In Situ Two-Phase Ligand Exchange: A New Method for the Synthesis

Apr 6, 2017 - Au21–xCux(SR)15 (x = 2–5) exhibits a significantly different configuration. ... Engineering a red emission of copper nanocluster sel...
0 downloads 0 Views 1MB Size
Communication pubs.acs.org/JACS

In Situ Two-Phase Ligand Exchange: A New Method for the Synthesis of Alloy Nanoclusters with Precise Atomic Structures Sha Yang,† Jinsong Chai,† Yongbo Song,† Jiqiang Fan, Tao Chen, Shuxin Wang, Haizhu Yu, Xiaowu Li, and Manzhou Zhu* Department of Chemistry and Center for Atomic Engineering of Advanced Materials, Anhui University, Hefei, Anhui 230601, People’s Republic of China S Supporting Information *

ABSTRACT: A new method termed “in situ two-phase ligand exchange” was developed to obtain alloy nanoclusters. With this approach, a series of alloy nanoclusters were obtained for the first time, including Au20Ag1(SR)15, Au21−xAgx(SR)15 (x = 4−8), Au21−xCux(SR)15 (x = 0, 1), and Au21−xCux(SR)15 (x = 2−5) (R = tert-butyl). Interestingly, single-crystal X-ray crystallography (SCXRD) shows that their frameworks are all alike except for Au21−xCux(SR)15 (x = 2−5), indicating that more Cu dopants alter the structure. Au21−xCux(SR)15 (x = 2−5) exhibits a significantly different configuration. The optical absorption spectra of these bimetal nanoclusters (NCs) show distinct characteristic peaks, indicating that the metal-doping remarkably affects the electronic structure of NCs. The DFT calculations were also employed for determination of NC 1−3 frameworks and understanding their optical properties.

N

oble metal nanoclusters (NCs) protected by thiolate ligands have emerged as a new class of nanomaterials that have been extensively studied.1−5 Their molecular-like properties (such as magnetism,6 HOMO−LUMO transitions,7 and enhanced luminescence8) are highly promising for applications in catalysis,9 biological imaging,10 sensoring,11 and biomedical fields.12 Specifically, recent studies have shown that the physicochemical properties of alloy NCs could be critically improved compared to those of the single component systems. Generally, the synthetic methods for the reported alloy NCs protected by thiolate ligands can be categorized into three types. First, the “one-pot” approach was widely used because of its operational simplicity, in which all requisite reagents were mixed together (Figure 1a). In this way, a series of alloy NCs have been prepared, including M x Au 25−x (SR) 18 , 13−15 Au38−xAgx(SR)24,16 Au3Ag38(SR)24X5,17 Au12Ag32(SR)30,18 Au24Ag46(SR)32,19 (AuAg)144(SR)60,20 and Au12+nCu32(SR)30+n.21 Second, the metal-exchange strategy utilizing the reaction of the templated, atomically precise metal NCs with foreign metal atoms or metal−SR complex has also been widely used to synthesis alloy NCs such as CdAu24(SR)18 and AuAg24(SR)18.22−24 Similar to the metalexchange approach, the ligand-exchange (Figure 1b) approach needs template alloy NCs to react with other ligands, which generally occurs under the two-phase circumstances. With this © 2017 American Chemical Society

Figure 1. Schematic of the synthetic method of (a) one-pot, (b) ligand exchange, and (c) in situ two-phase ligand exchange (color labels: green = Au, light blue = Ag, blue = Cu, red = S, purple = GSH, gray stick = C/H).

strategy, the product, such as Au 1 5 Ag 3 (SR) 1 4 and Au36−xAgx(SR)24, is often monodispersed with high purity.25 Inspired by the simplicity of the one-pot method and the effectiveness of the two-phase ligand-exchange approach, we tried to explore a combinative synthetic strategy, i.e., the in situ two-phase ligand-exchange method (Figure 1c, in situ for short). In this system, the two-phase interaction occurs at the early stage, and the target alloy (and even homometallic) NCs could be synthesized in a short amount of time. To our delight, a series of thiolate-stabilized alloy NCs, including Au20Ag1(SR)15 (NC-1), Au21−xAgx(SR)15 (x = 4−8, NC-2), Au21−xCux(SR)15 (x = 0 or 1, NC-3), and Au21−xCux(SR)15 (x = 2−5, NC-4) have been successfully synthesized. Interestingly, all of these NCs consist of six free electrons. The full details of the synthesis are provided in the Supporting Information (SI), and only a brief description is given here. As shown in Figure 1c, foreign metal salt and Received: January 20, 2017 Published: April 6, 2017 5668

DOI: 10.1021/jacs.7b00668 J. Am. Chem. Soc. 2017, 139, 5668−5671

Communication

Journal of the American Chemical Society

aqueous phase of the in situ method are mainly responsible for the successful synthesis of these alloy NCs. The crystals of these alloy NCs were all obtained via crystallization in CH2Cl2/CH3OH for 2−3 days. Interestingly, the SC-XRD shows that these alloy nanoclusters have two kinds of configuration. As for NC-1−3, they share a common structure, which contains a M14 core (Figure 2a, M = Au, Ag, and Cu) protected

glutathione (GSH) were codissolved in H2O. After a few minutes, HAuCl4·3H2O was added. Then, the toluene solution of borane tert-butylamine complex and tert-butyl (tBu) mercaptan was added. After a few hours, the mixture in the organic phase was washed several times with CH3OH to remove the excess ligands and byproducts. Because the ligand and the reducing agent were added simultaneously (Figure 1c), the selective reduction and ligand exchange are synchronous. In addition, the reduction promotes the ligand exchange of the hydrophilic GSH with the hydrophobic tert-butyl mercaptan. Therefore, compared to the traditional ligand exchange, the operation of this new synthetic method is easier, and the reaction time is shorter. For a better comparison, we also tried to synthesize the target alloy NCs with the conventional one-pot and ligandexchange methods (details are given in the SI). Interestingly, neither one-pot nor the conventional ligand-exchange method could be used to synthesize NC-2−4 (Figures S1−S3), and the synthesis of NC-1 with the conventional ligand exchange approach required a slightly longer amount of time than that with the in situ method. To interpret the difference between the in situ and the conventional ligand-exchange methods, we tracked the entire reaction process in aqueous phase by UV−vis absorption spectroscopy. As shown in Figure S4, for each synthetic system (NC-1−4), the intermediates formed via the in situ and the conventional ligand-exchange ways are totally different during the monitoring time. Taking NC-2 as an example, using the in situ system, the UV−vis spectrum of the precursor shows two peaks at 480 and 730 nm after 10 min. With prolonged time, the peak at 470 nm becomes more pronounced and a new peak at 640 nm appears. By contrast, using the conventional ligandexchange method, the UV−vis spectrum of the precursor shows only one weak peak at 490 nm throughout the reaction process (more detailed analysis on the other systems is shown in the SI). The distinct UV−vis spectra demonstrate the different compositions of the precursors in these two synthetic systems. To better characterize the compositions of the precursors, we carried out MALDI-MS tracking (Figure S5). Herein, NC-2 was also chosen as the example. Using the in situ method, three broad peaks at ∼5000, ∼7000, and ∼11000 Da gradually appear on MALDI-MS with prolonged time. By contrast, using the conventional ligand-exchange method, only one broad peak at ∼3000 Da appears even after 200 min. Consistent with the time-dependent UV−vis characterizations, the MALDI-MS analysis also implies that the precursors are totally different between these two synthetic systems. In addition, the precursors are more polydisperse in the in situ approach (compared to that in the conventional ligand-exchange method). Similar phenomena were also found in all other synthetic systems (i.e., NC-1, NC-3, and NC-4), and the details are provided in the SI. Interestingly, according to FTIR spectrometry analysis (Figure S6), the precursors formed via the in situ method show significant Au−StBu signals, whereas the signal is invisible in the FTIR spectra corresponding to the conventional ligandexchange method. The results indicate that the precursors of the in situ method were coprotected by t-BuSH and GSH. In other words, the reduction of Au(I)SG into polydispersed precursors occurs with simultaneous ligand exchange. According to the time-dependent UV−vis, MALDI-MS analysis, and FT-IR characterization, we suggest that the high dispersity of the precursors and the prior ligand exchange in the

Figure 2. (a) Fourteen-atom metal core, (b) surface staple motifs, and frameworks of (c) NC-1, (d) NC-2, and (e) NC-3 (color labels: green = Au, light blue = Ag, blue = Cu/Au, bluish-green = Au/Ag, magenta = Au/Ag/Cu, red = S; no C or H atoms are shown).

by two M2S3 staple motifs (Figure 2b, labeled with i), one M3S4 staple motif (Figure 2b, labeled with ii), and five bridging S atoms. The frameworks in NC-1−3 have been well-reproduced by DFT calculations (Figure S8; please see the SI for more details). Interestingly, two pairs of enantiomers were found in each unit cell of these NCs (Figures S10 and S11). For clarity, one enantiomer for each NC was chosen for the detailed structural analysis. To determine details of the atom-packing mode, we focused on their frameworks (Figure 2c−e). In NC-1, the single Ag atom is located at the 14-atom metal core, which results in an Au13Ag1 kernel. The bond lengths of the M−M (M = Au or Ag) range from 2.645 to 3.189 Å (average: 2.916 Å). The metal atoms in the staple motifs of NC-1 are all Au atoms. The Au−S bond lengths in the staples are in the range of 2.274−2.336 Å. The Au−S distances in the bridging SR moieties (average: 2.320 Å) are relatively longer than those in the staple motifs (average: 2.302 Å). The structure of NC-2 highly resembles that of NC-1 except for the sites and the number of doped Ag atoms. In NC-2, the doped Ag atoms locate on both the 14-atom core and the surface staple motifs (Figure 2d). The average bond length of M−M (M = Au or Ag) in the core is 2.613 Å, which is shorter than that of NC-1, indicating the additional Ag atoms in the core significantly affect the geometric structure. Moreover, the average M−S (M = Au or Ag) bond length in NC-2 (2.322 Å) is longer than that in NC-1. According to the SC-XRD analysis, NC-3 is made up of Au21(SR)15 and Au20Cu1(SR)15. The single Cu atom in Au20Cu1(SR)15 is located at the staple motif of NC-3 (Figure 2e). This position is different from that of the single Ag atom in NC-1 and is consistent with the position of one Ag atom in NC-2. The Au−Au distances in the core range from 2.645 to 3.331 Å (average: 2.927 Å). The M−S (M = Cu/Au) bond lengths in the staples are in the range of 2.262−2.317 Å (average: 2.297 Å), which are slightly shorter than those of Au−S bonds in NC-1 (average: 2.302 Å). Furthermore, the 5669

DOI: 10.1021/jacs.7b00668 J. Am. Chem. Soc. 2017, 139, 5668−5671

Communication

Journal of the American Chemical Society single Cu atom doping also results in slightly shorter Au−S bond lengths in the bridging SR cases (average: 2.315 Å) than those in NC-1 (average: 2.320 Å). As for NC-4, it also contains a 21-atom core protected by 15 thiolate ligands (Figure S12). However, NC-4 exhibits a similar configuration to that recently reported for Au21(S-Adm)15,26 which is totally different from those of the aforementioned three alloy NCs (Figure 3). First, an Au9Cu1 core (Figure 3b) is

Figure 4. UV−vis spectra of these alloy nanoclusters.

other three NCs, it is understandable that NC-4 displays a significantly different spectral profile with three peaks at 487, 520, and 610 nm and two shoulder bands at 375 and 422 nm (Figure 4, pink line). The dramatic changes in the optical spectra suggest that different foreign atom(s) doping largely changes the electronic structure of the NCs. In summary, a new method termed “in situ two-phase ligand exchange” has been developed. A series of alloy NCs with six free electrons have been synthesized, and their structures have been determined by SC-XRD, which shows that the NC-1−3 have similar structures and NC-4 has a distinct configuration. The optical absorption spectra of these bimetal NCs show obvious differences, indicating that the metal-doping can remarkably affect the electronic structure of the NCs. This in situ two-phase ligand-exchange method might potentially act as a powerful alternative for synthesizing alloy NCs and even homothallic ones with unprecedented atomic structures.

Figure 3. Crystal structure of NC-4: (a) framework, (b) core, (c, d) surface motifs, and (e, f) assembled motifs (color labels: green = Au, bluish-green = Au/Cu, blue = Cu, red = S; no C or H atoms are shown).

identified in the framework of NC-4 (Figure 3a), and the average M−M (M = Au or Cu) bond length in the core is 2.826 Å. The Au9Cu1 core can be viewed as two octahedron sharing one edge. Moreover, the core is protected by one long octameric M8(SR)9 motif (Figure 3c, labeled with i), one dimeric M2(SR)3 (Figure 3c, labeled with ii), one monomeric M1(SR)2 staple (Figure 3d, labeled with iii), and one bridging S atom. As shown in Figure 3e and f, the long octameric M8(SR)9 motif connects to the dimeric M2(SR)3 staple through two Cu−S bonds, which may contribute to the stability of NC-4. The bond lengths of Au−S (or Cu−S) in the staple motifs range from 2.185 to 2.303 Å (average: 2.263 Å). The accurate formulations of these alloy NCs were further determined by ESI-MS (Figures S13−16) and TGA (Figures S17−S20). All these results demonstrate that the formula matches well with the one obtained by single-crystal X-ray diffraction (see detailed discussion in the SI). Moreover, these alloy NCs were further studied by XPS (Figures S21−S25) in which the Au 4f peaks of these alloy NCs are all higher than that of Au(0) (84.0 eV), indicating that the Au atoms are all positively charged. Meanwhile, the Ag 3d peaks in both NC-1 (368.30 eV) and NC-2 (368.35 eV) are higher (i.e., reduction side) than that of Ag(0) (367.9 eV). The binding energies of Cu in NC-3 (933.15 eV) and NC-4 (933.05 eV) demonstrate that Cu atoms are close to that of Cu(0).23 The optical absorption spectra of the alloy NCs are shown in Figure 4. Despite the different numbers of Ag dopants, NC-1 and NC-2 exhibit similar optical absorption spectra. The UV− vis spectrum of the NC-1 displays an intense peak at 575 nm and two weak shoulder peaks at 385 and 452 nm (Figure 4, black line). Compared with NC-1, the spectrum of NC-2 (Figure 4, red line) slightly red-shifts with its main absorption band centered at 585 nm. A new shoulder band at 665 nm was also seen in NC-2. Unlike Au−Ag bimetal NCs, NC-3 exhibits two major absorption peaks centered at 580 and 780 nm as well as two shoulder peaks at 380 and 500 nm. The differences for the optical properties and electronic structures of NC 1−3 were evidenced and further analyzed by DFT calculations (Figure S9). Because the structure of NC-4 is distinct from those of the



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b00668. Details of the syntheses, crystallization, X-ray analysis, and supporting figures (PDF) Crystallographic information for NC-1, NC-2, NC-3, and NC-4



AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Shuxin Wang: 0000-0003-0403-3953 Haizhu Yu: 0000-0003-3010-1331 Manzhou Zhu: 0000-0002-3068-7160 Author Contributions †

S.Y., J.C., and Y.S. contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We acknowledge financial support from the National Natural Science Foundation of China (21631001, 21372006, and U1532141), the Ministry of Education, the Education Department of Anhui Province, and the 211 Project of Anhui University. 5670

DOI: 10.1021/jacs.7b00668 J. Am. Chem. Soc. 2017, 139, 5668−5671

Communication

Journal of the American Chemical Society



REFERENCES

(1) Jin, R.; Zeng, C.; Zhou, M.; Chen, Y. Chem. Rev. 2016, 116, 10346. (2) Schmid, G. Chem. Soc. Rev. 2008, 37, 1909. (3) Jin, R. Nanoscale 2010, 2, 343. (4) Desireddy, A.; Conn, B. E.; Guo, J.; Yoon, B.; Barnett, R. N.; Monahan, B. M.; Kirschbaum, K.; Griffith, W. P.; Whetten, R. L.; Landman, U.; Bigioni, T. P. Nature 2013, 501, 399. (5) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547. (6) Song, Y.; Wang, S.; Zhang, J.; Kang, X.; Chen, S.; Li, P.; Sheng, H.; Zhu, M. J. Am. Chem. Soc. 2014, 136, 2963. (7) Shichibu, Y.; Negishi, Y.; Watanabe, T.; Chaki, N. K.; Kawaguchi, H.; Tsukuda, T. J. Phys. Chem. C 2007, 111, 7845. (8) Das, A.; Li, T.; Nobusada, K.; Zeng, Q.; Rosi, N. L.; Jin, R. J. Am. Chem. Soc. 2012, 134, 20286. (9) Song, Y.; Fu, F.; Zhang, J.; Chai, J.; Kang, X.; Li, P.; Li, S.; Zhou, H.; Zhu, M. Angew. Chem. 2015, 127, 8550. (10) Yang, H.; Wang, Y.; Huang, H.; Gell, L.; Lehtovaara, L.; Malola, S.; Häkkinen, H.; Zheng, N. Nat. Commun. 2013, 4, 2422. (11) Zhu, M.; Aikens, C. M.; Hendrich, M. P.; Gupta, R.; Qian, H.; Schatz, G. C.; Jin, R. J. Am. Chem. Soc. 2009, 131, 2490. (12) Murray, R. W. Chem. Rev. 2008, 108, 2688. (13) (a) Negishi, Y.; Iwai, T.; Ide, M. Chem. Commun. 2010, 46, 4713. (b) Kumara, C.; Aikens, C. M.; Dass, A. J. Phys. Chem. Lett. 2014, 5, 461. (14) (a) Negishi, Y.; Kurashige, W.; Niihori, Y.; Iwasa, T.; Nobusada, K. Phys. Chem. Chem. Phys. 2010, 12, 6219. (b) Fields-Zinna, C. A.; Crowe, M. C.; Dass, A.; Weaver, J. E. F.; Murray, R. W. Langmuir 2009, 25, 7704. (15) Qian, H.; Jiang, D.; Li, G.; Gayathri, C.; Das, A.; Gil, R. R.; Jin, R. J. Am. Chem. Soc. 2012, 134, 16159. (16) Kumara, C.; Dass, A. Nanoscale 2012, 4, 4084. (17) Wang, Z.; Senanayake, R.; Aikens, C. M.; Chen, W.; Tung, C.; Sun, D. Nanoscale 2016, 8, 18905. (18) Yang, H.; Wang, Yu.; Huang, H.; Gell, L.; Lehtovaara, L.; Malola, S.; Häkkinen, H.; Zheng, N. Nat. Commun. 2013, 4, 2422. (19) Wang, S.; Jin, S.; Yang, S.; Chen, S.; Song, Y.; Zhang, J.; Zhu, M. Sci. Adv. 2015, 1, e1500441. (20) Kumara, C.; Dass, A. Nanoscale 2011, 3, 3064. (21) Yang, H.; Wang, Y.; Yan, J.; Chen, X.; Zhang, X.; Häkkinen, H.; Zheng, N. J. Am. Chem. Soc. 2014, 136, 7197. (22) Bootharaju, M. S.; Joshi, C. P.; Parida, M. R.; Mohammed, O. F.; Bakr, O. M. Angew. Chem., Int. Ed. 2016, 55, 922. (23) Wang, S.; Song, Y.; Jin, S.; Liu, X.; Zhang, J.; Pei, Y.; Meng, X.; Chen, M.; Li, P.; Zhu, M. J. Am. Chem. Soc. 2015, 137, 4018. (24) (a) Yao, C.; Lin, Y.; Yuan, J.; Liao, L.; Zhu, M.; Weng, L.; Yang, J.; Wu, Z. J. Am. Chem. Soc. 2015, 137, 15350. (b) Liao, L.; Zhou, S.; Dai, Y.; Liu, L.; Yao, C.; Fu, C.; Yang, J.; Wu, Z. J. Am. Chem. Soc. 2015, 137, 9511. (25) (a) Xiang, J.; Li, P.; Song, Y.; Liu, X.; Chong, H.; Jin, S.; Pei, Y.; Yuan, X.; Zhu, M. Nanoscale 2015, 7, 18278. (b) Fan, J.; Song, Y.; Chai, J.; Yang, S.; Chen, T.; Rao, B.; Yu, H.; Zhu, M. Nanoscale 2016, 8, 15317. (26) Chen, S.; Xiong, L.; Wang, S.; Ma, Z.; Jin, S.; Sheng, H.; Pei, Y.; Zhu, M. J. Am. Chem. Soc. 2016, 138, 10754.

5671

DOI: 10.1021/jacs.7b00668 J. Am. Chem. Soc. 2017, 139, 5668−5671