How do Water Solvent and Glutathione Ligands Affect the Structure

Letter pubs.acs.org/JPCL

How do Water Solvent and Glutathione Ligands Affect the Structure and Electronic Properties of Au25(SR)18−? Víctor Rojas-Cervellera,† Carme Rovira,†,‡ and Jaakko Akola*,§,∥ †

Departament de Química Orgànica and Institut de Química Teòrica i Computacional (IQTCUB), Universitat de Barcelona, Martí i Franquès 1, 08028 Barcelona, Spain ‡ Institució Catalana de Recerca I Estudis Avançats (ICREA), Passeig Lluís Companys, 23, 08018 Barcelona, Spain § Department of Physics, Tampere University of Technology, P.O. Box 692, FI-33101 Tampere, Finland ∥ COMP Centre of Excellence, Department of Applied Physics, Aalto University, FI-00076 Aalto, Finland S Supporting Information *

ABSTRACT: The effects of aqueous solvent and biological ligands on the structural and electronic properties of thiolate-protected Au25(SR)18− clusters have been studied by performing quantum mechanics/molecular mechanics (QM/MM) simulations. Analysis of bond distances and angles show that the solvated nanocluster experiences modest structural changes, which are reflected as flexibility of the Au core. The hydrophilic glutathione ligands shield the metallic core effectively and distort its symmetry via sterical hindrance effects. We show that the previously reported agreement between the calculated HOMO−LUMO gap of the cluster and the optical measurement is due to cancellation of errors, where the typical underestimation of the theoretical band gap compensates the effect of the missing solvent. The use of a hybrid functional results in a HOMO−LUMO gap value of 1.5 eV for the solvated nanocluster with glutathione ligands, in good agreement with optical measurements. Our results demonstrate that ligand/solvent effects should be considered for a proper comparison between theory and experiment.

T

binding target. To eliminate these interactions, some functionality needs to be introduced to the AuMPC to selectively bind the target molecules. Glutathione (GSH), a tripeptide formed by a glutamate, a cysteine, and a glycine (see Figure 1), is often used as a ligand8,18,19 because it is able to selectively bind proteins as glutathione-S-transferase3 and single-chain Fv antibody fragments.2 GSH can be written as γ-Glu-Cys-Gly, and it has 2 asymmetric carbons that introduce chirality. At physiological pH, both carboxylic acids are deprotonated (i.e., negatively charged), whereas the amino group is protonated (GSH total charge of −1). Although it has been demonstrated that Au25(GSH)18− is very stable,19 its crystallographic structure has not yet been resolved. The presence of a bulky biological ligand carrying several charged groups (GSH) is expected to modify the structural and electronic properties of AuMPC with respect to small monocharged ligands such as synthetic alkanethiols. In this Letter, we assess these differences by investigating the structure and electronic properties of two clusters of formula Au25(SR)18− (R = glutathione or alkanethiolate) in vacuum and aqueous solution. For computational reasons, we have used

hiolate monolayer-protected gold clusters (AuMPC) have been extensively used as carriers of biological molecules, namely DNA,1 antibodies,2 and specific proteins as glutathioneS-transferase.3 The synthesis of various AuMPCs of different sizes with a variety of protecting ligands for the gold core was reported recently.4,5 As a specific example, the structure of Au25(SR)18− (R being an alkyl substituent) was predicted computationally6 and verified experimentally by X-ray measurements of a crystallized sample.7 Au25(SR)18− is one of the smallest ligand-protected gold clusters for which the X-ray structure is known.8−10 It is spherical and symmetric, constituted by an icosahedral Au13 gold core (one central atom and 12 atoms in the vertices of the icosahedra) and six Au2(SR)3 dimeric staple motifs,4 which protect the gold core in an octahedral arrangement (Supporting Information Figure S1). Recently, structures of smaller ligand-protected gold clusters have been predicted computationally.11−14 Small alkanethiolates such as 2-phenylethanethiolate,7 ethanethiolate,15 or p-bromobenzenethiol16 have been used as ligand units for Au25(SR)18−. Furthermore, chiral ligands such as (R or S)-2-amino-2-phenylpropanethiolate (SCH2C*H(NH 2 )CH 2Ph) and (R or S)-2-methyl-2-phenylthiolate (SCH2C*H(CH3)CH2Ph), have been used as well.17 All these alkyl chains are known to be hydrophobic. Therefore, if these AuMPC are used to bind biological molecules, there may be nonspecific hydrophobic interactions that interfere in the © XXXX American Chemical Society

Received: June 29, 2015 Accepted: September 11, 2015

3859

DOI: 10.1021/acs.jpclett.5b01382 J. Phys. Chem. Lett. 2015, 6, 3859−3865

Letter

The Journal of Physical Chemistry Letters

solution, and (3) Au25(GSH)18− in solution. The structure of the Au25(SCH3)18− system was optimized with DFT, applying the PBE functional,29 commonly used to describe AuMPCs, followed by room-temperature ab initio molecular dynamics simulations for 7.5 ps using the CPMD program.30 An explicit water solvent was considered for the last two systems (Figure 1 and Supporting Information, SI), which were modeled with the CPMD quantum mechanics/molecular mechanics (QM/MM) approach.31 Classical molecular dynamics simulations, up to 8 ns, were used to equilibrate the solvated systems using the FF99SB and TIP3P force fields for the side-chains of GSH and the water molecules, respectively (see further details in the SI), followed by 7.5 ps of QM/MM MD simulations. As shown in Figure 2, there are three types of gold atoms (a, a′, and c) and two types of sulfur atoms (S and Sap) in the AuMPCs. The most relevant parameters defining the systems investigated are shown in Table 1. To facilitate the structural analysis, radar charts were constructed, taking Au25(SCH3)18− in the gas phase as the reference system. The results of Table 1 show that the solvent does not affect the global structure of the Au25(SCH3)18− framework, as the Au13 core and the six (protecting) staple motifs are preserved. However, the cluster expands: the largest difference between the isolated and the solvated Au25(SCH3 )18− at 0 K corresponds to Au(c)−Au(a′) distance (involving outer core gold atoms and those of the staple motifs), which increases by 0.11 Å (3.2%). As a consequence, the Au(c)−S−Au(c) angle involving the apex S atom and the Au(a′)−S−Au(c) angle at the core-ligand interface, change by −7.3° and 4.1°, respectively. During the MD simulation (300 K), the average values of the Au−Au bonds show a further increase up to 0.10 Å (for Au(a′)−Au(a′)), while the S−Au bonds remain almost intact (within ±0.03 Å). The S−Au(a′) bonds that link the staple motifs to the Au13 core and their variation during the of QM/MM MD simulations are illustrated in Figure S2. Comparing the structure of the AuMPCs with the two ligands (thiolate and glutathione), we again observe that the icosahedral Au13 gold core and the six dimeric staple motifs are not affected by ligand size, which is in agreement with a previous NMR and mass spectrometry experiment.32 However, significant variations are found in the GSH distances (Figure 2 and Table 1), and in particular, Au(a′)−Au(a′) bonds increase up to 0.08 Å. Furthermore, the increase of the Au−Au bond distances is even more pronounced for GSH at room temperature (0.09−0.14 Å). This is due to the solvent effects and steric hindrance of the bulky ligands, which modify the Au−Au bonding via mechanochemical coupling. Analysis of the radius of gyration shows that the Au13 core expands slightly when GSH is used to better accommodate the bulky ligands, and the first shell water solvent causes electrostatic effects on the electron density. The a′−S−c angle decreases due to the formation of hydrogen bond interactions among the GSH ligands. GSH contains three charged groups, and the tendency of these groups to form hydrogen bonds with each other causes ligand rotation. Figure 3A,B shows the change in the AuMPC symmetry from Ci for the methanethiolate ligand to C1 for GSH. Such transformation was previously suggested in ref 22, where N-acetyl-cysteine (NAC) was used to mimic GSH. A histogram of the main distances during the AIMD simulation of Au25(GSH)18− is shown in Figure S3. For the Au−Au bonds, the distributions are very broad, which demonstrates the flexibility of these bonds, termed previously as “fluxionality” for

Figure 1. Solvated Au25(SCH3)18− (top) and Au25(GSH)18− (bottom). For Au25(SCH3)18−, the QM region is zoomed on the right, where the QM zone is shown in color (Au in green, S in yellow, C in black and H in gray). For Au25(GSH)18−, the ball and stick representation of a GSH peptide (negatively charged) is shown on the right. Hydrogens are shown transparent for clarity. The black dashed line shows the gamma peptide linkage between glutamate and cysteine. The green dashed line shows the normal peptide linkage between cysteine and glycine. The images were generated with VMD.28

methanethiolate, i.e., the smallest alkanethiolate, as a model of the synthetic ligands. This system already captures the essential chemistry at the Au−S interface layer. We perform ab initio molecular dynamics (AIMD) simulations based on the density functional theory (DFT). One well-known issue of standard DFT methods is that the energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO), i.e., the HOMO−LUMO gap, tends to be severely underestimated with respect to the experimental values, although tendencies are correct.20 In principle, the absolute value of the HOMO−LUMO gap calculated by DFT should not be used as a reference for a direct comparison with optical measurements. Nevertheless, recent computational studies (with standard DFT) have reported values of the HOMO− LUMO gap for phosphine-halide and thiolate-protected gold (and silver) clusters in very close agreement with experiments.21−24 The role of implicit solvent in structural and electronic properties has been considered for small Au clusters25 and L-cysteine-covered Au20,26 both at the DFT level. Clearly, this aspect deserves more attention, and we have addressed it by modeling Au25(SR)18‑ in vacuum and explicit water solvation. Moreover, we consider as the R group both a small methanethiolate (SCH3) as well as the full GSH ligand (negative overall charge, see Figure 1) used in experiments.27 Ligands can induce distortions in the Au25(SR)18− framework via mechanochemical coupling,22 and we have quantified here how the distortion of the AuMPC caused by the bulky biological ligand (GSH), as well as the presence of water, affect the HOMO−LUMO gap. Three systems were prepared for the simulations: (1) Au25(SCH3)18− in the gas phase, (2) Au25(SCH3)18− in 3860


Letter


Figure 2. Top: Au25(SR)18− atom labeling used throughout the study. Color codes: Au (green), S (yellow). Center: Representative bond distances (left) and angles (right) of Au25(SCH3)18− isolated and solvated in water. Bottom: Representative bond distances (left) and angles (right) of Au25(GSH)18− solvated in water with respect to the isolated Au25(SCH3)18−. The labels 0 and 300 K are used to distinguish the optimized structure with respect to the average structure at room temperature, respectively. Values in the chart axes correspond to a percentage with respect to the reference Au25(SR)18− system in vacuum (shaded area).

Table 1. Structural Parameters of the Au25(SR)18− Nanoclusters Au25(SCH3)18‑ Au(a)−Au(a′) Au(a′)−Au(a′)b Au(a′)−Au(a′)c Au(c)−Au(a′) S−Au(a′) S−Au(c) Au(a′)−S−Au(c) Au(c)−S−Au(c) S−Au(c)−S a

Au25(GSH)18‑

vacuum (0K)

solvated (0 K)

solvated (300 K)a

vacuum (0 K)

solvated (300 K)a

2.85 2.84 3.04 3.41 2.40 2.32 92.5 100.6 172.9

2.86 2.83 3.06 3.52 2.40 2.32 96.6 93.3 173.4

2.90 2.93 3.09 3.52 2.43 2.33 95.60 94.60 170.10

2.91 2.85 3.12 3.42 2.43 2.32 92.3 94.7 167.3

2.99 2.93 3.17 3.47 2.44 2.33 93.9 91.8 164.9

Average values from the MD simulation at 300 K.

b,c

Au(a′)-Au(a′) distances display two values due to symmetry breaking.

bare Au clusters.33,34 Hence, the introduction of bulkier ligands can easily induce changes in the Au13 core. Another structural effect of the GSH ligands concerns the modification of the water structure. These ligands provide better “shielding” than small alkanethiolates, protecting the Au core more effectively from external agents. The solvent accessible surface (SASA) of AuMPCs and water molecules within 4.5 Å from any gold atom are presented in Figure 3C,D. The SASA value of the Au core is very different for the two systems: 16.5 Å2 and 3.5 Å2 for SCH3 and GSH, respectively. All the surface gold atoms of Au25(SCH3)18− are solventexposed (Figure 3B). On the contrary, the Au atoms in Au25(GSH)18− are mostly covered by the GSH ligands, which

effectively prevent the solvent from approaching the metallic core. However, a few water molecules are still able to overcome the shielding of peptides and reach some of the Au atoms (see also the radial distribution functions of the solvent in Figure S4). Figure 3E,F shows the hydrogen bonding network around the cluster staple motifs. Since the SCH3 ligands are hydrophobic, their interaction with water molecules is very weak, resulting in an organized second solvation shell around the cluster (Owater···Au distances around 7.5 Å; see Figure S4). On the other hand, the hydrophilic GSH ligands, with charged terminal groups (carboxylates, amines), form several hydrogen bonds with the solvent and other ligands. 3861


Letter


Figure 3. (A,B) Structures of Au25(SCH3)18− and Au25(GSH)18− belong to the Ci and C1 point groups, respectively. The side groups are shown semitransparent. (C,D) Solvent accessible surface area of Au25(SCH3)18− and Au25(GSH)18−. Water molecules that are up to 4.5 Å from gold atoms are also shown. The van der Waals surface of methanethiolate or GSH is shown as a semitransparent surface to illustrate their shielding effect. Gold atoms that are more exposed to the solvent are depicted in light green. (E) One dimeric staple motif in Au25(SCH3)18− and the single water that is less than 2 Å from the methanethiolates. (F) Hydrogen bonding of GSH in one dimeric staple motif in Au25(GSH)18−. Dashed red lines represent hydrogen bonds. GSH of the highlighted staple motif is represented by ball-and-stick, while the other ligands or water molecules are represented by sticks. Color codes: Au (green), S (yellow), N (blue), O (red), C (black), and H (gray).

Figure 4. (A) Change of the atomic Hirshfeld charges for the most representative atoms of Au25(SCH3)18− and Au25(GSH)18− solvated in water with respect to the ones obtained for the isolated Au25(SCH3)18− (blue area, change in percentage). (B) HOMO−LUMO gap of these systems with PBE and PBE0 exchange-correlation functionals. Green bars correspond to Au25(SCH3)18−, whereas yellow bars refer to Au25(GSH)18−.

Atomic (Hirshfeld35) charges of Au25 (SCH 3)18− and Au25(GSH)18− were computed to analyze the effect of the

solvent and the type of ligand on the electronic structure of the AuMPC. Figure 4a compares the atomic charges for the 3862


Letter


As a summary, we have performed QM/MM simulations for Au25(SCH3)18− and Au25(GSH)18− clusters in order to elucidate the effect of introducing a biological ligand (GSH) and explicit water solvent in the model. The overall changes in the cluster core structure and electronic properties are consistent for both side groups, although their character is different, and this suggests that our results can be extended to other AuMPCs in aqueous environment. Our findings demonstrate that the general structure of Au25(SR)18− is not modified, and that the Au13 core and the six dimeric SR−Au− SR−Au−SR staple motifs are always preserved. However, the Au−Au bond distances and angles between S and Au experience visible changes under different conditions. The shielding effect of the peptides is larger than that for methanethiolates due to their larger size, and GSH protects the gold surface toward nucleophilic attacks of external agents that can lead to undesired interactions. While methanethiolate is hydrophobic, the GSH ligands actively form hydrogen bonds with water and other neighboring peptides (amine and carboxylate groups). Upon changing the ligand from SCH3 to GSH, the changes in charge distribution affect the HOMO− LUMO gap, which is further modified by the loss of symmetry from Ci to C1 due to the flexible nature of Au−Au bonds. A lower symmetry is associated with a smaller HOMO−LUMO gap. Our main finding is that the HOMO−LUMO gap of AuMPCs depends sensitively on the ligands and solvent environment. In particular, the role of solvent has been typically neglected in the theoretical studies of monolayer-protected Au and Ag clusters, which have reported HOMO−LUMO gaps and/or optical absorption spectra in good agreement with experiments for standard DFT functionals (GGA).21−24,36−38 We have demonstrated here for Au25(SCH3)18− that the water solvent and ligand-exchange with GSH reduce the calculated band gap drastically down to 0.65 eV, underestimating the experiment by a factor of 2. Therefore, the good performance of GGA functionals (such as PBE) in reproducing the optical gap is due to a cancellation of errors. The calculations of AuMPCs in the vacuum give systematically larger values for the HOMO−LUMO gap than in the solvent. Hence, these larger values compensate the fact that DFT underestimates band gaps in general. The hybrid PBE0 functional is able to produce a computed value (1.54 eV) that is slightly above the experimental optical gap once the system is described realistically, i.e., the AuMPC composition (ligand) is the same as in experiments, and the system is surrounded by explicit solvent. Our result (HOMO−LUMO gap) represents an upper bound for the optical absorption gap,21 and future work is needed to compute the optical absorption spectrum of such AuMPCs by using time-dependent DFT and QM/MM.

different types of gold and sulfur atoms using a radar chart (see Table S1 for the full list of charges). As shown in Figure 4, introducing the solvent makes sulfur atoms more negative (lateral sulfur: −0.26e and −0.31e for Au25(SCH3)18− in vacuum and solvated, respectively). The corresponding charge is partially transferred from the negatively charged central Au atom (a), which has an effective charge of −0.65e. The effect becomes more pronounced for Sap (apex, −0.48e) when the ligand is GSH, and hence the staple motifs become less electrophilic. The Au atoms in the staple motifs (c) are positively charged (+0.40e) reflecting their different oxidation state. The gold atom charges in the Au13 icosahedron vertices are insensitive to the changes in environment. In the following, we shall compare the HOMO−LUMO gap with the optical absorption gap. Although these are not the same quantity, they are closely related, and the HOMO− LUMO gaps for generalized gradient approximation (GGA) functionals display a strong correlation with the computed and/ or experimental lowest optical transitions of ligand-protected Au clusters.21−24,36−38 For hybrid functionals, a recent study shows that the lowest optical transition is overestimated by the HOMO−LUMO gap,21 and this tendency should be kept in mind. Previous analysis has shown that the electronic structure of Au25(SR)18− corresponds to a closed-shell configuration of an 8-electron superatom with occupied 1S (two electrons) and 1P (six electrons, degenerate HOMO) orbitals, and that the LUMO orbitals have 1D symmetry (Figure S5 of the Supporting Information).7,39 The HOMO−LUMO gaps for the optimized structures of the studied systems are reported in Figure 4B. The results clearly evidence that the solvent reduces the gap significantly by 22% for the systems with SCH3 as ligand, with respect to the same system in vacuum (1.19 and 0.93 eV, respectively). Taking into account that the experimental optical gap (1.3 eV)27 is higher than the gas phase computed value, the reduction obtained by adding the solvent is in the wrong direction! However, there is another element missing in our analysis, which has been based so far in the Au25(SCH3)18− system. The HOMO−LUMO gap is expected to depend on the ligand as a consequence of the charge transfer effects and the variation in AuMPC symmetry (it has been reported that distortions in the AuMPC structure decrease the HOMO−LUMO gap22). In fact, replacing SCH3 by GSH distorts the AuMPC, which adopts a C1 symmetry instead of the original Ci symmetry of Au25(SCH3)18− (Figure 3). Most importantly, the change from SCH3 to GSH reduces the HOMO−LUMO gap by 31% (0.93 and 0.65 eV for methanethiolate or glutathione, respectively), i.e., the resulting value is now under-estimating the experimental reference by a factor of 2. It is well-known that hybrid functionals such as PBE0 give more accurate results (larger values) for electronic band gaps than GGA functionals. Correspondingly, the PBE0 value of Au25(GSH)18− (1.54 eV) is closer to the measurement once the ligand is the same as in experiments.25 Therefore, the previously reported poor performance of PBE0 for HOMO− LUMO gaps (severe overestimation21) is related to the limited description of the simulated system itself (absence of solvent, different ligand). For example, our PBE0 calculation gives a value of 2.20 eV for Au25(SCH3)18− in the gas phase, which is much larger, highlighting the importance of including the GSH ligands and the solvent in the model.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.5b01382. Detailed description of the computational approach, S− Au(a′) bond distances as a function of time, radial distribution functions for the nanocluster and water solvent, visualizations of molecular orbitals, Hirshfeld charges for Au and S atoms, and symmetry analysis (PDF) Coordinates of the cluster geometries (XYZ) 3863


Letter


■

(15) Dainese, T.; Antonello, S.; Gascon, J. A.; Pan, F.; Perera, N. V.; Ruzzi, M.; Venzo, A.; Zoleo, A.; Rissanen, K.; Maran, F. Au25(SEt)18, a Nearly Naked Thiolate-protected Au25 Cluster: Structural Analysis by Single Crystal X-ray Crystallography and Electron Nuclear Double Resonance. ACS Nano 2014, 8, 3904−3912. (16) Ni, T. W.; Tofanelli, M. A.; Phillips, B. D.; Ackerson, C. J. Structural Basis for Ligand Exchange on Au25(SR)18. Inorg. Chem. 2014, 53, 6500−6502. (17) Cao, T.; Jin, S.; Wang, S.; Zhang, D.; Meng, X.; Zhu, M. A Comparison of the Chiral Counterion, Solvent, and Ligand Used to Induce a Chiroptical Response from Au25- Nanoclusters. Nanoscale 2013, 5, 7589−7595. (18) Negishi, Y.; Nobusada, K.; Tsukuda, T. Glutathione-Protected Gold Clusters Revisited: Bridging the Gap Between Gold(I)-Thiolate Complexes and Thiolate-Protected Gold Nanocrystals. J. Am. Chem. Soc. 2005, 127, 5261−5270. (19) Shichibu, Y.; Negishi, Y.; Tsunoyama, H.; Kanehara, M.; Teranishi, T.; Tsukuda, T. Extremely High Stability of Glutathionateprotected Au25 Clusters Against Core Etching. Small 2007, 3, 835− 839. (20) Perdew, J. P. Density Functional Theory and the Band Gap Problem. Int. J. Quantum Chem. 1985, 28, 497−523. (21) Muniz-Miranda, F.; Menziani, M. C.; Pedone, A. Assessment of Exchange-Correlation Functionals in Reproducing the Structure and Optical Gap of Organic-Protected Gold Nanoclusters. J. Phys. Chem. C 2014, 118, 7532−7544. (22) Tlahuice-Flores, A.; Whetten, R. L.; Jose-Yacaman, M. Ligand Effects on the Structure and the Electronic Optical Properties of Anionic Au25(SR)18 Clusters. J. Phys. Chem. C 2013, 117, 20867− 20875. (23) Aikens, C. M. Origin of Discrete Optical Absorption Spectra of M25(SH)18− Nanoparticles (M = Au, Ag). J. Phys. Chem. C 2008, 112, 19797−19800. (24) Goh, J.-Q.; Malola, S.; Häkkinen, H.; Akola, J. Role of the Central Gold Atom in Ligand-Protected Biicosahedral Au24 and Au25 Clusters. J. Phys. Chem. C 2013, 117, 22079−22086. (25) Dufour, F.; Fresch, B.; Durupthy, O.; Chaneac, C.; Remacle, F. Ligand and Solvation Effects on the Structural and Electronic Properties of Small Gold Clusters. J. Phys. Chem. C 2014, 118, 4362−4376. (26) Tlahuice-Flores, A. Zwitterion l-cysteine adsorbed on the Au20 cluster: enhancement of infrared active normal modes. J. Mol. Model. 2013, 19, 1937−1942. (27) Link, S.; Beeby, A.; FitzGerald, S.; El-Sayed, M. A.; Schaaff, T. G.; Whetten, R. L. Visible to Infrared Luminescence from a 28-Atom Gold Cluster. J. Phys. Chem. B 2002, 106, 3410−3415. (28) Humphrey, W.; Dalke, A.; Schulten, K. VMD: visual molecular dynamics. J. Mol. Graphics 1996, 14, 33−38. (29) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (30) CPMD; Copyright IBM Corp. 1990−2001, Copyright MPI für Festkörperforscung, Stuttgart, 1997−2004. (31) Laio, A.; VandeVondele, J.; Rothlisberger, U. A Hamiltonian Electrostatic Coupling Scheme for Hybrid Car−Parrinello Molecular Dynamics Simulations. J. Chem. Phys. 2002, 116, 6941−6947. (32) Wu, Z.; Gayathri, C.; Gil, R. R.; Jin, R. Probing the Structure and Charge State of Glutathione-Capped Au25(SG)18 Clusters by NMR and Mass Spectrometry. J. Am. Chem. Soc. 2009, 131, 6535− 6542. (33) Kacprzak, K. A.; Akola, J.; Häkkinen, H. First-principles Simulations of Hydrogen Peroxide Formation Catalyzed by Small Neutral Gold Clusters. Phys. Chem. Chem. Phys. 2009, 11, 6359−6364. (34) Vargas, A.; Santarossa, G.; Iannuzzi, M.; Baiker, A. Fluxionality of Gold Nanoparticles Investigated by Born-Oppenheimer Molecular Dynamics. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 80, 195421. (35) Hirshfeld, F. L. Bonded-atom Fragments for Describing Molecular Charge Densities. Theoret. Chim. Acta 1977, 44, 129−138. (36) Yang, H.; Wang, Y.; Huang, H.; Gell, L.; Lehtovaara, L.; Malola, S.; Häkkinen, H.; Zheng, N. All-thiol-stabilized Ag44 and Au12Ag32

Coordinates of the cluster geometries (XYZ)

AUTHOR INFORMATION

Corresponding Author

*E-mail: jaakko.akola@tut.fi. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the computer support, technical expertise, and assistance provided by the CSC − IT Centre for Science (Espoo, Finland) and Barcelona Supercomputing CenterCentro Nacional de Supercomputación (BSC-CNS). Financial support was provided by the Academy of Finland through its Centres of Excellence Program (Project 251748) (J.A.), the Generalitat de Catalunya (Grant 2014SGR-987 (C.R.) and Ministerio de Economiá y Competitividad (MINECO) (Grant CTQ2014-55174-P) (C.R.). V.R. acknowledges a predoctoral FPU fellowship from MINECO (AP2009-3024). We thank Mauro Boero for insightful discussions regarding gold pseudopotentials.

■

REFERENCES

(1) Ackerson, C. J.; Sykes, M. T.; Kornberg, R. D. Defined DNA/ Nanoparticle Conjugates. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 13383−113385. (2) Ackerson, C. J.; Jadzinsky, P. D.; Jensen, G. J.; Kornberg, R. D. Rigid, Specific, and Discrete Gold Nanoparticle/Antibody Conjugates. J. Am. Chem. Soc. 2006, 128, 2635−2640. (3) Zheng, M.; Huang, X. Nanoparticles Comprising a Mixed Monolayer for Specific Bindings with Biomolecules. J. Am. Chem. Soc. 2004, 126, 12047−12054. (4) Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Bushnell, D. A.; Kornberg, R. D. Structure of a Thiol Monolayer-Protected Gold Nanoparticle at 1.1 Å Resolution. Science 2007, 318, 430−433. (5) Qian, H.; Eckenhoff, W. T.; Zhu, Y.; Pintauer, T.; Jin, R. Total Structure Determination of Thiolate-Protected Au38 Nanoparticles. J. Am. Chem. Soc. 2010, 132, 8280−8281. (6) Akola, J.; Walter, M.; Whetten, R. L.; Häkkinen, H.; Gronbeck, H. On the Structure of Thiolate-Protected Au25. J. Am. Chem. Soc. 2008, 130, 3756−3757. (7) Heaven, M. W.; Dass, A.; White, P. S.; Holt, K. M.; Murray, R. W. Crystal Structure of the Gold Nanoparticle [N(C8H17)4][Au25(SCH2CH2Ph)18]. J. Am. Chem. Soc. 2008, 130, 3754−3755. (8) Shichibu, Y.; Negishi, Y.; Tsukuda, T.; Teranishi, T. Large-scale Synthesis of Thiolated Au25 Clusters via Ligand Exchange Reactions of Phosphine-stabilized Au11 Clusters. J. Am. Chem. Soc. 2005, 127, 13464−13465. (9) Zeng, C.; Liu, C.; Chen, Y.; Rosi, N. L.; Jin, R. Gold−Thiolate Ring as a Protecting Motif in the Au20(SR)16 Nanocluster and Implications. J. Am. Chem. Soc. 2014, 136, 11922−11925. (10) Das, A.; Liu, C.; Byun, H. Y.; Nobusada, K.; Zhao, S.; Rosi, N.; Jin, R. Structure Determination of [Au18(SR)14]. Angew. Chem., Int. Ed. 2015, 54, 3140−3144. (11) Tlahuice, A.; Garzon, I. L. On the structure of the Au18(SR)14 cluster. Phys. Chem. Chem. Phys. 2012, 14, 3737−3740. (12) Tlahuice-Flores, A.; Jose-Yacaman, M.; Whetten, R. L. On the structure of the thiolated Au15 cluster. Phys. Chem. Chem. Phys. 2013, 15, 19557−19560. (13) Yu, Y.; Luo, Z.; Chevrier, D. M.; Leong, D. T.; Zhang, P.; Jiang, D.-e.; Xie, J. Identification of a Highly Luminescent Au22(SG)18 Nanocluster. J. Am. Chem. Soc. 2014, 136, 1246−1249. (14) Pei, Y.; Tang, J.; Tang, X.; Huang, Y.; Zeng, X. C. New Structure Model of Au22(SR)18: Bitetrahederon Golden Kernel Enclosed by [Au6(SR)6] Au(I) Complex. J. Phys. Chem. Lett. 2015, 6, 1390−1395. 3864


Letter

The Journal of Physical Chemistry Letters Nanoparticles with Single-crystal Structures. Nat. Commun. 2013, 4, 2422. (37) Jiang, D.-e.; Overbury, S. H.; Dai, S. Structure of Au15(SR)13 and Its Implication for the Origin of the Nucleus in Thiolated Gold Nanoclusters. J. Am. Chem. Soc. 2013, 135, 8786−8789. (38) Mustalahti, S.; Myllyperkiö, P.; Malola, S.; Lahtinen, T.; Salorinne, K.; Koivisto, J.; Häkkinen, H.; Pettersson, M. Molecule-like Photodynamics of Au102(pMBA)44 Nanocluster. ACS Nano 2015, 9, 2328−2335. (39) Walter, M.; Akola, J.; Lopez-Acevedo, O.; Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Whetten, R. L.; Gronbeck, H.; Häkkinen, H. A Unified View of Ligand-Protected Gold Clusters as Superatom Complexes. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 9157−9162.

3865


How do Water Solvent and Glutathione Ligands Affect the Structure

Recommend Documents