (5 mmol) dissolved in 10 mL of MeOH was added dropwise, and the mixture was stirred for 2 h at room temperature. The white MnSt2 precipitate was recovered by centrifugation (1700g for 15 min), washed several times with MeOH, dried under vacuum, and stored at 4 °C under N2 before use.

Preparation of Mn-Doped Ag–In–Ga–Zn–S (AIGZS) QDs

Mn-doped AIGZS QDs were synthesized by thermolysis of AgInGa2–Zn2(S2CN(C2H5)2)4 and MnSt2 using OAm as a solvent. First, the precursor (100 mg) and the appropriate amount of MnSt2 were dispersed in OAm (6 mL) and the mixture was stirred under an Ar flow until getting a clear solution. Then, the mixture was heated at 220 °C and maintained at this temperature for 10 min for completion of nanoparticle growth. After cooling, the large-sized crystals were removed by centrifugation (1700g for 10 min). Mn-doped AIGZS QDs were recovered by precipitation using MeOH followed by centrifugation (2500g for 10 min) and dried under vacuum. Next, the dried QDs were redispersed in OAm (6 mL), and the solution was heated at 220 °C for 30 min. After completion of the reaction, Mn:AIGZS QDs were precipitated by adding an excess of MeOH, recovered by centrifugation (2500g for 10 min), and purified by washing with MeOH. The concentrations of Mn2+ ions in AIGZS QDs used through the manuscript are based on the initial concentrations used during the synthesis.

Transfer of Mn:AIGZS QDs into Water Using Glutathione Tetramethylammonium (GTMA)

The ligand exchange with GTMA was conducted using a previously described method.[50] Briefly, GTMA was prepared by reacting GSH (30 mg) and TMAH (54 mg) in 1 mL of MeOH. Separately, 5 mg of the oleophilic Mn:AIGZS@OAm QDs was dispersed in 1 mL of toluene before adding 5 mL of chloroform. Next, 0.5 mL of the GTMA solution was added dropwise to the QD solution and the mixture was vigorously stirred for 2 h at room temperature. Then, water (4.5 mL) was added and the biphasic mixture was slightly shaken to transfer QDs into the aqueous phase before being left for 1 h to reach equilibrium. Finally, the top water layer containing GTMA-capped Mn:AIGZS QDs was collected, precipitated using an excess of MeOH, and centrifuged (2500g for 15 min). Mn:AIGZS QDs were stored at 4 °C for further use.

Morphological, Structural, and Optical Characterizations

The morphology and the microstructure of AIGZS QDs were investigated by transmission electron microscopy (TEM, JEOL-ARM200 Cold FEG microscope operating at 200 keV, equipped with double spherical aberration correctors and fitted with a JEOL SDD CENTURIO EDS system and a Gatan GIF quantum). Selected area electron diffraction (SAED) experiments and EDS analysis were performed on a JEOL ARM 200 Cold FEB transmission electron microscope operating at 200 keV, equipped with double spherical aberration correctors and fitted with a JEOL SD CENTURIO EDS system. The crystal structure of AIGZS QDs was examined by X-ray diffraction (XRD) using a Panalytical X’Pert Pro MPD diffractometer with Cu Kα radiation (λ = 0.15418 nm). X-ray photoelectron spectroscopy (XPS) measurements were carried out with a Gammadata Scienta SES 200-2 spectrometer. Inductively coupled plasma-optical emission spectrometer (ICP-OES) measurements were conducted on a Varian 720-ES equipment. Hydrodynamic QD sizes and ζ-potential measurements were determined by dynamic light scattering (DLS) on a Zetasizer Nano ZS at 25 °C in water (green laser beam 532 nm) with a disposable capillary cell (DTS1070) (Malvern Panalytical, U.K.).

Fourier transform infrared (FTIR) spectra were collected on a Brucker α spectrometer. Optical absorption and PL emission spectra were recorded on a UV–visible absorption spectrometer (Thermo Scientific Evolution 220) and spectrofluorimeter (Horiba Fluoromax-4 Jobin Yvon), respectively. PL spectra were spectrally corrected and PL QYs were determined relative to Rhodamine 6G in ethanol (PL QY = 94%). For the time-resolved photoluminescence (TR-PL) experiments, the QDs were pumped by the 355 nm line of a frequency-tripled YAG (yttrium aluminum garnet):Nd laser. The laser pulse frequency, energy, and duration were typically equal to 10 Hz, 50 μJ, and 10 ns, respectively. The PL signal was analyzed by a monochromator equipped with 600 grooves/mm grating and by a photomultiplier tube cooled at 190 K. The rise time of the detector is equal to around 3 ns.

Magnetic Measurements

X-band EPR spectra were recorded at 20 K in nonsaturating conditions on a Bruker ELEXSYS 500 spectrometer, equipped with an Oxford instrument continuous-flow liquid-helium cryostat and a temperature control system. Typical conditions were as follows: 5G amplitude modulation, 9.409 GHz, microwave power of 0.63 mW.

Magnetic properties of all synthesized samples were characterized by vibrating sample magnetometry (VSM) with an applied magnetic field (H) of up to ± 9 T. The magnetic moment values are reported in electromagnetic units (emu) per mass unit of the sample.

Relaxivity Measurements and MR Images

Relaxation rates R1 and R2 (inverse of relaxation times T1 and T2, respectively) at 60 MHz (1H resonance frequency, corresponding to a magnetic field of 1.4 T) were obtained using a Bruker Minispec MQ60 equipment. Longitudinal (T1) and transverse (T2) relaxation times of water protons were measured at 25 °C with the following parameters: T1 measurements—8 accumulations, repetition time 10 s, inversion recovery with 16 evolution times between 0.001 s and 4 × T1; T2 measurements—16 accumulations, repetition time 10 s, Carr–Purcell–Meiboom–Gill experiment with 2000 points, and an interpulse delay of 2 × 0.6 ms.

Relaxation rates at 100 MHz were measured using the same methods and parameters as the ones used at 60 MHz but on a Bruker Biospec Avance 24/40, operating at a magnetic field of 2.34 T (proton resonance frequency of 100 MHz). MR images were acquired on the same spectrometer. The 100 mg/mL solutions of 0, 1, and 2.5% Mn:AIGZS QDs were put in 4 mL vials and imaged with the following parameters: field of view: 6 cm; matrix: 256 × 256 pixels; and slice thickness: 2 mm. Two sets of spin-echo images were acquired: the first one with a 200 ms repetition time and 14 ms echo time to achieve a good T1 contrast and the T2-weighted images were recorded with a 5000 ms repetition time and 120 ms echo time.