Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Room-temperature formation of CdS magic-size clusters in aqueous solutions assisted by primary amines


Aqueous-phase approaches to semiconductor CdS magic-size clusters (MSCs) and the formation pathway have remained relatively unexplored. Here, we report the demonstration of an aqueous-phase, room-temperature approach to CdS MSCs, together with an exploration of their evolution pathway. The resulting CdS MSCs display a sharp optical absorption peak at about 360 nm and are labeled MSC-360. With CdCl2 and thiourea as the respective Cd and S sources, and 3-mercarpotopropionic acid as the ligand, CdS MSC-360 develops in a mixture of a primary amine and water. We argue that the primary amine facilitates room-temperature decomposition of thiourea when CdCl2 is present, and the formation pathway of MSCs is similar to that in organic-phase approaches. Our findings show there is a viable avenue to room-temperature aqueous-phase formation of CdS MSCs. Providing explanations of the procedure developed including the formation of large aggregates, the present study represents an important advance towards a mechanistic understanding of nanocrystal synthesis.


Crystallization from liquid phases is acknowledged to be central to materials science and vital in applications such as pharmaceutical and chemical manufacturing. In this regard, two models have been developed for nucleation, which are one-step or multistep based1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28. The one-step model, described by classical nucleation theory (CNT), involves the concept of supersaturation or supercooling1,2,3. The multistep model, designated by nonclassical nucleation theory, accounts for the presence of intermediates in the pre-nucleation stage4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28. Such intermediates have been illustrated in several materials systems including calcium-based inorganics4,5,6,7,8, organics9,10,11, polymers12,13,14, metals15,16,17, and semiconductor quantum dots (QDs)18,19,20,21,22,23,24,25,26,27,28.

For organic-phase approaches to semiconductor metal (M) chalcogenide (E) QDs, a two-pathway model has been proposed for the pre-nucleation stage18,19,20,21,22,23,24,25,26,27,28. One pathway involves the formation of monomers and fragments, which result in QDs as per the LaMer model of the CNT. The other pathway deals with the self-assembly of M and E precursors which occurs first followed by the formation of particular precursor compounds (PCs) for magic-size clusters (MSCs). The self-assembly pathway proposed for the MSC formation is consistent with the high-concentration approach to CdS MSCs29. These two individual pathways are interconnected; when QDs grow in size, fragmentation of PCs into monomers and fragments happens. Usually, MSCs exhibit relatively sharp optical absorption peaking at a persistent position compared to their corresponding QDs, while the PCs display no characteristic optical absorption. The PC is conceptually the same as the oligomer (CaCO3)n (n = 3–11) and Posner molecule (Ca3(PO4)2)3 reported for the pre-nucleation stage of calcium-based inorganics4,7,28. The PC to MSC transformation occurs via intramolecular reorganization, which follows first-order reaction kinetics19,23,26. A PC and its corresponding MSC appear to be one polymorphous pair, with similar semiconductor cores but different organic ligands. The PC to MSC transformation has been demonstrated only in nonaqueous-phase approaches for semiconductors, including CdS18,19,20, CdSe21,25, CdTe22,23,24,25, CdTeSe25, ZnS26, and ZnSe27.

A few examples of aqueous-phase approaches for the synthesis of colloidal semiconductor QDs and MSCs have been reported, while these studies have been largely empirical30,31,32,33,34,35,36,37,38,39,40,41. The complexity of aqueous environments including ionic interactions has been claimed to inhibit understanding the formation pathways of QDs and MSCs30,37. Supplementary Tables 1 and 2 summarize the published experimental results regarding aqueous-phase syntheses of CdE MSCs and QDs, respectively. Until the present work only CdSe MSCs, which exhibit sharp optical absorption peaking at about 420 nm with full width at half maximum of 20 nm, have been synthesized38,39,40,41. The approach to the evolution of the CdSe MSCs requires long reaction times (such as 7 days), and the Se source Na2SeSO3 has to be freshly prepared due to its limited stability. The aqueous-phase synthesis of CdS MSCs has not been demonstrated, and whether the formation pathway of aqueous-phase MSCs is similar to that of nonaqueous phase ones is not known18,19,20,21,22,23,24,25,26,27,28.

Here, we present the aqueous-phase approach to CdS MSCs and describe our understanding of the procedure developed and of the formation pathway. With CdCl2 and thiourea (TU, S = C(NH2)2) as the respective Cd and S sources with the feed molar ratio of 2 to 1, and 3-mercarpotopropionic acid (MPA, HS–CH2CH2–COOH) as the ligand, CdS MSCs in a single-ensemble form evolve in a mixture of a primary amine and water. The resulting MSCs exhibit sharp optical absorption peaking at about 360 nm, and are denoted as MSC-360 (in reference to their absorption peak position in wavelength). Importantly, the presence of a primary amine, such as butylamine (BTA, CH3–(CH2)3–NH2), is a prerequisite for the evolution of CdS MSC-360 at room temperature. BTA assists room-temperature decomposition of TU upon the presence of CdCl2. And the quantity of CdCl2 has a very significant influence; when the Cd concentration is above 4.0 mM, the formation of MSC-360 is suppressed due to the presence of large aggregates (above 100 nm). The higher the concentration is, the larger the aggregates are. We demonstrate that for an aqueous solution of CdCl2 and MPA in a basic environment (pH = ~12), aggregation occurs when the Cd concentration is above 4.0 mM. The resulting aggregates have relatively hydrophobic cores, as indicated by fluorescence spectroscopy of pyrene. Nuclear magnetic resonance (NMR) spectroscopy provides further evidence for the formation and aggregation of Cd−MPA complexes. Moreover, we show that purified CdS MSC-360 disappears gradually in deionized water, but reappears after a primary amine is added. This behavior indicates that a reversible PC  MSC transformation takes place, which is similar to that in the nonaqueous approaches18,19,20,21,22,23,24,25,26,27,28. The present findings introduce a room-temperature aqueous-phase approach to the production of single-ensemble CdS MSC-360 without the co-production of QDs, clarifying that the PC to MSC transformation is applicable for the aqueous-phase formation of semiconductor MSCs. The present study contributes to the advances of cluster science and the science of crystallization.


Aqueous-phase evolution of CdS MSC-360

Figure 1 presents the evolution of optical absorption spectra collected from one mixture in two solvents. The mixture contains CdCl2 (2.0 mM), MPA (8.0 mM), KOH (20.0 mM), and TU (1.0 mM). One solvent is a mixture of BTA (1.5 mL) and water (H2O, 1.5 mL) (Fig. 1a, b). The other solvent is H2O (3.0 mL) (Fig. 1c, d). For the BTA-water solution, the mixture is first placed in 1.5 mL of water followed by the addition of 1.5 mL of BTA. See the “Methods” section for details regarding the preparation of the two solutions. The absorption spectra are collected after elapsed times of 0, 0.5, 1, 3, 6, and 12 h (Fig. 1a, c); at each time point, an aliquot (50 μL) is extracted and diluted in deionized water (3.0 mL) and absorption measurements are performed (Fig. 1b, d).

Fig. 1: Absorption spectra of one mixture of CdCl2, MPA, KOH, and TU in two solvents.
figure 1

They are a mixture of BTA and H2O with equal volumes (a, b), and H2O (c, d). The spectra are collected at the various periods indicated (a, c); right after each measurement, an aliquot (50 μL) is diluted in 3.0 mL of water (b, d). The spectra are offset to have similar absorbance at 440 (a, c) or 400 nm (b, d). The dashed arrows (a, b) and line (d) signify the positions indicated. Evidently, the presence of BTA is a prerequisite for the room-temperature evolution of CdS MSC-360.

In the mixture of BTA and water (Fig. 1a), an absorption feature gradually develops to peak at ~356 nm, which indicates the presence of CdS MSC-360. At the beginning (0 h), the spectrum is quite featureless. At 3 h, a peak at 347 nm can be identified clearly. Afterwards, this peak increases in intensity considerately and red shifts slightly to 356 nm at the 12 h point. Simultaneously, the intensity of the absorption around 228 nm decreases (Fig. 1b). Supplementary Fig. 1 suggests such short wavelength absorption is due to CdCl2, MPA, and TU together. Accordingly, the overall conversion yield is estimated to be about 60% after a period of 6 h (as illustrated by Supplementary Fig. 2). In water (Fig. 1c), the absorption spectra are featureless, which suggests that little reaction takes place. Also, the absorption at 228 nm remains constant (Fig. 1d), in agreement with that there is no consumption of the Cd and S sources.

Therefore, the evolution of CdS MSC-360 at room temperature occurs only when the mixture of CdCl2, MPA, KOH, and TU is in the BTA and water environment (Fig. 1a), in which case we observe the consumption of the starting materials (Fig. 1b). When this mixture is placed in water without BTA, the starting materials are not consumed and the evolution of CdS MSC-360 does not take place (Fig. 1c, d). Figure 1a, c focuses on the possible development of MSCs, while Fig. 1b, d illustrates the likely consumption of the reactants. The synthesis conditions have been carefully optimized, such as the Cd to S feed molar ratio (Supplementary Fig. 3 with a constant Cd concentration of 2.0 mM). Clearly, the characteristic optical absorption peak of MSC-360 broadens gradually upon increasing the TU concentration. In this regard we use a 2CdCl2 to 1TU feed molar ratio to synthesize CdS MSC-360.

Transmission electron microscopy (TEM), powder X-ray diffraction (XRD), and thermogravimetric analysis (TGA) have been employed (Supplementary Fig. 4). Although the conventional characterization tools have some shortcomings with regard to providing precise structural and compositional information of colloidal semiconductor small-size QDs and MSCs20,25,26,27,28, Supplementary Fig. 4 suggests that the aqueous-phase CdS MSCs are spherical with a diameter smaller than 3 nm, and have a similar structure as that of organic-phase CdS MSC-36142, with the ligand to inorganic core weight ratio of 20 to 80 (Supplementary Note 1). The Fourier transform infrared (FT-IR) spectrum of the purified CdS MSC-360 sample (Supplementary Fig. 5) indicates that the purified MSCs are passivated only by MPA molecules which act as surface ligands. Therefore, there is no evidence that these MSCs have any of the primary amine, BTA, associated with them.

Effect of amounts and nature of primary amines

For the formation of MSC-360 in the mixture of CdCl2, MPA, KOH, and TU, Fig. 1 suggests that BTA plays an important role. In Fig. 2, we present the effects of the quantity of BTA (Fig. 2a) and of different primary amines (Fig. 2b). The absorption spectra are collected from the same mixture but in four solvents containing four different amounts of BTA (Fig. 2a) and in three solvents with three different primary amines (Fig. 2b). The mixture consists of CdCl2 (2.0 mM), MPA (8.0 mM), KOH (20.0 mM), and TU (1.0 mM); after the mixture is placed into water, a primary amine is added to result in a final volume of 3.0 mL. In Fig. 2a, the added BTA is 0.5 (gray trace), 1.0 (green trace), 1.5 (blue trace), and 2.0 mL (magenta trace). In Fig. 2b, 1.5 mL of a primary amine of BTA (blue trace), propylamine (PrA, CH3–(CH2)2–NH2, lighter blue trace), and ethylamine (ETA, CH3–CH2–NH2, lightest blue trace) is added. Supplementary Fig. 6 presents their chemical structures and 3D models. All the spectra are collected 24 h after the preparation of the solutions at room temperature. The temporal evolution of the absorption properties of these solutions are shown in Supplementary Figs. 7 (BTA), 8 (PrA), and 9 (ETA).

Fig. 2: Absorption spectra (offset) of the same mixture in different solvents.
figure 2

a The four solvents are 3.0 mL of BTA and water mixtures, with 0.5 (gray trace), 1.0 (green trace), 1.5 (blue trace), and 2.0 (magenta trace) mL BTA. b The three solvents are the mixtures of 1.5 mL primary amines (BTA (blue trace), PrA (lighter blue trace), and ETA (lightest blue trace)) and 1.5 mL water. All the spectra are collected from the solutions kept in cuvettes at room temperature for 24 h. The dashed lines signify the positions indicated. Evidently, increasing the amount of BTA, or using a primary amine with a shorter alkyl chain, facilitates the production of MSC-360.

In Fig. 2a, the absorption spectrum of the solution having 2.0 mL BTA displays a sharp optical absorption peaking at about 354 nm, while that of the solution with 0.5 mL BTA exhibits a relatively broad absorption. Supplementary Fig. 7 demonstrates that CdS MSCs with a sharper absorption are produced at 24 h in a solvent with more BTA. This seems to be the case with PrA as well (Supplementary Fig. 8). When 1.5 mL of ETA is added instead of BTA or PrA, more CdS MSC-360 is present at 24 h (Fig. 2b). Solutions with amine volumes up to 1.5 mL are relatively transparent at all times, and MSC-360 appears gradually (Supplementary Figs. 7–9). Solutions with 2.0 mL of BTA or ETA appear milky at the start and become clear after around 12 or 24 h, respectively, and the evolution of MSC-360 is observed (Supplementary Figs. 7d and 9d). Because of the precipitation (with 2.0 mL of amine) and the evaporation of PrA (boiling point 48 °C) and ETA (boiling point 17 °C), the mixture of BTA (1.5 mL), and water (1.5 mL) is used in this study to synthesize CdS MSC-360.

It is noteworthy that BTA has been claimed to assist the decomposition of a TU derivative in N,N-dimethylformamide (DMF) during the nucleation and growth of PbS nanocrystals43. In the absence of CdCl2, Supplementary Fig. 10 suggests that TU remains stable at room temperature in a mixture of BTA (1.5 mL) and water (1.5 mL). In this regard the room-temperature decomposition of TU requires both BTA and the Cd precursor. Moreover, only primary amines are found to facilitate the room-temperature evolution of CdS MSC-360. For example, when a secondary amine diethylamine is used (Supplementary Fig. 11), no CdS MSC-360 is observed.

CdCl2 concentration effect upon aggregation

To intensify the production of CdS MSC-360, we increase the concentration of the solution, the result of which is shown in Fig. 1a. Figure 3 presents the absorption spectra of four solutions with four different concentrations after 24 h of preparation (a), and a summary of time-dependent absorbance of CdS MSC-360 (at its lowest energy transition peak position) in these solutions for times up to 24 h (b). Mixtures with the same feed molar ratio as that used in Fig. 1a are placed in the same solvent of BTA (1.5 mL) and water (1.5 mL), but with resulting CdCl2 concentrations of 2.0 (magenta trace (a) and circular symbols (b)), 4.0 (blue trace (a) and square symbols (b)), 6.0 (green trace (a) and triangular symbols (b)), and 10.0 mM (gray trace (a) and diamond symbols (b)). The solutions are kept in cuvettes for the absorption measurements after different time durations.

Fig. 3: Absorption properties of four solutions with four CdCl2 concentrations.
figure 3

For the four solutions, the absorption spectra after 24 h of their preparation are shown in a and time-dependent absorbance of MSC-360 is summarized in b. In the solvent of BTA (1.5 mL) and water (1.5 mL), the four mixtures have the feed molar ratio of 2CdCl2-8MPA-20KOH-1TU with the CdCl2 concentrations of 2.0 (magenta trace for a and circular symbols for b), 4.0 (blue trace for a and square symbols for b), 6.0 (green trace for a and triangular symbols for b), and 10.0 mM (gray trace for a and diamond symbols for b). The spectra (a) are offset to have similar absorbance at 440 nm; the dashed line signifies the position of 356 nm. The absorbance of MSC-360 (b) is subtracted by the value at the corresponding wavelength evolved at 1 min; the solid lines in b are for visual guide. Obviously, the production of CdS MSC-360 is suppressed upon increasing the CdCl2 concentrations.

The solution with 2.0 mM CdCl2 effectively produces CdS MSC-360, with the absorbance of CdS MSC-360 increasing gradually to about 0.42 after 24 h. When the concentration of CdCl2 is increased to 4.0 mM, the absorbance of MSC-360 at 24 h is reduced to 0.31. With 6.0 mM of CdCl2, the strength of MSC-360 at 24 h is further reduced to 0.25. When the concentration of CdCl2 is increased to 10.0 mM, an evolution of MSC-360 does not occur. Accordingly, an increase in the solution concentration suppresses the production of CdS MSC-360.

For the two solutions of 2CdCl2-8MPA-20KOH-1TU in the BTA (1.5 mL) and water (1.5 mL) solvent with the CdCl2 concentrations of 5.0 and 10.0 mM, dynamic light scattering (DLS) (Supplementary Fig. 12) indicates large aggregates are present with hydrodynamic diameters (Dh) of 272 ± 4 and 379 ± 3 nm, respectively. The higher the solution concentration is, the larger the aggregates are. We would like to point out that these aggregates are much larger than those which are resulted from the self-assembly of Cd and S precursors in the organic-phase reactions18,19,20,28. The formation of such large aggregates has also been validated by TEM (Supplementary Fig. 13).

To investigate why the aggregation occurs, we use the fluorescence spectroscopy of pyrene. This molecule has been used as a solvatochromic probe for aqueous solutions to detect the onset of self-assembly and/or aggregation, which can result in moieties including micelles and nanofibers44,45,46,47,48. Pyrene has high affinity for nonpolar environments, such that when pyrene moves into a relatively more hydrophobic environment, the fluorescence intensity ratio of its third (I3) to first (I1) peaks increases. Supplementary Fig. 14 presents the I3/I1 ratios of pyrene for two types of mixtures, which are 2CdCl2-8MPA-20KOH (a) and 2CdCl2-8MPA-20KOH-1TU (b). The mixtures are placed in 3.0 mL of water and in the BTA (1.5 mL) plus water (1.5 mL) solvent, with different CdCl2 concentrations ranging from 0.5 to 80.0 mM. When the CdCl2 concentration is higher than 4.0 mM, the I3/I1 ratio increases significantly for the two types of mixtures without or with TU. From these observations, the critical aggregation concentration (CAC) of CdCl2 can be estimated to be ~4.0 mM. The term “Cd–MPA complex” was used for the aqueous-phase approach to CdTe QDs with a reaction of CdCl2, MPA, NaOH, and Te precursor in water34,36.

Accordingly, we hypothesize that the “Cd–MPA complex” forms in our 2CdCl2-8MPA-20KOH solutions and aggregates when the CdCl2 concentration is higher than the CAC. In our 2CdCl2-8MPA-20KOH-1TU solutions, when the CdCl2 concentration is lower than CAC, both the Cd–MPA complex and TU are present in the BTA-containing aqueous environment, and CdS MSC-360 evolves readily (as presented in Fig. 1a). When the CdCl2 concentration is higher than the CAC, it is the Cd–MPA complex that dominates the aggregation observed. The resulting aggregates effectively separate a majority of the Cd–MPA complex from the BTA molecule which stays in the solution phase (outside of the aggregates). Consequently, the production of CdS MSC-360 is suppressed almost completely when the CdCl2 concentration is as high as 10.0 mM (Fig. 3 gray trace and diamond symbols). Thus, a relatively low CdCl2 concentration of 2.0 mM is used in this study to synthesize CdS MSC-360 with a constant Cd to MPA feed molar ratio of 1 to 4. In the present work we also discovered that when the ratio was 1 to 1, precipitation took place, and when the ratio was 1 to 2, the CAC became larger at about 20.0 mM.


For the Cd–MPA complex that has been suggested to be present in the basic environment34,36, there has been no information available regarding their formation and aggregation. To have a fundamental understanding of such processes, we use NMR spectroscopy to study five MPA-containing solutions in D2O with an MPA concentration of 40.0 mM. Figure 4 shows the 1H NMR spectra of 8MPA (1), 2CdCl2-8MPA (2), 8MPA-8KOH (3), 8MPA-20KOH (4), and 2CdCl2-8MPA-20KOH (5). For Solutions 2 and 5, the CdCl2 concentration is 10.0 mM. Supplementary Fig. 15 displays the corresponding 13C NMR spectra of these five solutions. All the spectra are collected at room temperature. For the MPA molecule, HOOC–CH2–CH2–SH, the two methylene groups are referred to a and b, which are next to the carboxyl and thiol groups, respectively.

Fig. 4: 1H NMR spectra of five MPA-containing solutions in D2O.
figure 4

The MPA concentration is 40.0 mM for the five solutions, which consist of 8MPA (1), 2CdCl2-8MPA (2), 8MPA-8KOH (3), 8MPA-20KOH (4), and 2CdCl2-8MPA-20KOH (5). The feed molar ratio indicated follows that of the mixture studied in Fig. 1, while the CdCl2 concentration in Solutions 2 and 5 is 10.0 mM. For the MPA molecule, the two protons of the two methylene groups are labeled as a and b. Evidently, the formation of the Cd–MPA complex involves the deprotonation of the thiol group. Compared to trace 4, the chemical shift and broadening of the two resonance signals in trace 5 is indicative of the formation and aggregation of the Cd–MPA complex, which is in agreement with the 13C NMR and pyrene results shown in Supplementary Figs. 15 and 14 for the corresponding solutions, respectively.

The 1H NMR spectrum of MPA alone shows a resonance multiplet signal at about 2.63 ppm (trace 1), which changes little when CdCl2 is present (trace 2). For Solution 3 with the same KOH concentration as that of MPA (trace 3), two triplet resonance signals are observed. One triplet is located at 2.62 ppm (Proton b), while the other is at 2.46 ppm (Proton a). For Solution 4 with a KOH concentration 2.5 times of that of MPA (trace 4), the two triplet resonance signals are observed at 2.54 (proton b) and 2.29 ppm (proton a). For Solution 5 (trace 5) upon the presence of CdCl2, two broad resonance signals appear at 2.66 (proton b) and 2.38 ppm (proton a). It is noteworthy that the former resonance signal is broader than the latter.

From Solutions 1, 3–4, the upfield shift and separation of the two triplets is related to the deprotonation of the thiol and carboxyl groups in the course of the increase of the pH. Taking account of the pKa values of carboxyl (4.3) and thiol (10.8) groups in MPA49, it is reasonable that the two functional groups remain protonated in Solution 1 (pH ~2.6). In Solution 3 (pH ~5.2), the carboxyl group is deprotonated, while the thiol group is not. In Solution 4 (pH ~11.9), the two groups are deprotonated. From Solutions 4–5, the resonance signals display downfield shifts, suggesting the formation of the Cd–MPA complex. For the formation of Cd-based complexes, such as Cd–TOP and Cd–TOPO complexes, downfield shifts have been observed by 31P NMR50,51. TOP and TOPO represent tri-n-octylphosphine and tri-n-octylphosphine oxide, respectively. It is noteworthy that after the addition of CdCl2 to Solution 3, precipitation happens. Furthermore, due to the protonation of both the carboxyl and thiol groups in Solution 1, there is no obvious change of the 1H NMR spectra in the presence of CdCl2 (trace 2).

The 1H NMR results demonstrate that CdCl2 and MPA have no coordination interaction in an acidic environment (pH ~3.0), while the Cd–MPA complex forms in a basic environment (pH ~12.0) when the carboxyl and thiol groups are deprotonated. It has been known that NMR signal linewidth is inversely related to the spin-spin relaxation time (T2). Signal broadening suggests reduced T2, which may result from slower motion of larger aggregates with enhanced masses22,50. Therefore, the broadening of the two resonance signals of Solution 5 is indicative of the aggregation of the Cd–MPA complex. While Proton b shows the broadening more with a weaker feature than Proton a does, it is reasonable to conclude that the Cd2+ ion coordinates with the deprotonated thiol group for the formation of the Cd–MPA complex. The formation and aggregation of the Cd–MPA complex is also supported by the corresponding 13C NMR of MPA in the five solutions (Supplementary Fig. 15). We add TU into Sample 5 (Supplementary Fig. 16); the 1H and 13C NMR measurements suggest further that the aggregation is driven by the Cd–MPA complex.

In organic-phase syntheses, the formation pathway of binary colloidal semiconductor ME MSCs comprises an intramolecular reorganization of their particular PC molecules18,19,20,21,22,23,24,25,26,27,28. Whether the aqueous-phase MSCs follow the same pathway of formation is an important question. Figure 5 presents the absorption spectra collected at various elapsed time durations up to 1 day after a CdS MSC-360 purified sample is dispersed in deionized water at room temperature (a), with the concentration of MSC-360 estimated about 1.0 g ∙ L−1. After 1 day, 1.5 mL of the dispersion is mixed with 1.5 mL of BTA, and optical absorption measurements are performed at different periods up to 6 h (b).

Fig. 5: Absorption spectra of an aqueous dispersion of purified CdS MSC-360 before and after mixing with BTA.
figure 5

All spectra are offset to have similar absorbance at 440 nm. Purified CdS MSC-360 is dispersed in deionized water with a concentration of about 1.0 g ∙ L−1, and the absorption measurements last for 1 day at various durations indicated a. CdS MSC-360 disappears gradually. Afterwards, 1.5 mL of the dispersion is mixed with BTA (1.5 mL) for additional measurements up to 6 h b. CdS MSC-360 evolves again. In water (a), CdS MSC-360 transforms to its PC; upon the presence of BTA (b), the PC to MSC-360 transformation occurs, with almost complete recovery of MSC-360 in the relatively short period of 6 h.

Figure 5 suggests that such intramolecular reorganization remains important in the formation of the aqueous-phase CdS MSC-360. In Fig. 5a, the dispersion initially exhibits an absorption peaking at 369 nm with an optical density of ~1.3. In the following 1 day, the amount of MSC-360 decreases gradually. After 1 day, the absorbance at 369 nm decreases to ~0.6. Further decrease is observed up to 4 days (Supplementary Fig. 17a). In Fig. 5b, the absorption peaking at ~353 nm keeps increasing, and becomes the major peak after 6 h. These results suggest that the recovery of MSC-360 appears almost complete upon the presence of BTA (with the dilution). When deionized water is used instead of BTA (Supplementary Fig. 17b), a recovery of CdS MSC-360 is not observed. With a smaller amount of BTA (Supplementary Fig. 18) and with different primary amines such as PrA and ETA (Supplementary Fig. 19), a recovery of MSC-360 can also be obtained. We would like to point out that since the peak position of the lowest energy absorption for the present MSCs varies significantly with solvent from 369 nm in deionized water to 353 nm in the BTA and water mixture, a label of MSC-360 would seem reasonable for these CdS MSCs.

In water (Fig. 5a), the disappearance of CdS MSC-360 indicates that an MSC to PC transformation has happened. Upon the presence of a primary amine (Fig. 5b), the process of the CdS MSC-360 recovery is quite similar to the PC to MSC transformation that occurs upon the presence of a primary amine such as BTA or octylamine (OTA) for two-step approaches which are carried out in nonaqueous environments, such as CdSe, and CdTe MSCs in toluene21,22,23,24, and ZnSe MSCs in cyclohexane27. The reversible transition process may be related to changes in surface passivation26. Close examination of panels a and b in Fig. 1 suggests that it is reasonable to infer that the consumption of the reactants (which resulted in the formation of the MSC PC) took place mainly in the first 6 h, while the PC to MSC-360 transformation (via intramolecular organization) continued for up to 12 h (shown). The relatively fast PC formation compared to the PC transformation to MSC-360 supports that the ~60% conversion yield was estimated for the PC formation (Supplementary Fig. 2). The different rates between the PC formation and PC transformation to MSCs can be found elsewhere25,52.

In conclusion, we have developed an effective aqueous-phase approach to CdS MSC-360 at room temperature and have explored the formation pathway. CdCl2 and TU are used as the respective Cd and S sources, with MPA as the ligand. For the synthesis, a mixture with the feed molar ratio of 2CdCl2-8MPA-20KOH-1TU is prepared; the mixture is then added to deionized water followed by the addition of BTA. The CdCl2 concentration in the resulting solution is 2.0 mM. The presence of BTA is a prerequisite for the room-temperature evolution of CdS MSC-360 (Fig. 1). While the amount of BTA matters, other primary amines such as PrA and ETA also promote the room-temperature evolution of MSC-360 (Fig. 2). The primary amine assists the room-temperature decomposition of TU in the presence of CdCl2 (Supplementary Fig. 10). Furthermore, the CdCl2 concentration matters (Fig. 3), due to the possible aggregation with the CAC of ~4.0 mM. Larger CdCl2 concentration give rise to larger aggregates. We argue that a Cd–MPA complex forms in a 2CdCl2-8MPA-20KOH solution with a basic environment (pH = ~12), in which the thiol and carboxyl groups of MPA are both deprotonated and a coordination interaction occurs between the former group and CdCl2 (Fig. 4). The comprehensive identification of the aggregation, which is dominated by the Cd–MPA complex with the CAC of ~4.0 mM, supports that the mechanism of the nucleation and growth of QDs is more complicated in aqueous-phase syntheses than in organic-phase ones30,37. We propose that the reactivity of TU is enhanced via the coordination with the Cd–MPA complex, followed by the BTA-assisted decomposition of the activated TU, which facilitates the room-temperature formation of Cd–S covalent bonds. Supplementary Fig. 20 depicts the possible reactions that could be involved in the BTA-assisted decomposition of TU, similar to those hypothesized for the BTA-assisted decomposition of diphenyl TU in DMF43. In deionized water, CdS MSC-360 (purified) disappears gradually; however, when BTA is introduced, the population of CdS MSC-360 recovers (Fig. 5). Thus, the formation pathway of aqueous-phase MSCs follows the PC to MSC transformation, similar to that in the organic-phase evolution18,19,20,21,22,23,24,25,26,27,28. The findings introduce a room-temperature aqueous-phase approach to forming semiconductor CdS MSC-360. As always, it is impossible to have a complete understanding of the formation mechanism from one study. The formation and aggregation of the Cd–MPA complex in basic aqueous environments have been quite unexpected and should be conceptually helpful even when other ligands are used. For the approach developed, we are actively exploring its applicability for the synthesis of other II–VI semiconductor MSCs, such as those of CdSe and CdTe. The present study brings deeper insights for the formation pathway of colloidal semiconductor MSCs, conveying comprehension into the pre-nucleation stage and contributing to the advance of the nanocrystal synthesis from an empirical art to science.



MPA (99%), BTA (99%), ETA aqueous solution (ETA, 70%), deuterium oxide (D2O, 99%), and pyrene (99%) were purchased from Aldrich. CdCl2·2.5H2O (99%) and TU (99%) were purchased from Chengdu Kelong Chemical. Ethanol (99%) was from Chengdu Haixin Chemical. Potassium hydroxide (KOH, 99%) and propylamine (PrA, 99%) were obtained from Tianjin Zhiyuan Chemical. All the chemicals are used without purification (except stated otherwise).

Preparation of Cd and S precursor stock solutions and CdS MSCs

A Cd precursor stock solution was prepared by dissolving CdCl2·2.5H2O (0.1827 g, 0.8 mmol) in deionized water (5.0 mL) with MPA (279 μL, 1.6 mmol), to which KOH solution (5 M, about 1.6 mL) was added to adjust the pH to about 11.7. Additional deionized water was added to make the total volume of 10.0 mL, with the pH changed little. A S precursor stock solution was obtained by dissolving TU (0.0304 g, 0.4 mmol) in deionized water (10.0 mL). For the room-temperature aqueous-phase synthesis of CdS MSCs (such as shown in Fig. 1a), 75 μL Cd precursor solution and 75 μL S precursor solution were mixed together; the resulting mixture was added into 1.35 mL deionized water, followed by the addition of 1.5 mL BTA. It is noteworthy that the aqueous solution has apparent pH values of 11.0 and 13.0 before and after the addition of BTA. The resulting solution consisting of CdCl2 (2.0 mM), TU (1.0 mM), and MPA (8.0 mM) was then allowed to react at room temperature for various durations. The colloidal reaction system remains stable up to 3 days.

Purification of CdS MSCs

The as-synthesized CdS MSC aqueous sample (3.0 mL) was precipitated upon the addition of ethanol (9.0–12.0 mL). The resulting turbid solution was then centrifuged on a Beckman Coulter Allegra 64R at room temperature with a speed of 8000 rpm for about 2 min. After centrifugation, the resulting precipitate was collected and evacuated for 1 h to remove the solvent residual.

Ultraviolet absorption spectroscopy

Absorption spectra were collected using a TECHCOMP UV 2310 II ultraviolet–visible (UV–vis) absorption spectrometer, with a 1-nm data collection interval. A cuvette with deionized water was used as a background sample.

Photoluminescence spectroscopy

Pyrene (0.3 μM) was used as a probe to investigate the aggregation. A solution of pyrene was prepared in water or the mixture of BTA and water with equal volumes, to which a mixture of 2CdCl2-8MPA-20KOH-1TU was added with various CdCl2 concentrations. The samples prepared in deionized water were kept at room temperature for 1 day before emission measurements. Those prepared in the mixture of BTA and water were stored for about 10 min only before measurements; longer storage would consume the precursors and form CdS MSCs. The photoluminescence spectra were recorded with an excitation wavelength of 335 nm and with both excitation and emission slit widths of 2.5 nm. For the pyrene spectra, the intensity of the third (I3, at about 380 nm) and the first peaks (I1, at about 370 nm) was extracted, and the I3/I1 ratio was then plotted vs the CdCl2 concentrations.

Dynamic light scattering (DLS)

DLS measurements were carried out on a Zetasizer Nano ZS90 instrument. The two 2CdCl2-8MPA-20KOH-1TU samples were placed in the solvent of BTA (1.5 mL) and water (1.5 mL). Each sample was kept for about 10 min at room temperature before measurement; longer storage would consume the precursors and form CdS MSCs.

1H and 13C NMR spectroscopy

The samples for both 1H and 13C NMR measurements were prepared in D2O. Each sample was kept for about 2 h at room temperature before measurement. The NMR spectra were recorded on Bruker Avance II spectrometers. The 1H NMR spectra were corrected with the characteristic resonance signals of the solvent D2O. The 1H and 13C NMR spectra were collected at 400 and 200 MHz, respectively.

Data availability

The authors declare that all relevant data supporting the findings of this study are available from the authors on request.


  1. 1.

    Bai, G., Gao, D., Liu, Z., Zhou, X. & Wang, J. Probing the critical nucleus size for ice formation with graphene oxide nanosheets. Nature 576, 437–441 (2019).

    ADS  CAS  PubMed  Google Scholar 

  2. 2.

    Thanh, N. T., Maclean, N. & Mahiddine, S. Mechanisms of nucleation and growth of nanoparticles in solution. Chem. Rev. 114, 7610–7630 (2014).

    CAS  PubMed  Google Scholar 

  3. 3.

    García-Rodríguez, R., Hendricks, M. P., Cossairt, B. M., Liu, H. & Owen, J. S. Conversion reactions of cadmium chalcogenide nanocrystal precursors. Chem. Mater. 25, 1233–1249 (2013).

    Google Scholar 

  4. 4.

    Liu, Z. et al. Crosslinking ionic oligomers as conformable precursors to calcium carbonate. Nature 574, 394–398 (2019).

    ADS  CAS  PubMed  Google Scholar 

  5. 5.

    Habraken, W. J. et al. Ion-association complexes unite classical and non-classical theories for the biomimetic nucleation of calcium phosphate. Nat. Commun. 4, 1507 (2013).

    ADS  PubMed  Google Scholar 

  6. 6.

    Demichelis, R., Raiteri, P., Gale, J. D., Quigley, D. & Gebauer, D. Stable prenucleation mineral clusters are liquid-like ionic polymers. Nat. Commun. 2, 590 (2011).

    ADS  PubMed  Google Scholar 

  7. 7.

    Dey, A. et al. The role of prenucleation clusters in surface-induced calcium phosphate crystallization. Nat. Mater. 9, 1010–1014 (2010).

    ADS  CAS  PubMed  Google Scholar 

  8. 8.

    Gebauer, D., Volkel, A. & Colfen, H. Stable prenucleation calcium carbonate clusters. Science 322, 1819–1822 (2008).

    ADS  CAS  PubMed  Google Scholar 

  9. 9.

    Chen, H. et al. Multistep nucleation and growth mechanisms of organic crystals from amorphous solid states. Nat. Commun. 10, 3872 (2019).

    ADS  PubMed  Google Scholar 

  10. 10.

    Wiedenbeck, E., Kovermann, M., Gebauer, D. & Colfen, H. Liquid metastable precursors of ibuprofen as aqueous nucleation intermediates. Angew. Chem. Int. Ed. 58, 19103–19109 (2019).

    CAS  Google Scholar 

  11. 11.

    Davey, R. J., Schroeder, S. L. & Horst, J. H. Nucleation of organic crystals—a molecular perspective. Angew. Chem. Int. Ed. 52, 2166–2179 (2013).

    CAS  Google Scholar 

  12. 12.

    Tang, X., Chen, W. & Li, L. The tough journey of polymer crystallization: battling with chain flexibility and connectivity. Macromolecules 52, 3575–3591 (2019).

    ADS  CAS  Google Scholar 

  13. 13.

    Ma, X. et al. Tuning crystallization pathways through sequence engineering of biomimetic polymers. Nat. Mater. 16, 767–774 (2017).

    ADS  CAS  PubMed  Google Scholar 

  14. 14.

    Sleutel, M. & Van Driessche, A. E. Role of clusters in nonclassical nuclation and growth of protein crystals. Proc. Natl Acad. Sci. 111, 546–553 (2014).

    Google Scholar 

  15. 15.

    Jin, B. et al. Revealing the cluster-cloud and its role in nanocrystallization. Adv. Mater. 31, e1808225 (2019).

    PubMed  Google Scholar 

  16. 16.

    Yang, J. et al. Amorphous-phase-mediated crystallization of Ni nanocrystals revealed by high-resolution liquid-phase electron microscopy. J. Am. Chem. Soc. 141, 763–768 (2019).

    CAS  PubMed  Google Scholar 

  17. 17.

    Loh, N. D. et al. Multistep nucleation of nanocrystals in aqueous solution. Nat. Chem. 9, 77–82 (2017).

    CAS  PubMed  Google Scholar 

  18. 18.

    Zhang, J. et al. Individual pathways in the formation of magic-size clusters and conventional quantum dots. J. Phys. Chem. Lett. 9, 3660–3666 (2018).

    CAS  PubMed  Google Scholar 

  19. 19.

    Zhu, T. et al. Two-step nucleation of CdS magic-size nanocluster MSC–311. Chem. Mater. 29, 5727–5735 (2017).

    CAS  Google Scholar 

  20. 20.

    Zhang, B. et al. Thermally-induced reversible structural isomerization in colloidal semiconductor CdS magic-size clusters. Nat. Commun. 9, 2499 (2018).

    ADS  PubMed  Google Scholar 

  21. 21.

    Zhu, D. et al. Interpreting the ultraviolet absorption in the spectrum of 415 nm-bandgap CdSe magic-size clusters. J. Phys. Chem. Lett. 9, 2818–2824 (2018).

    CAS  PubMed  Google Scholar 

  22. 22.

    Liu, M. et al. Probing intermediates of the induction period prior to nucleation and growth of semiconductor quantum dots. Nat. Commun. 8, 15467 (2017).

    ADS  CAS  PubMed  Google Scholar 

  23. 23.

    Luan, C. et al. Evolution of two types of CdTe magic-size clusters from a single induction period sample. J. Phys. Chem. Lett. 9, 5288–5295 (2018).

    CAS  PubMed  Google Scholar 

  24. 24.

    Luan, C. et al. Four types of CdTe magic-size clusters from one prenucleation stage sample at room temperature. J. Phys. Chem. Lett. 10, 4345–4353 (2019).

    CAS  PubMed  Google Scholar 

  25. 25.

    Gao, D. et al. Formation of colloidal alloy semiconductor CdTeSe magic-size clusters at room temperature. Nat. Commun. 10, 1674 (2019).

    ADS  PubMed  Google Scholar 

  26. 26.

    Liu, S. et al. Transformation of ZnS precursor compounds to magic-size clusters exhibiting optical absorption peaking at 269 nm. J. Phys. Chem. Lett. 11, 75–82 (2020).

    CAS  PubMed  Google Scholar 

  27. 27.

    Wang, L. et al. Precursor self-assembly identified as a general pathway for colloidal semiconductor magic-size clusters. Adv. Sci. 5, 1800632 (2018).

    Google Scholar 

  28. 28.

    Palencia, C., Yu, K. & Boldt, K. The future of colloidal semiconductor magic-size clusters. ACS Nano 14, 1227–1235 (2020).

    CAS  PubMed  Google Scholar 

  29. 29.

    Nevers, D. R. et al. Mesophase formation stabilizes high-purity magic-sized custers. J. Am. Chem. Soc. 140, 3652–3662 (2018).

    CAS  PubMed  Google Scholar 

  30. 30.

    Jing, L. et al. Aqueous based semiconductor nanocrystals. Chem. Rev. 116, 10623–10730 (2016).

    CAS  PubMed  Google Scholar 

  31. 31.

    Aboulaich, A. et al. One-pot noninjection route to CdS quantum dots via hydrothermal synthesis. ACS Appl. Mater. Interfaces 4, 2561–2569 (2012).

    CAS  PubMed  Google Scholar 

  32. 32.

    Vossmeyer, T. et al. CdS nanoclusters: synthesis, characterization, size dependent oscillator strength, temperature shift of the excitonic transition energy, and reversible absorbance shift. J. Phys. Chem. 98, 7665–7673 (1994).

    CAS  Google Scholar 

  33. 33.

    Fojtik, A., Weller, H., Koch, U. & Henglein, A. Photo-physics of extremely small CdS particles: Q-state CdS and magic agglomeration numbers. Ber. Bunsenges. Phys. Chem. 88, 969–977 (1984).

    CAS  Google Scholar 

  34. 34.

    Zhou, D. et al. Hydrazine-mediated construction of nanocrystal self-assembly materials. ACS Nano 8, 10569–10581 (2014).

    CAS  PubMed  Google Scholar 

  35. 35.

    Zhou, D. et al. Simple synthesis of highly luminescent water-soluble CdTe quantum dots with controllable surface functionality. Chem. Mater. 23, 4857–4862 (2011).

    CAS  Google Scholar 

  36. 36.

    Han, J. et al. Growth kinetics of aqueous CdTe nanocrystals in the presence of simple amines. J. Phys. Chem. C. 114, 6418–6425 (2010).

    CAS  Google Scholar 

  37. 37.

    Zhou, D. et al. Nucleation of aqueous semiconductor nanocrystals: a neglected factor for determining the photoluminescence. J. Phys. Chem. C. 114, 22487–22492 (2010).

    CAS  Google Scholar 

  38. 38.

    Kurihara, T., Noda, Y. & Takegoshi, K. Quantitative solid-state NMR study on ligand-surface interaction in cysteine-capped CdSe magic-sized clusters. J. Phys. Chem. Lett. 8, 2555–2559 (2017).

    CAS  PubMed  Google Scholar 

  39. 39.

    Baker, J. S., Nevins, J. S., Coughlin, K. M., Colón, L. A. & Watson, D. F. Influence of complex-formation equilibria on the temporal persistence of cysteinate-functionalized CdSe nanocrystals in water. Chem. Mater. 23, 3546–3555 (2011).

    CAS  Google Scholar 

  40. 40.

    Park, Y.-S. et al. Aqueous phase synthesized CdSe nanoparticles with well-defined numbers of constituent atoms. J. Phys. Chem. C. 114, 18834–18840 (2010).

    CAS  Google Scholar 

  41. 41.

    Park, Y.-S. et al. Size-selective growth and stabilization of small CdSe nanoparticles in aqueous solution. ACS Nano 4, 121–128 (2010).

    CAS  PubMed  Google Scholar 

  42. 42.

    Tang, J. et al. CdS magic-size clusters exhibiting one sharp ultraviolet absorption singlet peaking at 361 nm. Nano Res. 12, 1437–1444 (2019).

    CAS  Google Scholar 

  43. 43.

    Wang, Y. et al. Room-temperature direct synthesis of semi-conductive PbS nanocrystal inks for optoelectronic applications. Nat. Commun. 10, 5136 (2019).

    ADS  PubMed  Google Scholar 

  44. 44.

    Kalyanasundaram, K. & Thomas, J. K. Environmental effects on vibronic band intensities in pyrene monomer fluorescence and their application in studies of micellar systems. J. Am. Chem. Soc. 99, 2039–2044 (1977).

    CAS  Google Scholar 

  45. 45.

    Palos Pacheco, R. et al. Synthesis and characterization of four diastereomers of monorhamnolipids. J. Am. Chem. Soc. 139, 5125–5132 (2017).

    CAS  PubMed  Google Scholar 

  46. 46.

    Zhang, X. & Waymouth, R. M. 1,2-Dithiolane-derived dynamic, covalent materials: cooperative self-assembly and reversible cross-linking. J. Am. Chem. Soc. 139, 3822–3833 (2017).

    CAS  PubMed  Google Scholar 

  47. 47.

    Zhang, M., Fives, C., Waldron, K. C. & Zhu, X. X. Self-assembly of a bile acid dimer in aqueous solutions: from nanofibers to nematic hydrogels. Langmuir 33, 1084–1089 (2017).

    CAS  PubMed  Google Scholar 

  48. 48.

    Chen, Y. L., Luo, J. T. & Zhu, X. X. Fluorescence study of inclusion complexes between star-shaped cholic acid derivatives and polycyclic aromatic fluorescent probes and the size effects of host and guest molecules. J. Phys. Chem. B 112, 3402–3409 (2008).

    CAS  PubMed  Google Scholar 

  49. 49.

    Bhattacharyya, L., Rohrer, J. S. Applications of ion chromatography for pharmaceutical and biological products. (John Wiley & Sons, Inc., 2012).

  50. 50.

    Yu, K. et al. The formation mechanism of binary semiconductor nanomaterials: shared by single-source and dual-source precursor approaches. Angew. Chem. Int. Ed. 52, 11034–11039 (2013).

    CAS  Google Scholar 

  51. 51.

    Yu, K. et al. General low-temperature reaction pathway from precursors to monomers before nucleation of compound semiconductor nanocrystals. Nat. Commun. 7, 12223 (2016).

    ADS  CAS  PubMed  Google Scholar 

  52. 52.

    Zhang, H. et al. A room-temperature formation pathway for CdTeSe alloy magic-size clusters. Angew. Chem. Int. Ed. (2020).

Download references


Me.Z. is grateful to China Postdoctoral Science Foundation 2020T130441, Sichuan University Postdoctoral Research Fund 2019SCU12073 and 0020224153002, and the Fundamental Research Funds for the Central Universities. K.Y. thanks National Natural Science Foundation of China (NSFC) 21773162, the Fundamental Research Funds for the Central Universities, the Applied Basic Research Programs of Science and Technology Department of Sichuan Province 2020YJ0326, the State Key Laboratory of Polymer Materials Engineering of Sichuan University respectively for Grant No. sklpme2018-2-08, and the Open Project of Key State Laboratory for Supramolecular Structures and Materials of Jilin University for SKLSSM 201935 and 202035. H.F. and W.H. appreciates the National Major Scientific and Technological Special Project for “Significant New Drugs Development” (2019ZX09201005-005-001 and 2019ZX09201005-005-002). We thank Sichuan Univ of Analytical & Testing Center for TEM, XRD, TGA, and NMR.

Author information




Me.Z. and K.Y. designed and directed this study. W.W. and Me.Z. performed the synthesis and characterization, with NMR and TEM by C.Z. and S.W., respectively. W.W. and Me.Z. summarized and analyzed the experimental data. W.W., Me.Z., Mi.Z., N.R., T.K., H.F., W.H., and K.Y. contributed to data analysis and writing and editing of the paper.

Corresponding author

Correspondence to Kui Yu.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Guohua Jia and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wan, W., Zhang, M., Zhao, M. et al. Room-temperature formation of CdS magic-size clusters in aqueous solutions assisted by primary amines. Nat Commun 11, 4199 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing