Transformative synthesis of semiconductor nanoclusters with atomic precision via metal-ligand coordination exchange

Cation exchange reactions with atomically defined precursors and coordination complexes address the challenge of synthesizing semiconductor nanocrystals with homogeneous structures, achieving precise control over sizes and compositions, leading to improved nanocluster properties.

WO2024243127A1PCT designated stage expired Publication Date: 2024-11-28UNIV OF FLORIDA RESEARCH FOUNDATION INC
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/US2024/030202
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-07-24
Filing Date
2024-05-20
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Current methods for synthesizing semiconductor nanocrystals, or quantum dots, face challenges in achieving homogeneous sizes and structures, which is crucial for enhancing device performance and understanding structure-property correlations, due to complexities in precursor conversion, nucleation, and surface coordination, especially for binary compound semiconductors like CdSe.

Method used

The method involves cation exchange reactions using atomically defined precursors and coordination complexes, allowing for precise control of cation and surface ligand exchange, resulting in atomically precise semiconductor nanoclusters with tunable sizes, shapes, and compositions, such as the transformation of Cu26Se13(PEt2Ph)14 to Cd26Se17I18(PPr3)10, preserving the anion lattice and introducing novel core and surface structures.

Benefits of technology

This approach enables the synthesis of nanoclusters with near-unity conversion yield and precise control over sizes, shapes, and surface ligands, overcoming previous challenges in achieving atomic precision and stability, facilitating the production of nanoclusters with enhanced optical and electronic properties.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024030202_28112024_PF_FP_ABST
    Figure US2024030202_28112024_PF_FP_ABST
Patent Text Reader

Abstract

Precision nanosynthesis is crucial for atomic engineering in nanostructures and mechanistic understanding of nanoscale reactions. But it is a daunting task due to emerging complexities in nanostructures and nanosynthesis. Here, we report the precision synthesis of semiconductor nanocluster via synergizing coordination, cluster, and colloidal chemistry at the nanoscale. In particular, by employing Cu26Se13(PEt2Ph)14 cluster and Cdl2(PPr3)2 coordination complex as precise precursors in colloidal cation exchange reaction, an atomically precise Cd26Se17I18(PPr3)10 cluster is produced at near- unity yield. X-ray crystallography reveals that the child CdSe cluster inherits its icosahedral-packed anion lattice and halide-phosphine-rich surface from parent cluster and complex respectively and further evolves into an enlarged structure. The hybridization of achiral precursors leads to chiral CdSe clusters co-crystallized as a 1:1 mixture of enantiomers. In-situ optical spectroscopy unveils that the reaction proceeds with well-defined pathways. Based on charge and coordination conservations, the structure of child cluster can be atomically linked to precursors via a series of transformations including surface coordination exchange, intra-cluster reorganization, and inter-cluster digestive ripening. The precision nanosynthesis with atomically defined precursors, pathways, and products is expected to further rationalize and diversify nanochemistry.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Attorney Ref. No.10457-559PC0 Transformative synthesis of semiconductor nanoclusters with atomic precision via metal-ligand coordination exchange BACKGROUND Semiconductor nanocrystals, also known as quantum dots (QDs), are important building blocks for energy and information technologies1,2. Despite significant advancements in the past three decades3–5, a major challenge in QD synthesis is their polydisperse sizes and inhomogeneous structures (especially at the surface)6. Synthesizing QDs with homogeneous sizes and structures is important to further enhance the performance of QD-based devices, such as consistent emissions in light-emitting diodes, coherent charge transport in field-effect transistors, and uniform fields for quantum computing.1,2,7It is also essential for the fundamental understanding of structure-property correlations, such as size- and surface-dependent absorption and emission properties6,8. Atomically precise nanoclusters are ultrasmall nanocrystals (~1 nm to ~3 nm) with identical core, interface, and surface structures9. These clusters can self-assemble into single crystals with translational, orientational, and rotational orders10. Thus, the total structure of clusters (core and surface) and their assembly can be determined by single-crystal X-ray diffraction (sc- XRD) at atomic resolution. While the past decades have seen significant advances in atomically precise metal clusters9–12, the progress in achieving atomic precision in semiconductor clusters is limited. Compared to metals, binary compound semiconductors contain both cations and anions (e.g., Cd2+and Se2–), which leads to more complex precursor conversion13, surface coordination14, and charge balancing15,16issues. In addition, the lability of surface ligands and ionic bonding of the core often result in aggregation or transformation of clusters during the isolation or characterization processes17. Due to these challenges, only a few structures of semiconductor clusters have been reported, including early works on chalcogenolate-protected CdS and CdSe clusters18–21and recent reports on carboxylate-protected clusters22,23. In addition to atomic structures, achieving atomic precision in chemical synthesis of nanomaterials is another long pursuing goal in nanochemistry. Akin to molecular synthesis, precision nanosynthesis is crucial for rational constructing of nanostructures, precise tuning of properties, and mechanistic understanding of reactions24. However, extending precision synthesis Attorney Ref. No.10457-559PC0 from molecular scale to nanoscale has been a formidable challenge due to emerging complexity in chemical reactions of nanomaterials25. Compared to a single or a few collision events in a molecular reaction, constructing a nanostructure requires numerous collisions that occur hierarchically at the atomic core, coordination interface, organic surface, and inter-particle levels6,14,26,27. This leads to a broad energy landscape spanning from ~1000 kJ / mol (covalent bonds) to ~1 kJ / mol (intermolecular interactions) as well as entangled kinetics including precursor conversion, nucleation-growth, and ripening. BRIEF DESCRIPTION OF DRAWINGS Fig. 1 | Cation exchange reactions across the length scale. a, Reaction of bulk crystal. b, Reaction of nanocrystals with typical conservations in shapes and anion lattices. c, Atomically precise reaction of nanocluster demonstrated in this work, which unveils the cation exchange is accompanied by surface ligand exchange. Yellow / blue / green spheres are anions (E) / original cations (M) / incoming cations (M’); blue / green shades are original ligands (L) / incoming ligands (L’), which includes negatively charged X-type ligands and neutral L-type donors. Fig. 2 | Precise precursors for cation exchange reaction and corresponding characterization of precursor and product clusters. a, Cu26Se13(PEt2Ph)14cluster (center) and CdI2(PPr3)2 coordination complexes (periphery) as precise reaction precursors. The structures are determined by X-ray crystallography. b, Shell-by-shell anatomy of Cu26Se13P14 framework viewed along the pseudo C3axis (top row) and C2axis (bottom row). Superscripts o, i, and ii refer to the core, facet-capping, and edge-bridging Cu+cations respectively. c to g, Crystals and reaction solution, UV-vis, powder XRD,1H NMR, and31P NMR characterization of Cu26Se13(PEt2Ph)14 cluster, respectively. h to l, characterization of the exchanged cluster product. The red spectral lines in f and g are obtained at −40oC (* indicates toluene-d8and CDCl3solvent peaks). Fig. 3 | Cd26Se17I18(PPr3)10 as the precise product of cation exchange. a, Overall structure. b, Shell-by-shell anatomy of Cd26Se17 core viewed from the C3 axis (top row) and the C2 axis (bottom row). Superscripts o, i, ii, and iii refer to the core, facet-capping, edge-bridging, and vertex-terminating Cd respectively. c, Surface ligand coordination at facet-, edge-, corner-, and all Cd sites, respectively (side view). d-e, Chiral structures of left (green) and right (red) enantiomers viewed from top and base, respectively. f, Inter-cluster packing structures of enantiomers in the crystal. Attorney Ref. No.10457-559PC0 Fig. 4 | Transformation pathway of cation exchange reaction. a, Overall reaction from Cu26Se13(PEt2Ph)14 to Cd26Se17(PPr3)10. b to d, Reaction process monitored by photos and in-situ UV-vis absorption spectroscopy. e, Four stages of the transformation and corresponding reaction intermediates derived based on X-ray structures of precursor and product clusters (viewed from tetrahedron base) following charge and coordination conservations. Fig. 5 | NMR characterization of Cu26Se13(PEt2Ph)14 clusters. a,31P{H} NMR of free PEt2Ph ligand at room temperature (top) and Cu26Se13(PEt2Ph)14cluster at variable temperatures (bottom). Seven peaks of the cluster can be resolved at low-temperature spectrum, matching with seven chemically inequivalent phosphorous (P1 to P7) in the crystal structure (left). b, Corresponding1H NMR spectra. c,1H-1H COSY spectrum of the cluster measured at −40oC (left) and 1D spectrum Seven groups of correlated protons in the phenyl group of the PEt2Ph ligands can be identified. The sample is measured in toluene-d8. Solvent peaks at 7.09, 6.97-7.01, and 2.08 ppm are labeled by *. Fig. 6 | X-ray crystallographic structure of CdI2(PPr3)2. a, Unit cell structure. b, Tetrahedron coordination of Cd(II) center in individual complex. c, Bond lengths and angles around Cd(II) center. Fig. 7 | MALDI-MS characterization of precursor and product clusters. a, Spectrum of Cu26Se13(PEt2Ph)14cluster and corresponding assignment of fragmentation peaks. All PEt2Ph ligands dissociate during the ionization process. The most intense peak is assigned to a dimeric (Cu26Se13)2Cu fragment. b, Spectrum of Cd26Se17I18(PPr3)10 cluster and corresponding assignment of fragmentation peaks. The cluster framework dissociates during the ionization process and reorganizes into characteristic fragments of CdSe materials. Fig. 8 | Overview of X-ray crystallographic structure of the Cd26Se17I18(PPr3)10 cluster. a, Cd26Se17 core, Cd26Se17I18P10 inorganic framework, and Cd26Se17I18(PPr3)10 cluster viewed from three orientations. The corresponding sizes are measured based on the atom-to-atom distance along the tetrahedron edge. b, Packing and coordination of Se17 anion lattice (top) and Cd26 cation lattice (bottom). The anion lattice features a central icosahedron capped by four apical Se, and the cation lattice can be viewed as four interpenetrating icosahedra. All Se and Cd are tetrahedrally coordinated except for the four apical Se (light yellow). Green: Cd; orange: Se; pink: P; purple: I; grey: C; and white: H. Attorney Ref. No.10457-559PC0 Fig. 9 | NMR characterization of Cd26Se17I18(PPr3)10 cluster. a,31P{1H} NMR spectra of free PPr3 ligand, CdI2(PPr3)2 complex, and Cd26Se17I18(PPr3)10 cluster. A 3:3:3:1 splitting of 10 phosphines is observed for the cluster sample. The satellite peaks observed in both complex and cluster are due to Cd-P couplings. b, Corresponding1H NMR spectra. c, Amplified1H NMR spectrum of the cluster at aliphatic region. The four inequivalent are indicated by numbers (P1 to P4), which matches with the C3 symmetry of the structure (left). The sample is measured in CDCl3at room temperature. The solvent peaks are labeled with *. Fig.10 | The overall transformation reaction. a, the reaction between Cu26Se13(PEt2Ph)14 cluster and CdI2(PPr3)2 complex leads to an atomically precise Cd26Se17I18(PPr3)10 cluster. b, Shell- by-shell structural comparison of the precursors (top) with the product cluster (bottom). c, Comparison of Se-Se distances of Se13icosahedral frameworks in Cu26Se13(PEt2Ph)14(left) and Cd26Se17I18(PPr3)10 (right). Fig.11 | The origin of chirality in Cd26Se17I18(PPr3)10 clusters and their chiral packing in the unit cell. a, Adaption of tetrahedral coordination into icosahedral framework induces chirality in the core. Left: icosahedral framework in left Cd26Se17 enantiomer with the Cd-Se bonds leaning to the left against the principal axis. Middle: cuboctahedron in zincblende lattice with vertical alignments of all the Cd-Se bonds. Right: right-leaning Cd-Se bonds in the right Cd26Se17enantiomer. b, Each unit cell contains two left (dark and light green) and two right (dark and light red) enantiomers. Arrow indicates the direction of the principal C3 axis. The corresponding orientations of enantiomers are shown in the tetrahedral outlines (right). Solvent molecules (toluene) in the unit cell are omitted for clarity. Fig.12 | Symmetry breaking induced by surface ligands. The uneven distribution of six phosphines (P atom highlighted in yellow) to four corners of the tetrahedron leads to the symmetry breaking from T to C3, with three PPr3concentrated on the top corner and one PPr3on each of the base corners of the tetrahedron. Fig.13 | Cation exchange reactions with different CdI2(PR3)2 (R = octyl, butyl, propyl, and ethyl) complexes. The same 365 nm (dark blue curves) and 355 nm (light blue curves) intermediates are observed when different phosphines are used in the complex. The 355 nm → 372 nm conversion is slower when bulkier phosphine is used. Reaction with CdI2(PEt3)2leads to aggregation at the end. CdI2(PPr3)2 shows the best results in balancing reaction rates and Attorney Ref. No.10457-559PC0 preventing aggregations. In-situ UV-vis spectra are monitored every 10 minutes in the first hour (left) and every 2 hours for the rest of the reaction (right). Fig. 14 | Cation exchange reaction with different CdX2(PPr3)2 (X = I−, Br−, Cl−) complexes. Compared to the well-defined 365 nm → 355 nm → 372 nm transitions in CdI2(PPr3)2, the cation exchange reaction using CdBr2(PPr3)2results in the aggregation of larger and less- defined clusters after around one hour. The aggregation is even faster (within a few minutes) in the case of CdCl2(PPr3)2. Fig. 15. Transformative synthesis of atomically precise II-VI semiconductor nanoclusters (II = Zn, Cd, Hg; VI = Se). (A) Precise precursors: Cu26Se13(PEt2Ph)14cluster (center) as anion template and three MI2(tmeda) complexes (periphery) as cation carriers. (B) Corresponding reaction schemes. (C-E) Absorption spectra of the product clusters. Dashed line: reaction mixture. Solid line: solution of single crystals in CH3OH for Zn14Se13, toluene-tmeda for Cd14Se13, and toluene-tmeda for Hg14Se13, respectively. Inset: photos of reaction mixtures and crystals. Fig. 16. Atomic structures of parent Cu26Se13(PEt2Ph)14 cluster and child [Zn14Se13(tmeda)6]2+(1), Cd14Se13(tmeda)6I2 (2), Hg14Se13(tmeda)6I2 (3) clusters. (A) Overall crystallographic structures. (B) Se13 icosahedral anion lattices. (C) Shell-by-shell assembly of cation and anion polyhedrons. (D) Arrangements of core cationscM inside Se13shell. (E) Positions of saturated face-capping cationsfM connected to surface halides and unsaturated face-capping cationsf’M without ligands. (F) Arrangement of edge-bridging cationseM at three orthogonal C2 axes of icosahedron. The size difference of II and VI ions is illustrated in the color legend: Se2–(198 pm), Cu+(60 pm), Zn2+(60 pm), Cd2+(78 pm), and Hg2+(96 pm) (71). Fig. 17. Structural variations of II14VI13 clusters under different synthesis and crystallization conditions. (A-C) Crystal structures of Zn14Se13(tmeda)6Cl2 (4), [Cd14Se13(tmeda)6Cl2]n(5), and [Hg14Se13(tmeda)6Cl2]n(6). The bottom panels highlight the coordination environment of eight dynamic cations. (D-F) Comparison of absorption spectra of six II14VI13 clusters. The Zn14Se13 clusters were measured in CH3OH. The Cd14Se13 and Hg14Se13 clusters were measured in CH2Cl2with the presence of tmeda. Fig. 18. Optical properties of II14VI13 clusters. (A) Absorption spectra of clusters 1-3 plotted in energy scale. (B) Plots of first (E1) and second (E2) exciton transition energy of M14Se13 Attorney Ref. No.10457-559PC0 clusters against corresponding bulk band gap (Eg), with linear fittings of E1= 0.655 Eg+ 2.411 and E2 = 0.687 Eg + 2.619. Fig. 19.1H NMR of free tmeda ligand and three MI2(tmeda) complexes (M = Zn, Cd, Hg) in CDCl3. The satellite peaks in CdI2(tmeda) corresponding to the coupling between1H and113Cd (J = 90 Hz). Fig. 20. Transformative synthesis of II-VI nanoclusters from Cu26Se13 template cluster and MI2(tmeda) cation carriers (M = Zn, Cd, Hg). (A) Reaction with ZnI2(tmeda). (B) Reaction with CdI2(tmeda) using excess PPr3 to increase its solubility in toluene. (C) Reaction with HgI2 and excess tmeda. The initial color changes of the reaction mixtures are shown in the photos (left). The reaction mixture was then diluted and monitored by in-situ UV-Vis spectroscopy (right). The in-situ spectra were recorded every 10 minutes in the first two hours and every one hour for the rest of the time. Fig. 21. Crystal structures of transformed II-VI clusters and their packing in unit cells. (A) [Zn14Se13(tmeda)6]2+cluster co-crystallized [BPh4]‒counterions and CH2Cl2solvent molecules. Each unit cell contains two whole clusters and four counterions with balanced charges. The counterions and solvent molecules are sandwiched between two layers of clusters. (B) Cd14Se13(tmeda)6I2cluster co-crystallized with one CdI2(tmeda) complex and several toluene molecules. (C) Hg14Se13(tmeda)6I2 cluster. Top: individual cluster; middle: top view of unit cell; bottom: side view of unit cell. The cations are drawn in space-filling mode to highlight the orientation of clusters in unit cells. Fig. 22 Comparison of Se13 anion lattices of parent and child clusters. Se-Se distances in Se13 icosahedral frameworks are compared among the parent Cu26Se13(PEt2Ph)14 cluster and three child clusters: [Zn14Se13(tmeda)6]2+, Cd14Se13(tmeda)6I2, and Hg14Se13(tmeda)6I2 (unit: Å). Top: peripheral Se-Se distances; middle: radial Se-Se distances; bottom: tabulated comparison with corresponding Se-Se distances in bulk lattices. Fig. 23. Synthesis and crystal structure of Zn14Se13(tmeda)6Cl2 cluster (4). (A) Reaction scheme. (B) Corresponding UV-Vis spectra. The crystals were dissolved in CH3OH for measurement. The E2 peak was slightly affected by the absorption of CH2Cl2 (measured in CH3OH). Inset: crystal blocks with nearly square shape. (C) Crystal structure of individual cluster. (D) Unit cell structure of the cluster with surrounding CH2Cl2solvent molecules. Left: top view along c axis; right: side view from bc plane. Attorney Ref. No.10457-559PC0 Fig. 24. Synthesis and crystal structure of [Cd14Se13(tmeda)6Cl2]n cluster (5). (A) Reaction scheme. (B) Corresponding UV-Vis spectra. The crystals were dissolved in CH2Cl2 with the presence of tmeda. The two peaks are at 354 and 328 nm in both spectra. Inset: Crystals with the shapes of triangular plates. (C) Crystal structure of the individual cluster, featuring four bridging chlorides. Each chloride is shared by two clusters, giving the overall composition of Cd14Se13(tmeda)6Cl2. (D) Unit cell structure of the cluster viewed along the body diagonal of the cube (left) and along a axis (right). Inter-cluster connection is built through Cl‒tetrahedrally, forming the extended 3D network. The chlorides are shown in space-filling mode to highlight the connection pattern. Fig. 25. Synthesis and crystal structure of [Hg14Se13(tmeda)6Cl2]n cluster (6). (A) Reaction scheme. (B) Corresponding UV-Vis spectra. The crystals were dissolved in CH2Cl2with the assistance of tmeda. Inset: Yellow crystals as thin rods. (C) Crystal structure of the individual cluster. (D) Unit cell structures of the cluster viewed from ab plane (left) and ac plane with only one chain displayed (right). Each cluster features one terminal chloride and two bridging chlorides. Each bridging chloride is shared by two clusters, giving the overall composition of Hg14Se13(tmeda)6Cl2 and inter-connected polymeric structure along the a axis. The chlorides are shown in space-filling mode to highlight the connection pattern. Fig. 26. Synthesis and crystal structure of [Zn14Se13(tmeda)6(CH3CN)]2+cluster (7) crystallized from CH3CN. (A) Reaction scheme. (B) Corresponding UV-Vis spectra. After precipitation, the crude cluster was redissolved in CH3CN for crystallization. The E2peak was better resolved when using CH3OH as solvent, showing two peaks at 288 nm and 270 nm. The obtained crystals were also dissolved in CH3OH. Inset: crystals as rhombic blocks. (C) The crystal structure of the cluster is co-crystallized with two equivalent of [BPh4]‒ions and one CH3CN solvent molecule. (D) Unit cell structure of the cluster viewed along c axis (left) and from bc plane (right). Fig. 27. Synthesis and crystal structure of Cd14Se13(tmeda)6I2 cluster (8) crystallized from CH2Cl2. (A) Reaction scheme. (B) Corresponding UV-Vis spectra. The crystal solution was made in CH2Cl2 with the presence of tmeda. The two peaks are at 355 and 329 nm in both spectra. Inset: rhombic shaped crystals. (C) The crystal structure of the product as a mixture of two cluster components with different occupancy factors: 75% of [Cd14Se13(tmeda)6]2+and 25% of Attorney Ref. No.10457-559PC0 [Cd14Se13(tmeda)6I8]6‒. (D) Unit cell structure of the cluster viewed along c axis (left) and from ac plane (right). Fig. 28. Structural variations of Cd14Se13 and Zn14Se13 clusters induced by solvents. (A) Top: overall crystal structures of [Zn14Se13(tmeda)6]2+and [Zn14Se13(tmeda)6(CH3CN)]2+; bottom: the arrangement of core and / or face-capping cations. (B) Comparison of UV-Vis spectra for three Zn14Se13 clusters measured on crystal solutions in CH3OH. (C) The corresponding structural analysis for [Cd14Se13(tmeda)6]2+and [Cd14Se13(tmeda)6I8]6–. (D) The comparison of UV-Vis spectra for three Cd14Se13 clusters measured on crystal solutions CH2Cl2 with the presence of tmeda. Fig.29. The comparison of the linear fittings of E1energy to bulk band gap of II-VI clusters in literature and this work. The E1peak positions of (CdS)13(CdTe)13, (CdSe)13, (ZnTe)13, and (ZnSe)13 are extracted from ref. (13, 14). The trendlines share similar slopes: 0.655 (black) and 0.679 (orange). DETAILED DESCRIPTION Presented herein embodiments related to precision synthesis of semiconductor nanoclusters involving atomically defined precursors, pathways and products via synergizing colloidal, cluster, and coordination chemistry in the reaction. Typically, cation exchange is selected as a model reaction, inspired by its recent advances in the synthesis of colloidal ionic nanocrystals28–31as well as previous success in the transformative synthesis of atomically precise metal clusters32,33. Compared to de novo synthesis with entangled processes, cation exchange provides a facile pathway for compound nanostructures by fixing the anion framework (E) in the template (ME) and exchanging cations (M to M’, Fig. 1a), as shown by early work of solid-state transformation of bulk single crystals (e.g., CdS → Cu2S)34–36. Since its introduction to nanocrystal synthesis (e.g., Ag2Se ↔ CdSe)37,38, cation exchange has become a versatile method to access nanocrystals with novel shapes, phases, and heterostructures39–46. The rigidity of the anion lattices leads to the typical preservation of sizes, shapes, and phases of nanocrystals after transformation (Fig. 1b)39–46. Despite the advances, atomic insights into the reaction mechanism, especially surface reaction, were still lacking28. Extending cation exchange to clusters has been challenging Attorney Ref. No.10457-559PC0 due to the difficulties to characterize ultrasmall clusters using conventional methods such as powder XRD and electron microscope47,48. To realize atomic precision in cation exchange reaction (e.g., Cu2Se → CdSe), atomically defined clusters and coordination complexes were developed as precise precursors for the host anion template (Cu2Se) and incoming cation carrier (Cd2+), respectively (Fig. 1c). The reaction successfully produces an atomically precise CdSe cluster with near-unity conversion yield. The crystallographic structures of precursors and product provide atomic evidence that the anion lattice of the ultrasmall cluster can be preserved during the conversion and unveil that cation exchange is accompanied by surface ligand exchange (Fig. 1c). The precise conversion enables the synthesis of novel CdSe cluster with chiral core (icosahedron-based tetrahedron) and polar surface (halide and phosphine protection). Together with in-situ absorption spectra of the reaction, the precise reaction pathway can be mapped to provide atomic insights into the mechanism of cation exchange and subsequent conversion processes. Based on the teachings herein, embodiments of the invention may synthesize atomically precise nanoclusters with controllable sizes, shapes, compositions, surface ligands, dopants, and heterostructures. In certain embodiments, the methods provided herein can provide predictable sizes of nanoclusters as reaction products, typically ranging from 1 nm to 3 nm. For example, 1.1 nm Cd26Se17(demonstrated). The sizes of product clusters can be tuned by using different templating clusters. Because of its rigidity, the anion framework in the templating cluster can be preserved during cation exchange. Therefore, the core size of the product clusters can be controlled by the choice of templating clusters. Other than Cu26Se13(PEt2Ph)14cluster, the reported collection of Cu2Seclusters provide variable sources of anion frameworks. For example, Cu44Se22(PEt2Ph)18 cluster has an ellipsoidal shape with a length of ~1.1 nm. Another cluster, Cu70Se35(PEt2Ph)24, is a triangular plate with an edge of ~1.6 nm and a thickness of ~0.7 nm. In this way, semiconductor nanoclusters can be synthesized with predictable core sizes. Moreover, shapes of the synthesized nanocluster reaction product can be controlled and include tetrahedrons (demonstrated in Cd26Se17), spheres, and triangular plates. Similar to the sizes, the shapes of the nanocluster reaction product can be adjusted by the shapes of the templating clusters. Attorney Ref. No.10457-559PC0 Cations used in the transformation reactions may include Cd2+, Zn2+, Hg2+, Pb2+and other transition metals to access various metal chalcogenide nanoclusters. Atomically precise nanoclusters with different core compositions can be obtained using the corresponding metal complexes. With the presence of a soft Lewis base such as phosphine, Cu(I) ions in the template cluster can be extracted and replaced by incoming cations to form the new compositions. In addition to Cd(II) ions, other metal precursors such as Zn(II), Hg(II), Pb(II), and various transition metal complexes could also replace the original Cu(I) cations and lead to ZnSe, HgSe, PbSe, and other metal selenide nanoclusters, respectively. By exchanging with various cations, the core composition can be further diversified to show various properties. Anions utilized in the templating clusters may include S2–and Te2–,other than Se2–. As is described further herein, the templating clusters may also comprise one or more surface ligands, and typically the coordination complex will comprise different surface ligands and carry them into the product clusters. Examples of suitable surface ligands include but are not limited to, halides, phosphines, amines, carboxylates, carbenes, and the like. Surface modification is important for tuning the chemical (stability and solubility), optical (absorption and emission), electronic (charge transport), and catalytical properties of semiconductor nanoclusters. Therefore, controlling the surface structure is critical for applications in LEDs, solar cells, photocatalysis, etc. As demonstrated herein for the transformation of Cu26Se13(PEt2Ph3)14 cluster into a Cd26Se17I18(PPr3)10 cluster via cation exchange with a CdI2(PPr3)2 coordination complex, the capping ligands of the product cluster are inclusively inherited from the incoming metal complex. Thus, the surface passivating ligands of nanocluster can be tailored by altering ligands in the metal complex. For instance, tetramethylethylenediamine can be introduced as a replacement for the neutral phosphine ligand. As for the anionic ligand, carboxylate could take the position of iodide to balance the charge as well as protect the core. The templating clusters may also include one or more dopants, such as, but not limited to Mn2+. Doping with extraneous cations can also be achieved by cation exchange to tune the properties of semiconductor nanoclusters. Although doping has been widely utilized in conventional quantum dots, it is challenging to control and characterize the precise concentrations and positions of dopants. Cation exchange provides an efficient pathway to achieve homogeneous doping, therefore enabling the precise control of dopant concentrations and positions. For example, extraneous metal complex Mn(II) and host metal complex of Cd(II) Attorney Ref. No.10457-559PC0 can be pre-mixed to react with the cluster template. Dopant concentration and reactivity can be adjusted by changing the ratio of cations or the ligands of the complexes. The positions of dopants in the core framework will be determined by the thermodynamically favored arrangements depending on the strength of host cation-ligand interactions. Nanocluster with heterostructures could be obtained as cation exchange reaction products, i.e., core-shell structure (such as CdSe core and ZnSe / ZnS shell). Heterostructures are important for improving the photoluminescent efficiency and the stability of semiconductor clusters. Similar to the synthesis of colloidal heterostructures, precise core / shell clusters can also be realized by sequential cation exchange. For example, Cu2Se cluster will first be converted to CdSe cluster, followed by exchange with Zn(II) complex to replace surface Cd(II). The reaction can be limited on the surface by controlling the amounts and reactivity of the Zn(II) complex. Advantages Cation exchange between clusters and metal complexes is a transformative approach to synthesizing atomically precise semiconductor nanoclusters. As compared to de novo synthesis from molecular precursors, the disclosed method circumvents a series of obstacles in the synthetic system, including the identification of metal and chalcogenide sources, control of reactivities, precursor conversion, nucleation, growth kinetics, size distribution, etc. Cation exchange will be developed into a versatile tool for a library of atomically precise semiconductor nanoclusters with various sizes, shapes, core compositions, surface ligands, and dopants. Potential applications • Photoluminescent emission for light-emitting diodes (e.g., ZnS / CdSe core / shell clusters) • Building blocks for solution-processible solar cells (e.g., PbSe clusters) • Photocatalysis (e.g., CdSe clusters) • Chiral optics According to certain method embodiments, disclosed is a method for precision synthesis of nanoclusters that involves mixing a templating nanocluster comprising a plurality of each of Cat1-An1-L1, wherein Cat1is Cu(I) or Ag(I), An1is selected from S, Se, or Te, and L1is a surface ligand, with a coordination complex comprising a Cat2-X2-L2, wherein Cat2is selected from Cd2+, Zn2+, Pb2+, Hg2+, or other transitional metal cations but, optionally, not Cat1, X2is selected from Attorney Ref. No.10457-559PC0 halides, carboxylates, thiolates, and other anionic ligands, and L2is a neutral ligand, under conditions to form a nanocluster as reaction product that comprises Cat2, An1, X2, and L2. In a specific embodiment, L1is usually a phosphine, including PEt3, PBu3, PEt2Ph, etc. In another specific embodiment, L2is selected from phosphines (PEt3, PPr3, PBu3, P(octyl)3), amines, carbenes, alcohol, or other neutral ligands. In a more specific embodiment L2is PPr3, NR3 (R is various CxHy group, including H), or multidentate ligands (R2N-CxHy-NR2), with a specific multidentate ligand being (CH3)2NCH2CH2N(CH3)2. Typically, X2may be halides (Cl–, Br–, and I–). In an exemplary embodiment of the method, the first reagent is a Cu26Se13(PEt2Ph)14 template nanocluster the second reagent is a CdI2(PPr3)2coordination complex. Using these clusters and coordination complexes in the reaction produces a Cd26Se17I18(PPr3)10product nanocluster as the reaction product. As used herein, the term “nanocluster” refers to a nanomaterial with a diameter less than 5 nanometers, typically about 1 to about 5 nanometers, that includes a core structure comprising at least 2 or 3 elements. The term “coordination complex” refers to a reagent with less than a few metal cations, typically one metal center, a few anionic ligands, and a few neutral ligands. In an exemplary embodiment of the method, the first reagent is a Cu26Se13(PEt2Ph)14template nanocluster, and the second reagent is a CdI2(PPr3)2 coordination complex. Their reaction produces a Cd26Se17I18(PPr3)10 nanocluster as the reaction product. According to another embodiment, provided is a nanocluster comprising Cat2-An1-X2-L2, wherein Cat2is selected from Cd2+, Zn2+, Hg2+, Pb2+or other transitional metal cations, wherein An1is selected from S, Se, or Te, wherein X2is selected from halides (Cl–, Br–, I–), carboxylates (–O2CR, R is various CxHy group), or other anionic ligands, and wherein L2is a neutral surface ligand. In a specific example, L2is selected from a phosphine, amine, carbene, alcohol, or other neutral ligands, wherein L2is, optionally, is PPr3, NR3 (R is various CxHy group, including H), multidentate ligands (R2N-CxHy-NR2). In another specific example, the nanocluster comprises a size of 0.5-5 nm. The nanocluster may optionally include at least one dopant such as Mn2+. In a specific embodiment, the nanocluster is Cd26Se17I18(PPr3)10. Nanoclusters described herein may take possess a shape selected from tetrahedron, sphere, ellipsoidal, or triangular plate. Attorney Ref. No.10457-559PC0 Another embodiment relates to device that incorporates nanoclusters described herein. In a specific example, provided is a solar cell that includes a plurality of nanoclusters as described herein. In another example, provided is a light-emitting diode comprising a plurality of nanoclusters as described herein. In a further example, provided is a photocatalysis process that involves use of a nanocluster as described herein. EXAMPLES Example 1: Methods Chemicals Selenium (Se, pellets, < 5 mm, ≥ 99.99%, trace metal basis), lithium triethylborohydride solution (LiBEt3H, 1.0 M in tetrahydrofuran), chlorotrimethylsilane (SiMe3Cl, ≥99%), copper (I) acetate (97%), diethyl ether (Et2O, anhydrous, ≥99.7%), methanol (MeOH, anhydrous, 99.8%), dichloromethane (CH2Cl2, anhydrous), tetrahydrofuran (THF, anhydrous, ≥99.9%), trioctylphosphine (P(octyl)3, 97 %), tri-n-butylphosphine (PBu3, 99%), tri-n-propylphosphine (PPr3, 97%), triethylphosphine (PEt3, 99%), cadmium chloride (CdCl2, 99.99%), toluene (anhydrous, 99.8%), and toluene-d8 (anhydrous, 99.6 atom % D) were purchased from Sigma- Aldrich. Diethylphenylphosphine (PEt2Ph, 97%), cadmium iodide (CdI2, 98%), cadmium bromide (CdBr2, 99.99%), and anhydrous n-heptane (99%) were purchased from Alfa Aesar. Chloroform- d (CDCl3, ≥99.8 atom % D + 0.05% v / v TMS) was purchased from Cambridge Isotope Laboratories and stored in a nitrogen-filled glovebox after degassing via freeze-pump-thaw. Synthesis of precursors and products General considerations: All synthetic steps were carried out in an air-free environment using a nitrogen-filled glovebox or standard Schlenk techniques. Synthesis and crystallization of Cu26Se13(PEt2Ph)14 cluster precursor: Cu26Se13(PEt2Ph)14 cluster was prepared based on the reported method49. In a typical synthesis, Cu(I) acetate (0.37 g, 3 mmol) was suspended in 27 mL of diethyl ether. A clear colorless solution was obtained after adding PEt2Ph (1.0 mL, 6 mmol). The solution was cooled to −78 °C, and bis(trimethylsilyl)selenide70 Attorney Ref. No.10457-559PC0 (0.37 mL, 1.5 mmol) was added. The reaction mixture turned from yellow, orange, to dark red in ten minutes. Then the reaction vessel was stored at −20 °C. Red rhombic crystals formed after one to two weeks. The crystals were washed with cold methanol to remove the impurities on the surface and dried under a vacuum. Yield: 0.403 g (69.7 % based on Se). Synthesis and crystallization of CdI2(PPr3)2complex precursor: CdI2(200 mg, 0.546 mmol) was suspended in 4 mL of toluene. Two equivalents of PPr3(220 μL, 1.092 mmol) were added at room temperature while stirring, yielding a clear colorless solution. The reaction solution was stored at −20 °C, and colorless needle-like crystals of CdI2(PPr3)2came out after one week. Synthesis and crystallization of Cd26Se17I18(PPr3)10cluster product via cation exchange reaction: Two equivalents of PPr3(208 µL, 1.04 mmol) were added into a suspension of CdI2(190.4 mg, 0.52 mmol) in 3 mL of toluene. After stirring for 10 minutes at room temperature, a clear colorless solution formed. Cu26Se13(PEt2Ph)14 cluster (50 mg, 0.01 mmol) was dissolved in 1 mL of toluene and injected into the vigorously stirred CdI2(PPr3)2solution at room temperature. The color of the solution changed from red to colorless in a few seconds. The reaction mixture was stirred at room temperature for 24 hours. To isolate the product, the reaction mixture was first concentrated under vacuum to remove most of the solvent. Around 1.2 mL of n-heptane was added to the concentrated mixture to form a suspension. The mixture was centrifuged at 10,000 rpm for 3 minutes, and the supernatant was discarded. The pellets were redispersed in 0.8 mL of toluene and 1.2 mL of n- heptane followed by centrifugation. The washing cycle was repeated one more time. The final precipitates were dissolved in CH2Cl2and dried under vacuum, resulting in 41 mg of light-yellow powder. The powder was dissolved in 0.5 mL of CH2Cl2 and layered with 1 mL of toluene. Colorless hexagonal crystalline plates were observed after three days at room temperature. The crystals were washed with toluene and dried, giving a yield of 26 mg (41.7% based on Se). The obtained clusters were stable in crystalline form for more than a month but gradually aggregate in CH2Cl2. The aggregation can be slowed down by diluting the CH2Cl2 solution of clusters with a large amount of toluene. Characterization Ultraviolet-visible (UV-vis) absorption spectra were recorded on a Cary 5000 spectrophotometer in the range from 200 nm to 800 nm with a 1 nm step size. The spectra of Attorney Ref. No.10457-559PC0 Cu26Se13(PEt2Ph)14crystals and Cd26Se17I18(PPr3)10crystals were measured in toluene and CH2Cl2, respectively. For the in-situ experiments, a time interval of five minutes was set for the first hour of data collection and then changed to one hour in the following period. The toluene solution of CdI2(PPr3)2 (1.07 mg / mL, 1.56 mM) and Cu26Se13(PEt2Ph)14 (1 mg / mL, 0.2 mM) were prepared separately. In a typical experiment, Cu26Se13(PEt2Ph)14 solution (1 mL, 0.2 μmol) was injected into the CdI2(PPr3)2solution (10 mL, 15.6 µmol) upon stirring. The final concentration was 18.2 μM based on initial Cu26Se13(PEt2Ph)14 (or 236 μM based on Se element). The color of the mixture turned colorless within one second. The reaction solution was transferred into a cuvette and sealed before bringing out of the glovebox for in-situ measurements. The time between mixing and the first scan was around two minutes. In-situ experiments with different CdI2(PR3)2complexes, temperatures, and concentrations were carried out in a similar procedure by changing the corresponding parameter accordingly. The stock solutions of CdI2(PR3)2 complexes (PR3 = P(octyl)3, PBu3, or PEt3) were prepared in a similar method as CdI2(PPr3)2. NMR data were recorded on a 600 MHz Bruker Avance III HD NMR spectrometer using 16 scans for1H spectra and 64 scans for31P spectra. The gradient-selected Double-Quantum Correlation Spectroscopy (gDQCOSY) was used for two-dimensional NMR spectra at low temperature (−40 °C). The temperature controller functioned with liquid nitrogen and a heater was employed to collect data at variable temperatures (−40 to 20 °C). Samples of PEt2Ph (20 µL) and Cu26Se13(PEt2Ph)14cluster (50 mg) were prepared in 0.6 mL of toluene-d8. Samples of PPr3(20 µL), CdI2(PPr3)2 (20 mg) and Cd26Se17I18(PnPr3)10 (40 mg) were made in 0.6 mL of CDCl3. Chemical shifts are given in δ ppm and residual solvent peaks were used as internal reference in1H spectra (δCH3= 2.08 ppm for toluene-d8and δ = 7.26 ppm for CDCl3). Powder XRD characterization was carried out on a Panalytical X’Pert Powder diffractometer using Cu Kα radiation (λ=1.54056 Å) operated at 45 kV and 40 mA. Finely-ground powder samples were deposited on a glass plate holder and loaded into the flat stage. Data were collected at 2θ values from 5° to 90° using stepwise scans of 0.008° with a time per step of 10.16 seconds under ambient conditions. The reference XRD patterns (see Fig. 2e, j) were simulated from bulk bellidoite Cu2Se (ICSD-30230), zinc blende CdSe (ICSD-41528), and wurtzite CdSe (ICSD-41491). Attorney Ref. No.10457-559PC0 Matrix-Assisted Laser Desorption Ionization-Time of Flight (MALDI-TOF) Mass Spectrometry was performed on a Bruker Autoflex LRF instrument with a laser (335 nm). For sample preparation, 1,8,9-trihydroxyanthracene was used as a matrix, which was dissolved in toluene or CH2Cl2with a concentration of 88 mM (i.e., 20 mg in 1.0 mL of solvent). Solid samples of Cu26Se13(PEt2Ph)14 cluster and Cd26Se17I18(PPr3)10 cluster were dispersed in toluene and CH2Cl2, respectively, with a concentration of 1 mg / mL. 50 μL of matrix solution and 50 μL of analyte solution were mixed in air.0.8 μL of the mixture was spotted onto a stainless-steel target plate and allowed to dry followed by spotting another 0.8 μL of the mixture. A mixture of insulin, cytochrome, myoglobin, and ubiquitin was used as the external standard for calibration in the mass range of 4-20 kDa. The sample was measured at the laser power of 80-90%. The detector gain (linear detector, positive mode) was set as 17.6´ (2936 V). X-ray crystallographic determination The X-ray intensity data were collected at 100 K on a Bruker Dual micro source D8 Venture diffractometer equipped with PHOTON III detector using Mo Kα radiation (λ = 0.71073 Å). The data frames were integrated, and multi-scan scaling was applied using APEX3 software package of programs. Intrinsic phasing structure solution provided all the non-H atoms. The structure was refined using full-matrix least-squares refinement in the Bruker SHELXTL Software Package. The non-H atoms were refined with anisotropic displacement parameters (ADPs). All the H atoms were calculated in idealized positions and refined riding on their parent atoms. For the Cd26Se17I18(PPr3)10cluster, a hexagonal crystal specimen (approximate dimensions: 0.404 mm × 0.240 mm × 0.172 mm) was used for X-ray crystallographic analysis. The nanocluster crystallizes in the space group P3^^1, with Z = 4 (Z’ = 0.3333) for the formula-unit content Cd26Se17I18P10C142.5H270. The final cell constants are a = 26.2849(4) Å, b = 26.2849(4) Å, c = 42.3692(11) Å, α = 90°, β = 90°, γ = 120°, volume = 25350.9(10) Å3. The integration of the data yielded a total of 459723 reflections to a maximum θ angle of 27.5° (0.75 Å resolution), of which 19433 were independent (average redundancy = 4.32, completeness = 100.0%, Rint= 6.68%, Rsig= 2.23%). Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.640. For the Cu26Se13(PEt2Ph)14cluster, a red rhombic crystalline plate (approximate dimensions: 0.178 mm × 0.176 mm × 0.122 mm) was used for the X-ray crystallographic analysis. Attorney Ref. No.10457-559PC0 The nanocluster crystallizes in the space group P1^, with Z = 2 for the formula-unit content Cu13Se6.5P7C74H115O1. The final cell constants are a = 17.432(3) Å, b = 18.014(3) Å, c = 18.257(3) Å, α = 98.808(5)°, β = 109.168(4)°, γ = 118.101(5)°, volume = 4439.1(13) Å3. The integration of the data yielded a total of 260538 reflections to a maximum θ angle of 32.802° (0.75 Å resolution), of which 30802 were independent (average redundancy = 7.53, completeness = 99.9%, Rint = 6.99%, Rsig= 4.19%). Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.708. For the CdI2(PPr3)2 complex, a colorless needle-like crystal (approximate dimensions: 0.337 mm × 0.100 mm × 0.057 mm) was used for the X-ray crystallographic analysis. The complex crystallizes in the space group P21 / c, with Z = 4 for the formula-unit content Cd1I2P2C25H50. The final cell constants are a = 7.9162(3) Å, b = 14.8295(7) Å, c = 28.1826(13) Å, α = 90°, β = 94.3040(10) °, γ= 90°, volume = 3299.1(3) Å3. The integration of the data yielded a total of 142002 reflections to a maximum θ angle of 35.273° (0.75 Å resolution), of which 13999 were independent (average redundancy = 6.30, completeness = 99.6%, Rint = 5.22%, Rsig = 2.95%). Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.381. Example 2: Precise precursors for cation exchange reaction To achieve atomic precision in the cation exchange reaction, an atomically defined Cu26Se13(PEt2Ph)14 cluster was selected, where PEt2Ph is diethylphenylphosphine49, as a precise reaction precursor (Fig. 2a, center). Copper(I) chalcogenide (Cu2-xE, E = S, Se, Te) NCs are common host templates for cation exchange40–46. The high mobility of Cu+as well as its favorable solvation with soft Lewis bases (e.g., phosphines) drive the complete exchange of Cu+with various harder cations50. In addition, an existing library of atomically precise Cu2-xE clusters has been developed by Fenske and coworkers51, which serves as ideal cluster templates for precise transformation. The Cu26Se13(PEt2Ph)14cluster was synthesized based on previously reported method49(see Methods). The cluster crystallized as red rhombic crystals (Fig. 2c) and its structure was verified by sc-XRD. As shown in the shell-by-shell anatomy of Cu26Se13P14 framework (Fig. 2b), the cluster features a Se center surrounded by twelve core Cu(I) cations (labeled asoCu) in a Attorney Ref. No.10457-559PC0 distorted icosahedron mode, which is orthogonally covered by another Se12icosahedral shell. Eight Cu cap the triangular facets of the Se12 icosahedron (labeled asiCu) and another six Cu bridge its edges (labeled asiiCu), forming the surface Cu14 rhombic dodecahedral shell. Each surface Cu is datively coordinated to a PEt2Ph ligand. The overall symmetry of the cluster is Ci. The absorption spectrum of the cluster dissolved in toluene shows a decay curve (Fig. 2d), which is consistent with that of non-plasmonic Cu2E nanocrystals52. Powder XRD shows broad diffraction peaks because of the ultrasmall size of the cluster and its icosahedron-based packing (Fig. 2e, red line), which matches the pattern simulated from Cu26Se13(PEt2Ph)14 structure (Fig. 2e, grey line). NMR is performed to examine cluster structure in the solution phase. Due to the fluxionality of phosphine ligands53, broad peaks are observed at room temperature. But seven groups of peaks with two-fold degeneracy can be resolved in low-temperature1H,31P and1H-1H COSY spectra (Fig. 2f, g and Fig.5). The splitting matches with seven chemical of fourteen PEt2 Ph ligands under Ci symmetry, suggesting the cluster remains intact in toluene solution54. Another crucial reagent is the cation precursor. We identify a precisely defined CdI2(PPr3)2coordination complex, where PPr3 is tri-n-propylphosphine, as the carrier for incoming cation (Fig. 2a, periphery) after multiple optimizations (vide infra). Common cation carriers used in colloidal cation exchange, such as metal oleate or oleylamine compounds28,41–45, often form coordination polymers with undefined coordination centers and flexible hydrocarbon groups13,55, which introduce randomness to the reaction system. In contrast, the Cd(II) complex coordinated to halides and phosphines shows a discrete and tetrahedrally coordinated cation center56, which can simplify diffusion kinetics and enable precise collision pathways. In addition, the phosphine ligands in the complex concurrently serve as soft bases to extract Cu+and drive the forward exchange reaction50. It was also found that the coordination complex not only controls the reactivity of the cation precursor but also determines the stability of the product cluster (vide infra). The CdI2(PPr3)2complex was prepared by reacting CdI2 and PPr3 in toluene (see Methods). Colorless needle-like crystals were obtained at −20 °C and the tetrahedrally coordinated Cd(II) complex was confirmed by sc-XRD (Fig. 6). In a typical cation exchange reaction, one equivalent of Cu26Se13(PEt2Ph)14was mixed with 52 equivalents of CdI2(PPr3)2in toluene at room temperature. The solution color changes from dark red to colorless within a few seconds (Fig. 2h, top), and two sharp absorption peaks at 372 nm and 347 nm are observed after one day (Fig.2i, solid line). The distinct peaks are characteristics Attorney Ref. No.10457-559PC0 of excitonic transitions of CdSe nanomaterial22,57. The same absorption profile can be reproducibly observed in different batches of syntheses, indicating the two peaks belong to a single CdSe cluster. The cluster was isolated by precipitation with heptane. Powder XRD shows two broad peaks centered at ~26oand ~45o(Fig. 2j, green line), which deviate from the bulk zincblende (ZB) and wurtzite (WZ) phases (Fig. 2j, black lines) but match the simulated pattern of resolved structure described below (Fig.2j, grey line). Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) reveals a series of characteristic fragmentation peaks of CdSe materials22,57,58, which are different from Cu26Se13 fragments (Fig. 7). Interestingly, the1H NMR spectrum of the isolated product only shows peaks in the aliphatic region (Fig. 2k), suggesting that the original PEt2Ph ligands are replaced with PPr3. Although the characterization cannot provide detailed composition and structure of the product, it confirms the reaction proceeds and results in a well- defined CdSe cluster. Example 3: Atomically precise CdSe cluster as the product of cation exchange reaction The CdSe cluster produced by the reaction of Cu26Se13(PEt2Ph)14 and CdI2(PPr3)2 can be crystallized into colorless hexagonal plates in CH2Cl2 / toluene solution (Fig.2h, bottom). The total structure of a Cd26Se17I18(PPr3)10 cluster was solved by sc-XRD (Fig. 3a). The cluster features a tetrahedron-based but non-close-packed Cd26Se17 core with an edge length of 1.1 nm (Fig. 8a). The overall size including the ligands is 2.2 nm. The Se17anion lattice is composed of a Se13icosahedron and four apical Se, while the Cd26cation lattice can be viewed as four interpenetrating icosahedra in a tetrahedral arrangement (Fig. 8b). All the Se and Cd are tetrahedrally coordinated except the four apical µ3-Se. The shell-by-shell anatomy of the Cd26Se17 core is shown in Fig. 3b. The cluster features a Se center surrounded by four core Cd (labeled asoCd). EachoCd is further connected to three Se and the total twelve Se assemble into an icosahedron shell. The Se12 shell is face-capped by four Cd (labeled asiCd), edge-bridged by six Cd (labeled asiiCd), and vertex- terminated by twelve Cd (labeled asiiiCd). TheiCd4iiCd6iiiCd12forms the surface Cd22shell, withiiiCd12 capped by four apical Se. Significantly, the icosahedral Se13 lattice of Cd26Se17 contrasts with zincblende and wurtzite lattices commonly observed in CdSe NCs3,4,59and clusters21,22, but is nearly identical to that of Cu26Se13(Fig. 10). The resemblance indicates the Se13lattice of the child Cd26Se17cluster Attorney Ref. No.10457-559PC0 is inherited from that of the parent Cu26Se13cluster, and the anion sublattice can be preserved during cation exchange in the ultrasmall cluster. The rigidity of anion lattice can be explained by the favored cohesion energy between cations and anions in the cluster framework, while it is energetically unfavored to dissociate and solvate ionic building blocks in the nonpolar reaction medium. It is worth noting that the icosahedron is a frequently observed structural motif in metal clusters due to its low surface energy9,24. A few reported semiconductor clusters also show deviation from bulk phases (e.g., In37P20, Cd14Se13, and Zn14Se13)23,60,61. A careful examination of these structures shows that their anion lattices are also based on icosahedron (e.g., interpenetrating bi-icosahedra P19 in In37P20). It is interesting that semiconductor clusters could share similar atomic packings with metal clusters when cation and anion lattices are considered independently.23,60,61Note that the expansion of the anion lattice from Se13to Se17is attributed to the subsequent reaction of CdSe clusters after cation exchange (vide infra). In contrast to core anions, the surface ligands of child Cd26Se17I18(PPr3)10 cluster are exclusively inherited from CdI2(PPr3)2precursor, revealing that cation exchange is accompanied by surface ligand exchange. All the ligands are coordinated to surface Cd2+(Fig.3c). While iodides can be found on facets, edges, and corners (Fig. 3c, purple), phosphines are mainly on vertexes except for one at the bottom facet (Fig. 3c, pink). Based on the tetrahedral coordination of Cd(II), each surface Se3iCd, Se2iiCd, and Se2iiiCd site exposes 1, 2, and 2 dangling bonds, respectively. The empty sites on four facetiCd are saturated by three terminal iodide (tI) and one PPr3. The six edgeiiCd are passivated by twelve bridging iodides (bI), which further connect toiiiCd. The remaining sites on twelve vertexiiiCd are protected by threetI and nine PPr3. In addition to saturating coordination sites, the eighteen iodides also balance the positive charges of the Cd26Se17 core and the ten phosphines provide steric protection around the cluster. The ligands also partially shield the undercoordinated apical Se sites (Fig. 8a). While the Cd26Se17core has a T symmetry, the symmetry of entire cluster is reduced to C3 due to the distribution of ligands around the principal axis.31P and1H NMR spectra display a 3, 3, 3, 1 distribution of the ten PPr3, which matches with the C3 of the ligand shell (Fig. 2l, Fig. 9). The product cluster not only hybridizes the structural features from both parent precursors but also evolves into an enlarged anion lattice with extended surface patterns. Linked by bridging iodides, three P-iiiCd-I-iiCd-I-iiiCd-P chains emanate from the top corner of the tetrahedron (Fig. 3d), and another three P-iiiCd-I-iiCd-I-iiiCd-I chains wrap the base (Fig.3e). While both precursors Attorney Ref. No.10457-559PC0 are achiral, chirality emerges in the product, as reflected in the clockwise or counterclockwise rotation of surface patterns (Fig. 3d, e). Surface chirality can be traced back to the core, which originates from the integration of tetrahedral coordination of Cd into the icosahedral framework of Se (Fig. 11a). The core chirality is manifested by the twisted alignment of Cd-Se bonds along the principal axis, while achiral Se13 cuboctahedron based on the zincblende phase exhibits vertically aligned Cd-Se bonds. The cluster also features an intrinsic dipole moment due to the uneven distribution of ligands, with phosphines concentrated at the top and iodides at the base (Fig.8a). The inter-cluster dipole interactions27facilitate the head-to-tail alignment of enantiomers in the unit cell (Fig. 11b) and their ordered assembly into single crystals (Fig. 3f). The Cd26Se17I18(PPr3)10crystals dissolved in CH2Cl2show two sharp excitonic absorption transitions at 371 nm (E1) and 346 nm (E2) and a hump at 317 nm (E3) (Fig. 2i, dash line). The peaks blueshift by 1 nm compared to that of the reaction mixture due to different solvent permittivity. The full widths at half maximum of E1 and E2 are 18 nm (161 meV) and 20 nm (207 meV), respectively. The molar extinction coefficients are 1.08 × 105M−1cm−1for E1and 5.83 × 104M−1cm−1for E2. The E1peak position is comparable to that of a zincblende-based selenolate- protected Cd32Se14cluster (373 nm)21. In addition, many CdSe magic-sized clusters (MSCs) have been identified by their characteristic absorption57,58,62, including a series of carboxylate-protected MSCs with peaks at ~360 nm, 384 nm, 406 nm, and 431 nm. These MSCs are theoretically assigned as (CdSe)13, (CdSe)19, (CdSe)33, and (CdSe)66, respectively63. The Cd26Se17seems to fit into this sequence when the total number of CdSe units is considered. We shall note that excitonic transitions of ultrasmall clusters could be sensitive to structural details such as core shape and symmetry64as well as surface ligands59,65in addition to sizes. No photoluminescence is detected at room temperature. This is consistent with other structurally resolved semiconductor clusters19–22, where broad emissions could only be observed at cryogenic temperature. The lack of emission can be attributed to the ultrasmall core size and large surface area, which enhance the coupling between excitons and surface / lattice vibration and increase the energy dissipation via nonradiative pathways.19–22Also, the dangling bonds on apical Se could lead to the trap states in the optical gap that quench the emission6,15,16. Example 4: Precise transformation pathways and atomic insights into the reaction mechanism Attorney Ref. No.10457-559PC0 Two interesting questions arise in Cu26Se13(PEt2Ph)14 to Cd26Se17I18(PPr3)10 transformation (Fig. 4a): how does the anion framework expand from Se13 to Se17? and how are surface ligands exchanged? To answer these questions, the reaction was monitored by in-situ UV-vis spectroscopy. Upon injecting one equivalent of Cu26Se13 solution into 78 equivalents of CdI2(PPr3)2 solution, the reaction mixture changes to colorless within one second (Fig. 4b). Interestingly, the 2 min reaction mixture shows a well-defined peak at 365 nm, which blueshifts to 355 nm in the next ~10 min and stabilizes for another ~20 min (Fig. 4c). In the next 24 hours, the 355 nm peak slowly redshifts to the characteristic 372 nm of the final product (Fig. 4d). Both shifts (365 nm → 355 nm → 372 nm) occur in a quantized manner with the presence of isosbestic points, indicating the reaction proceeds with direct conversion among well-defined intermediates. The final spectrum is stabilized for more than 72 hours with only a subtle change in absorbance. According to the final absorbance of the 372 nm peak and its extinction coefficient, the conversion yield from Cu26Se13 to Cd26Se17 is near unity on the Se basis. The precise structures of precursors and products determined by X-ray crystallography and the pathways mapped by in-situ spectroscopy enable us to study reaction mechanisms at the atomic level (Fig. 4e). Remarkably, the structure of the child cluster (including core and surface) can be atomically linked to precursors by following two assumptions: (i) the reaction intermediates should be nearly neutral as highly charged species are unstable in a nonpolar medium; (ii) the coordination of intermediates should be close to saturation to minimize uncontrolled aggregations. Based on these charge and coordination conservations, the overall reaction can be roughly divided into the following stages (Fig. 4e). The first stage can be described as surface coordination exchange: instead of simply exchanging cations, the exchange occurs between the entire surface coordination moiety and the incoming complex (Fig.4e, I). For example, upon interacting with CdI2(PPr3)2, each edge-bridging [iiCu(PEt2Ph)]+moiety gains an I−and a PPr3 and leaves the host cluster as a charge-neutral CuI(PEt2Ph)(PPr3) complex (Fig. 4e, I.1). Concurrently, the residual [CdI(PPr3)]+re-bridges the exposed Se edge sites and restores its tetrahedral coordination. In this way, the charge and coordination of the cluster and the complex are conserved during the conversion. Similarly, each face-capping [iCu(PEt2Ph)]+sites can be exchanged into [iCdI]+, with CuI(PEt2Ph)(PPr3)2as the Attorney Ref. No.10457-559PC0 leaving group (Fig.4e, I.2). Altogether, the SeoCu12Se12[iCu(PEt2Ph)]8[iiCu(PEt2Ph)]6is converted to SeoCu12Se12[iCdI]8[iiCdI(PPr3)]6. The second stage should be intra-cluster cation diffusion to exchange coreoCu (Fig. 4e, II). This can be achieved by the inward diffusion of fouriCd to form the SeoCd4center and the outward diffusion of 12oCu to occupy the vertex positions of Se12 icosahedron. TheoCu+can subsequently be replaced by [iiiCdI]+and removed as CuI(PPr3)2, leading to a fully exchanged Cd26Se13I26(PPr3)6intermediate (Fig. 4e, I.3). The following stage could be surface reorganizations, as the newly exchangediiiCd sites are undercoordinated and unstable (Fig. 4e, III). One way to saturateiiiCd sites is to convert twelve terminal iodides into bridging ones, which links originally isolatediiCd andiiiCd sites into six chains. In the meantime, the PPr3oniiCd is shuffled to the end of the chain. The uneven distribution of six PPr3to four corners breaks the T symmetry into C3 (Fig. 12). The last stage should be inter-cluster digestive ripening (Fig. 4e, IV). The reorganized Cd26Se13I26(PPr3)6is still unstable due to high-energy face-capping iodides (fI−) at four corners and insufficient steric shielding from PPr3. During inter-cluster collisions, some clusters could dissociate and release Se2−, which can be readily captured by other clusters to replacefI−and form [Cd26Se17I22(PPr3)6]4−(Fig. 4e, IV.1). The introduced negative charges can be rebalanced by converting four terminal iodides (three atiiiCd and one atiCd) to PPr3(Fig. 4e, IV.2), which concurrently provides more shielding to the cluster. Finally, a stable Cd26Se17I18(PPr3)10 product is reached with balanced charges, saturated surface Cd sites, and sufficient steric protection. The observed ligand exchange during cation exchange can be readily explained by surface coordination exchange (Fig. 4e, I). This process can be very fast considering the lability of phosphine and its favorable solvation with Cu+, which is reflected in the initial rapid color exchange from red to colorless in seconds (Fig. 4b). The expansion of the anion lattice from Se13to Se17 can be explained by the digestive ripening (Fig. 4e, IV), which is evidenced in the slow 355 nm → 372 nm redshift in hours (Fig. 4d). The inter-cluster reaction is expected to be slow because it requires the dissociation and reassembly of the entire cluster framework. The ripening process can also be understood as precise aggregation, which is indeed frequently observed in semiconductor cluster synthesis, including the earlier work of converting Cd10S4(SPh)12cluster into Cd32S14(SPh)36 in coordination solvent of pyridine19as well as converting 350 nm CdSe cluster into 380 nm and 408 nm species22. Attorney Ref. No.10457-559PC0 While the redshift period can be readily correlated to the anion lattice enlargement, the initial blueshift period (365 nm → 355 nm in minutes, Fig. 4c) is more puzzling. One possibility is to assign the 365 nm species to partially exchanged core-shell Cu12Cd14Se13I14(PPr3)6 (Fig. 4e, I.2). This is reasoned by the slower exchange of core cations compared to surface ones36, leading to the temporally trapped core-shell species. The blueshift can thus be explained by the exclusion of coreoCu, since the excitonic transition is known to be shifted in heavily doped or core-shell semiconductor nanocrystals66,67. The other possibility is to assign the 365 nm species as the fully exchanged iodide rich Cd26Se13I26(PPr3)6(Fig. 4e, I.3) and the 355 nm species as alternative surface reorganized [Cd14Se13I4(PPr3)12]2−. The blueshift can thus be explained by the depletion of iodides on the surface (from I22to I4). This assignment is also supported by a recently resolved Cd14Se13Cl2L6 cluster (L = Me2NC2H4NMe2), which shows E1 at 354 nm60. It is noted that the intermediates are highly dynamic and constantly exchanging (especially on surfaces). The 365 nm and 355 nm species could correspond to a group of clusters with the same core but dynamic surface structures. It is unclear yet to what extent the depletion of core Cu+or surface I−will cause the absorption shift. Future studies on precise structure-property correlations for doped or ligand- exchanged clusters are necessary to confirm the mechanism. The effects of various reaction conditions on cation exchange were investigated. Varying alkyl chains of phosphine from octyl, butyl to ethyl in CdI2(PR3)2complex does not alter the reaction pathway and intermediates. The same 365 nm → 355 nm → 372 nm conversions are observed (Fig. 13). But the reaction is much faster with CdI2(PEt3)2 due to reduced diffusion and collision barriers for complexes with shorter ligands. In contrast, changing halide ligands has more significant effects in the final product: the reaction with CdBr2(PPr3)2or CdCl2(PPr3)2complex leads to uncontrolled aggregations into larger CdSe clusters (Fig. 14). This can be explained by the smaller sizes of chloride (3.3 Å) and bromide (3.9 Å)68, making them less effective than iodide (4.4 Å) as bridging ligands required by Cd26Se17structure (the averageiiCd-iiiCd distance is 4.46 ± 0.07 Å, Fig. 3c). The reaction is faster when performed at 40oC, but further increasing the temperature to 75oC leads to the aggregation of CdSe clusters. The reaction is nonstoichiometric but kinetically controlled: insufficient amounts of CdI2(PPr3)2 (13 to 26 equivalents to Cu26Se13) result in irreversible aggregations at the initial stage of reaction probably due to the lack of instant surface passivation, while excess amounts (52 to 117 equivalents) promote the reaction and accelerate the final product formation. These results further demonstrate the complexity of cation Attorney Ref. No.10457-559PC0 exchange on nanoclusters, which require precise reactions not only in the atomic core but also at the coordination surface and inter-cluster level to reach a stable product. Thus, fine-tuning reaction conditions and balancing hierarchical interactions are important for precise transformations. Conclusion In summary, Examples 1-4 report the precision synthesis of Cd26Se17I18(PPr3)10 cluster via cation exchange reaction between Cu26Se13(PEt2Ph)14cluster and CdI2(PPr3)2complex. The employment of atomically defined precursors is crucial to the precise transformation. While Cu26Se13(PEt2Ph)14 cluster provides the template for anion framework, the coordination complex CdI2(PPr3)2controls the reactivity of incoming cations, the solvation of outgoing cations, and the surface structures and stability of the resulting cluster. It is shown that cation exchange on nanocluster bridges the solid-state conversion in the core and molecular coordination reaction on the surface. The precise reaction provides atomic evidence that anion lattice can be preserved in ultrasmall clusters and unveils that cation exchange is accompanied by surface ligand exchange. This enables the access of atomically precise CdSe nanoclusters with novel core and surface structures. The child Cd26Se17I18(PPr3)10 cluster inherits its icosahedral anion lattice from the parent cluster and its tetrahedrally coordinated cation surface from the parent complex, and further evolves into an enlarged core and extended surface with emerging chirality and polarity. The precise reaction also provides atomic insights into the reaction mechanism, which include surface coordination exchange occurred in seconds; intra-cluster core or surface reorganizations in minutes; and inter-cluster digestive ripening in hours. Despite numerous collisions, exchanges, and reorganizations required to construct a nanostructure, Examples 1-4 demonstrate that precision nanosynthesis with atomically defined precursors, pathways, and products can be realized via synergizing coordination, cluster, and colloidal chemistry in the nanoscale reaction. The emergence of atomic precision in the complex nanoscale reaction is reminiscent of biomolecular reactions in living systems, which all proceed with atomically defined pathways despite the intricacies of the process69. It is expected that precision nanosynthesis will provide a designable approach to access a diverse library of atomically precise nanoclusters and further facilitate atomic engineering of functional nanostructures for energy and information applications. Attorney Ref. No.10457-559PC0 Example 5: Transformative syntheses of atomically precise II14VI13 clusters using MX2(tmeda) complexes Introduction nanocrystals or quantum dots (QDs) are often considered “artificial atoms” because of their discrete electronic states with confined excitonic transitions resembling those of atoms (1–3) as well as their spherical shapes allowing their assembly into exotic “artificial solids” (4, 5). An important challenge in artificial atoms is their size dispersity and structural defects, which hinder the precise structure-property correlation. A generalizable and precise synthetic strategy is highly desired for the atomic engineering of artificial atoms to enhance their performance in energy conversion applications and emerging technologies including quantum computing and exotic artificial solids (6). Compared to small molecules, colloidal QDs have much more complex structures composed of tens to thousands of core atoms and surface-capping molecules. The precise construction of colloidal nanocrystals is more challenging, which requires controlling not only individual bonds between atoms and molecules but also their ensemble packing at hierarchical levels (7, 8). Despite the challenges, a few atomically precise semiconductor nanoclusters have been discovered during the past four decades, including Cd32S14, Cd54S20, and In37P20(9–17). While these clusters have precise structures, their syntheses are less defined, which has hindered further investigation of their properties. On the other hand, semiconductor magic-sized clusters (MSCs) are often observed during the colloidal synthesis of QDs. They are characterized by distinct and consistent absorption features, such as CdSe MSCs with peaks at 350, 384, and 406 nm (18–23). Despite their precise optical properties, their atomic structures remain largely elusive, as their characterization is hampered by ultrasmall sizes and dynamic structures (23). To decipher the enigma of precise structure-property correlation in artificial atoms calls for a generalizable method that could bridge the gaps between conventional coordination and colloidal chemistry. Compared to the direct synthesis of QDs from molecular precursors via nucleation and growth (24–26), colloidal transformation reaction via cation exchange (e.g., CdSe QDs + Ag+→ Ag2Se QDs + Cd2+) has provided a shortcut to access QDs with novel phases, shapes, and heterostructures (27–29). Examples 1-4 above, present the “precisionalization” of cation exchange reaction: by using atomically defined precursors, including a parent Cu26Se13(PEt2Ph)14 cluster as the template and a CdI2(PPr3)2complex as cation carrier, a child Cd26Se17I18(PPr3)10cluster was Attorney Ref. No.10457-559PC0 produced with near-unity yield (see also 30). While the child cluster inherits its anion lattice from the parent cluster, its cations and surface ligands are dictated by the parent coordination complex. It is envisioned that such a transformative approach is highly modulable, as it decouples the controls of cations, anions, and surface ligands of nanoclusters into different reaction precursors, therefore allowing rational control and atomic engineering of semiconductor nanoclusters. In this Example 5, the generalization of transformative synthesis to access a library of atomically precise II-VI clusters is demonstrated. To introduce different IIB cations, a universal MX2L coordination molecule (“coordination complex”) was identified as the cation carrier, where M (“Cat2”) = Zn2+, Cd2+, Hg2+; X = I–, Cl–; L = tetramethylethylenediamine (tmeda). Their reactions with the parent Cu26Se13cluster lead to a family of Zn14Se13, Cd14Se13, and Hg14Se13clusters with high purity. The precise synthesis enables the resolution of both atomic structures and distinct optical absorption of the sequence of II-VI clusters, which further facilitates the theoretical calculation to decipher the electronic structures and “periodic” rules of these artificial atoms. Chemicals Selenium (Se, pellets, < 5 mm, ≥ 99.99%, trace metal basis), lithium triethylborohydride solution (LiBEt3H, 1.0 M in tetrahydrofuran), chlorotrimethylsilane (SiMe3Cl, ≥99%), copper (I) acetate (CuOAc, 97%), zinc iodide (ZnI2, ≥98%), mercury iodide (HgI2, ≥99%), zinc chloride (ZnCl2, ≥98%), cadmium chloride (CdCl2, 99.99% trace metal basis), N,N,N′,N′- tetramethylethylenediamine (tmeda, 99%), tri-n-propylphosphine (PPr3, 97%), sodium tetraphenylborate (NaBPh4, 99.5%), diethyl ether (Et2O, anhydrous, ≥99.7%), methanol (CH3OH, anhydrous, 99.8%), dichloromethane (CH2Cl2, anhydrous), pentane (anhydrous, ≥99%), toluene (anhydrous, 99.8%), and acetonitrile (CH3CN, anhydrous, 99.8%) were purchased from Sigma- Aldrich. Diethylphenylphosphine (PEt2Ph, 97%), cadmium iodide (CdI2, 98%), and anhydrous n- heptane (heptane, 99%) were purchased from Alfa Aesar. Mercury chloride (HgCl2, ≥99.5%) was purchased from Thermo Scientific Chemicals. Chloroform-d (CDCl3, ≥99.8 atom % D + 0.05% v / v TMS) was purchased from Cambridge Isotope Laboratories and stored in a nitrogen-filled glovebox after being degassed using the freeze-pump-thaw method. Synthesis of the anion template cluster and cation carrier complexes Attorney Ref. No.10457-559PC0 General considerations. All synthetic steps were performed in air-free environment using nitrogen-filled glovebox or standard Schlenk techniques. Cu26Se13(PEt2Ph)14 cluster. The synthesis was based on the previous report.(1, 2) CuOAc (0.37 g, 3 mmol) and PEt2Ph (1.0 mL, 6 mmol) were added into 27 mL of diethyl ether. After stirring for 2 hours at room temperature, a clear solution formed and was cooled to –78 °C. Se(SiMe3)2(0.37 mL, 1.5 mmol) synthesized based on the literature (3) was cooled too and then injected into the Cu(I) precursor solution. The color of the reaction mixture changed from yellow, orange, to dark red in about ten minutes, indicating the formation of cluster species. Red rhombic crystals were obtained after storing the reaction mixture at –20 °C for one to two weeks. After separating the mother liquor, the crystals were rinsed with cold methanol and dried under vacuum. Yield: 0.47 g (81.4 %). ZnI2(tmeda) complex. ZnI2(319 mg, 1 mmol) was suspended in ~3 mL of CH2Cl2. Tmeda (300 µL, 2 mmol) was added, and the suspension became a clear colorless solution after stirring for 15 minutes at room temperature. The solution was filtered to remove a trace amount of unreacted ZnI2. After reducing the volume to 1 mL under vacuum, heptane (4 mL) was added to precipitate the complex. The resulting white suspension was brought out of the glovebox and centrifuged at 6,000 rpm for 5 min. The precipitates were extracted with ~5 mL of CH2Cl2. After filtration, the CH2Cl2 solution was dried under vacuum and white powders were obtained. Yield: 364 mg (83.6 %). Colorless needle-like crystals were grown in one day by layering heptane onto the CH2Cl2 solution of the product. The structure was confirmed by X-ray crystallography, which matched the reported structure of ZnI2(tmeda).(4) Unit cell parameters: a = 12.905(5) Å, b = 7.751(2) Å, c = 13.351(7) Å, α = 90.000°, β = 110.247(9)°, γ = 90.000°, and a cell volume of 1253.04 Å3in C2 / c space group. CdI2(tmeda) complex. CdI2(366.2 mg, 1 mmol) was suspended in ~3 mL of CH2Cl2,and tmeda (150 µL, 1 mmol) was added. After ten minutes of stirring at room temperature, the suspension became a clear colorless solution. The solution was filtered and layered with heptane. Needle-like crystals were obtained after one day. The crystals were rinsed with heptane and dried under vacuum. Yield: 275 mg (57.0 %). The product can be recrystallized from CH2Cl2 / heptane for X-ray crystallography. The structure matched the reported CdI2(tmeda) complex.(5) Unit cell Attorney Ref. No.10457-559PC0 parameters: a = 8.4514(8) Å, b = 12.1012(12) Å, c = 12.4691(12) Å, α = 90.000°, β = 97.209(2)°, γ = 90.000°, and a cell volume of 1265.16 Å3in P21 / n space group. HgI2(tmeda) complex. HgI2(91 mg, 0.2 mmol) was suspended in ~2 mL of CH2Cl2,and tmeda (60 µL, 0.4 mmol) was added. After ten minutes of stirring at room temperature, the red powders disappeared, and a clear colorless solution was obtained. The solution was filtered and dried under vacuum. Yield: 99 mg (86.8 %). Needle-like crystals were obtained after two to three days by layering heptane on top of a CH2Cl2 solution. The structure matched the reported HgI2(tmeda) complex.(5) Unit cell parameters: a = 23.182(2) Å, b = 7.2605(7) Å, c = 15.0566(13) Å, α = 90.000°, β = 111.233(3)°, γ = 90.000°, and a cell volume of 2362.17 Å3in C2 / c space group. Transformative synthesis and crystallization of II-VI child clusters [Zn14Se13(tmeda)6][BPh4]2 cluster (1). The as-synthesized ZnI2(tmeda) complex (26.1 mg, 60 µmol) and PPr3(24 µL, 120 µmol) were mixed in 0.5 mL of CH2Cl2. Cu26Se13(PEt2Ph)14cluster (10 mg, 2 µmol) dissolved in 0.5 mL of CH2Cl2 was injected into the complex solution at room temperature. The reaction mixture gradually changed from red to colorless in about two minute. After stirring for another 16 hours, the reaction volume was reduced to 0.5 mL and transferred outside the glove box. 4.5 mL of pentane was added to precipitate the cluster. White precipitate was collected after centrifuging at 6,000 rpm for 5 minutes and redispersed in 1.5 mL of CH2Cl2 followed by filtration. To grow high-quality crystals, a solution of NaBPh4(20 mg, 58 µmol) in 0.5 mL of CH3CN was combined with the CH2Cl2solution. Around 1 mL of heptane was layered on top of the mixture. Rod-like crystals of [Zn14Se13(tmeda)6][BPh4]2 were obtained after two to three days at room temperature. Yield: ~ 4.1 mg (77.7% on Se basis). Cd14Se13(tmeda)6I2 cluster (2). CdI2(tmeda) complex (29 mg, 60 µmol) and PPr3(72 µL, 360 µmol) were combined in 0.5 mL of toluene to form a clear colorless solution at room temperature. Cu26Se13(PEt2Ph)14 cluster (10 mg, 2 µmol) dissolved in toluene (0.5 mL) was injected. The reaction mixture changed to colorless within one second. The reaction vial was stored at ‒20 °C. Rod-like crystals of Cd14Se13(tmeda)6I2 were obtained after a couple of weeks. The cluster was found to co-crystallize with one CdI2(tmeda) complex and two toluene molecules. Yield: ~ 3.6 mg (50.7% on Se basis). Attorney Ref. No.10457-559PC0 Hg14Se13(tmeda)6I2 cluster (3). The addition of tmeda (1 mL) to HgI2(27.2 mg, 60 µmol) led to a clear colorless solution after stirring for 20~30 minutes, which was used as a stock solution. In a typical synthesis, 0.1 mL of the stock solution (6 µmol of Hg2+) and PPr3 (2.4 µL, 12 µmol) were combined with 0.8 mL of tmeda. The total volume of tmeda is 0.9 mL (6 mmol). Cu26Se13(PEt2Ph)14 cluster (1 mg, 0.2 µmol) dissolved in 0.1 mL of toluene was injected at room temperature. The color of the reaction mixture changed to bright yellow in one minute. After about one hour, the reaction mixture was filtered into an NMR tube and layered with heptane. Yellow needle-like crystals of Hg14Se13(tmeda)6I2 formed after one to two days, together with small amounts of black precipitates. Yield: ~ 0.8 mg (83.6% on Se basis). Zn14Se13(tmeda)6Cl2 cluster (4). ZnCl2(8.2 mg, 60 µmol) was suspended in 0.5 mL of CH2Cl2. Tmeda (18 µL, 120 µmol) and PPr3(24 µL, 120 µmol) were added and stirred for 20 minutes until the suspension turned into a clear colorless solution. A fresh solution of Cu26Se13(PEt2Ph)14 cluster (10 mg, 2 µmol) in 0.5 mL of CH2Cl2 was made and injected into the precursor solution at room temperature. The color of the reaction mixture gradually changed into colorless in about 20 minutes. After reacting overnight, the reaction mixture was layered with ~ 1 mL of pentane and stored at ‒20 °C. Rectangular block-shaped crystals were obtained after one day. X-ray crystallography unveiled the structure of a Zn14Se13(tmeda)6Cl2surrounded by four CH2Cl2molecules, which matches the reported structure.(6) Yield: ~ 4.5 mg (83.0% on Se basis). [Cd14Se13(tmeda)6Cl2]n cluster (5). CdCl2 (11 mg, 60 µmol) was combined with tmeda (36 µL, 240 µmol) and PPr3 (24 µL, 120 µmol) in 0.5 mL of CH2Cl2. A clear colorless solution was obtained after five minutes of stirring. Cu26Se13(PEt2Ph)14cluster (10 mg, 2 µmol) dissolved in 0.5 mL of CH2Cl2was injected. The reaction mixture changed from red to colorless in 2~3 seconds. The addition of 4 mL of pentane generated white precipitates, which were washed with 1 mL of toluene and dried under vacuum. The crude product was redissolved in 1 mL of CH2Cl2 together with the help of 50 µL tmeda. The solution was filtered and layered with 1 mL of toluene. After two days at ‒20 °C, crystals with triangular plate shape were obtained, which were composed of Cd14Se13(tmeda)6Cl2 linked into a three-dimensional network. Yield: ~5.1 mg (75.7% on Se basis). [Hg14Se13(tmeda)6Cl2]n cluster (6). HgCl2 (16.3 mg, 60 µmol) and 0.5 mL of tmeda were combined with 0.5 mL of CH2Cl2. After stirring for 20~30 minutes, a clear colorless solution was Attorney Ref. No.10457-559PC0 obtained, which was used as a stock solution. In a typical synthesis, 0.05 mL of the stock solution (6 µmol of Hg2+) and PPr3 (2.4 µL, 12 µmol) were combined with 0.85 mL of tmeda. The total volume of tmeda is 0.9 mL (6 mmol). Cu26Se13(PEt2Ph)14 cluster (1 mg, 0.2 µmol) dissolved in 0.1 mL of toluene was injected at room temperature. The color of the reaction mixture changed to bright yellow in 2~3 minutes. After one hour, the reaction mixture was filtered into an NMR tube and layered with heptane. Yellow crystals formed after 4~5 days at room temperature, which were composed of Hg14Se13(tmeda)6Cl2interconnected into a three-dimensional network. Yield: ~0.7 mg (76.1% on Se basis). [Zn14Se13(tmeda)6(CH3CN)][BPh4]2 (7). The synthesis and purification procedure were the same as (1). For crystallization, the white precipitate was redissolved in about 0.5 mL of CH3CN instead of 1.5 mL of CH2Cl2. After adding the CH3CN solution of NaBPh4and layering the mixture with heptane, small rhombic crystals came out after a couple of hours, which were identified to be [Zn14Se13(tmeda)6(CH3CN)][BPh4]2. Yield: 5.4 mg (80.4% on Se basis). [Cd14Se13(tmeda)6]0.75[Cd14Se13(tmeda)6I8]0.25 cluster (8). The synthesis was similar to (2), except CH2Cl2 was used as the reaction solvent instead of toluene. For crystallization, around 1 mL of heptane was layered on top of the reaction mixture. Colorless rhombic crystals were observed after sitting at room temperature for two days, which were identified to be [Cd14Se13(tmeda)6]0.75[Cd14Se13(tmeda)6I8]0.25. The overall composition is Cd14Se13(tmeda)6I2. Yield: 2.8 mg (39.4% on Se basis). Characterization Ultraviolet-visible (UV-Vis) absorption spectroscopy. The absorption spectra were recorded on a Cary 5000 UV–Vis−NIR spectrophotometer with a 1 nm step size. For in-situ measurements, the time interval was set to 10 minutes for the first two hours and 60 minutes for the next 14 hours. The first scan was measured around two minutes after mixing the reactant solutions. Nuclear Magnetic Resonance (NMR) spectroscopy. The1H NMR data were collected on a 400 MHz Bruker Avacne III HD spectrometer with 16 scans. Chemical shifts are given in δ (ppm) and the solvent peak was used as internal reference (δ = 7.26 ppm for CDCl3). Single-crystal X-ray diffraction (sc-XRD) Attorney Ref. No.10457-559PC0 All the data were collected at 100 K on a Bruker Dual micro source D8 Venture diffractometer with Mo Kα radiation (λ= 0.71073 Å) and PHOTON III detector. The data frames were integrated and scaled using APEX4 software package. Multi-scan method (SADABS) was used for absorption correction. Intrinsic phasing structure solution provided all the non-H atoms. The structures were refined using full-matrix least-squares refinement using SHELXL program. The non-H atoms were refined with anisotropic displacement parameters. All the H atoms were calculated in idealized positions and refined riding on their parent atoms. The refinement cycles were carried out by minimizing the wR2function using F2rather than F values. R1is calculated to provide a reference to the conventional R value but its function is not minimized. For [Zn14Se13(tmeda)6][BPh4]2 (1), a rhombic crystal specimen (approximate dimensions: 0.371 mm × 0.050 mm × 0.050 mm) was used for analysis. The nanocluster crystallizes in the space group P21 / c, with Z = 4 for the formula-unit content Zn7Se6.5N6Cl4B1C44H72. The integration of the data yielded a total of 192546 reflections to a maximum θ angle of 31.6° (0.68 Å resolution), of which 20332 were independent (completeness = 99.9%, Rint = 7.59%, Rsig = 4.57%). The final cell constants are a = 17.3474(5) Å, b = 21.3861(6) Å, c = 17.9790(5) Å, α = 90°, β = 112.5730(10)°, γ = 90°, volume = 6159.1(3) Å3. The asymmetric unit consists of half of a [Zn14Se13(tmeda)6]2+cluster around an inversion center, one [BPh4]−counterion, and two CH2Cl2 solvent molecules. In the final cycle of refinement, 20332 reflections (of which 15557 are observed with I > 2σ (I)) were used to refine 630 parameters. The final refinement on F2converged at R1= 3.99%, wR2= 8.89% and S (goodness of fit) = 1.040. For Cd14Se13(tmeda)6I2 (2), a rod-like crystal specimen (approximate dimensions: 0.176 mm × 0.104 mm × 0.045 mm) was used for analysis. The nanocluster crystallizes in the space group P1^, with Z = 2 for the formula-unit content Cd15Se13N14I4C56H128. The final cell constants are a = 15.0219(17) Å, b = 21.0500(2) Å, c = 21.3200(2) Å, α = 93.781(3)°, β = 97.483(3)°, γ = 100.579(4)°, volume = 6542.4(13) Å3. The integration of the data yielded a total of 410183 reflections to a maximum θ angle of 30.7° (0.70 Å resolution), of which 40113 were independent (completeness = 100.0%, Rint = 10.13%, Rsig = 5.39%). The asymmetric unit consists of one Cd14Se13(tmeda)6I2cluster, one CdI2(tmeda) complex, and two toluene solvent molecules. There is a positional disorder around Cd6 atom, which is refined in two parts with site occupation factors of 60% and 40%. Three of the tmeda ligands possess disordered areas around the -CH2CH2- moieties. All the disordered areas were refined in two parts, each with their site occupation factors Attorney Ref. No.10457-559PC0 fixed in the latter stages of refinement. One of the toluene molecules is also disordered. The two parts are refined with their geometry restricted to ideal hexagonal geometry (AFIX66) and the carbon atoms are restrained to idealized values (EADP, SADI). Program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent accessible void and remove its contribution to the overall intensity data. The total voids added up to a volume of 1578 Å3and a total electron count of 467 e–. In the final cycle of refinement, 40113 reflections (of which 27989 are observed with I > 2σ (I)) were used to refine 980 parameters. The final refinement on F2converged at R1 = 4.72%, wR2 = 11.05% and S (goodness of fit) = 1.033. For Hg14Se13(tmeda)6I2(3), a yellow needle-like crystal specimen (approximate dimensions: 0.107 mm × 0.060 mm × 0.053 mm) was used for analysis. The nanocluster crystallizes in the space group P212121 with Z = 4 for the formula-unit content Hg14Se13N12I2C36H96. The final cell constants are a = 15.3894(9) Å, b = 24.8074(14) Å, c = 26.3346(14) Å, α = β = γ = 90°, volume = 10053.8(10) Å3. The integration of the data yielded a total of 142799 reflections to a maximum θ angle of 28.4° (0.75 Å resolution), of which 25228 were independent (completeness = 99.9%, Rint = 7.12%, Rsig = 5.21%). The asymmetric unit consists of one Hg14Se13(tmeda)6I2 cluster. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 2358 Å3and a total electron count of 1480 e–. In the final cycle of refinement, 25228 reflections (of which 22982 are observed with I > 2σ (I)) were used to refine 718 parameters. The final refinement on F2converged at R1= 5.18%, wR2= 12.32% and S (goodness of fit) = 1.119. For Zn14Se13(tmeda)6Cl2 (4), a rectangular block crystal specimen (approximate dimensions: 0.095 mm × 0.067 mm × 0.048 mm) was used for analysis. The nanocluster crystallizes in the space group P1^, with Z = 2 for the formula-unit content Zn14Se13N12Cl10C40H104. The final cell constants are a = 13.609(3) Å, b = 13.646(3) Å, c = 27.393(6) Å, α = 90.058(7)°, β = 103.997(5)°, γ = 95.430(4)°, volume = 4912.2(18) Å3. The integration of the data yielded a total of 161107 reflections to a maximum θ angle of 28.0° (0.76 Å resolution), of which 22987 were independent (completeness = 99.9%, Rint = 22.56%, Rsig = 12.11%). The data quality was affected by the degradation of crystal during data collection. The asymmetric unit consists of a Zn14Se13(tmeda)6Cl2cluster surrounded by four CH2Cl2solvent molecules, which is possibly the reason of the crystallinity loss. Two of the Cl atoms in solvent molecules are disordered and refined Attorney Ref. No.10457-559PC0 as two parts. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 665 Å3and a total electron count of 267 e–. In the final cycle of refinement, 22987 reflections (of which 13011 are observed with I > 2σ (I)) were used to refine 844 parameters. The final refinement on F2converged at R1= 9.26%, wR2= 22.08% and S (goodness of fit) = 1.035. For [Cd14Se13(tmeda)6Cl2]n (5), a triangular crystal specimen (approximate dimensions: 0.155 mm × 0.068 mm × 0.052 mm) was used for analysis. The nanocluster crystallizes in the space group Fd3^, with Z = 16 for the formula-unit content Cd7Se6.5N6Cl1C18H48. The final cell constants are a = b = c = 29.1627(14) Å, α = β = γ = 90°, volume = 24801.81(1) Å3. The integration of the data yielded a total of 35638 reflections to a maximum θ angle of 30.9° (0.69 Å resolution), of which 3283 were independent (completeness = 99.8%, Rint = 6.19%, Rsig = 3.09%). The asymmetric unit consists of 1 / 12 of a Cd14Se13(tmeda)6Cl2 cluster owing to the high symmetry in cubic space group. Individual clusters are linked by bridging Cl atoms into a polymeric structure in the 3D space. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 10544 Å3and a total electron count of 2225 e–. In the final cycle of refinement, 3283 reflections (of which 2654 are observed with I > 2σ (I)) were used to refine 62 parameters. The final refinement on F2converged at R1= 2.29%, wR2= 6.07% and S (goodness of fit) = 0.727. For [Hg14Se13(tmeda)6Cl2]n(6), a thin needle-like crystal specimen (approximate dimensions: 0.159 mm × 0.039 mm × 0.017 mm) was used for analysis. The nanocluster crystallizes in the space group P21 / n, with Z = 4 for the formula-unit content Hg14Se13N12Cl2C36H96. The final cell constants are a = 20.7519 (7) Å, b = 25.9453 (10) Å, c = 22.7039 (8) Å, α = γ = 90°, β = 108.1230 (10)°, volume = 11617.7 (7) Å3. The integration of the data yielded a total of 303921 reflections to a maximum θ angle of 28.3° (0.75 Å resolution), of which 28967 were independent (completeness = 100%, Rint = 10.75%, Rsig = 5.63%). The asymmetric unit consists of a Hg14Se13(tmeda)6Cl2cluster, which contains one terminal chloride and two bridging chloride ligands that shared by neighboring clusters. Two organic groups in the tmeda ligands are disordered and refined as two parts. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 4403 Å3and a total electron count of 1535 e–. In the final cycle of refinement, 28967 reflections (of which 20766 are observed with I > 2σ (I)) were used to refine Attorney Ref. No.10457-559PC0 758 parameters. The final refinement on F2converged at R1= 3.31%, wR2= 7.56% and S (goodness of fit) = 1.017. For [Zn14Se13(tmeda)6(CH3CN)][BPh4]2 (7), a rhombic crystal specimen (approximate dimensions: 0.209 mm × 0.095 mm × 0.067 mm) was used for X-ray crystallographic analysis. The nanocluster crystallizes in the space group P1^, with Z = 2 for the formula-unit content Zn7Se6.5N7.5B1C45H72.5. The final cell constants are a = 13.789(7) Å, b = 14.268(7) Å, c = 17.402(8) Å, α = 101.262(13)°, β = 99.500(10)°, γ = 106.776(12)°, volume = 3124.0(3) Å3. The integration of the data yielded a total of 173286 reflections to a maximum θ angle of 30.7° (0.70 Å resolution), of which 19205 were independent (completeness = 100.0%, Rint= 8.48%, Rsig= 4.51%). The asymmetric unit consists of half of a [Zn14Se13(tmeda)6(CH3CN)]2+cluster, one BPh4–counterion and one CH3CN solvent molecule. There is positional disorder around the Zn4 atom, which is refined in two parts. Another CH3CN solvent molecule around the cluster is disordered and could not be properly refined. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 248 Å3and a total electron count of 76 e–. In the final cycle of refinement, 19205 reflections (of which 15240 are observed with I > 2σ (I)) were used to refine 642 parameters. The final refinement on F2converged at R1= 5.38%, wR2= 14.73% and S (goodness of fit) = 1.064. For [Cd14Se13(tmeda)6]0.75[Cd14Se13(tmeda)6I8]0.25 (8), a rhombic crystal specimen (approximate dimensions: 0.176 mm × 0.104 mm × 0.045 mm) was used for X-ray crystallographic analysis. The nanocluster crystallizes in the space group R3^, with Z = 6 for the formula-unit content Cd7Se6.5N6I1C18H48. The final cell constants are a = 16.0727(11) Å, b = 16.0727(11) Å, c = 35.1330(4) Å, α = 90°, β = 90°, γ = 120°, volume = 7850.1(14) Å3. The integration of the data yielded a total of 44266 reflections to a maximum θ angle of 27.5° (0.77 Å resolution), of which 3967 were independent (completeness = 98.2%, Rint = 7.36%, Rsig = 3.61%). The asymmetric unit consists of one sixth of a Cd14Se13(tmeda)6I2cluster around the inversion center and a three-fold rotation axis. Due to the positional disorder at the Cd atoms around center Se1, the whole structure can be considered as a mixture of a Cd14Se13(tmeda)6 and a Cd14Se13(tmeda)6I8 cluster with their occupation factor of 75% and 25%, respectively. The disordered areas on the tmeda ligand are refined in two parts. Program SQUEEZE was used to remove the solvent accessible voids that added up to a total volume of 1198 Å3and a total electron count of 168 e–. In the final cycle of refinement, 3967 reflections (of which 3088 are observed with I > 2σ (I)) were used to refine 980 Attorney Ref. No.10457-559PC0 parameters. The final refinement on F2converged at R1= 6.95%, wR2= 22.64% and S (goodness of fit) = 1.072. Results Transformative syntheses of atomically precise II14VI13 clusters Metal-ligand coordination complexes play crucial roles in precise cation exchange, as they not only control the reactivity of cations but also determine the stability, solubility, and surface structures of the products (30). In this example, MX2(tmeda) complexes were selected as the common carriers for IIB cations (Fig. 15A, B), as the tmeda ligand could coordinate with various IIB metal halides to form well-defined tetrahedral complexes (31). The chelating effect enhances complex stability, while the methyl groups increase their solubility in various organic solvents. MX2(tmeda) complexes were synthesized by mixing MI2 salts with tmeda ligands in CH2Cl2 solution (see SI for details).1H NMR measurements of synthesized complexes showed that the ZnI2(tmeda) exhibited the downshift among three IIB cations (Fig. 19), indicating the strongest interaction between Zn and tmeda. This is consistent with hard-soft acid-base theory (HSAB), as Zn2+is a harder acid (η = 10.88 eV) compared with Cd2+(10.29 eV) and Hg2+(7.7 eV) and should bind more tightly to hard base amine (32). The relative strength of metal-ligand interactions is also assessed by the formation constants (K) of analogous metal-ethylenediamine complexes, with KZn(3.1´1011) > KCd(4.0´1010) > KHg(5.4´109) (33). The template Cu26Se13(PEt2Ph)14cluster was prepared based on the previously reported method (30, 34). It is noted that similar to colloidal cation exchange, soft base trialkylphosphine (PR3) was added to promote the extraction of soft acid Cu+(η = 6.28 eV) and their complete exchange with harder M2+ions (27, 28, 35). Without phosphine, the Cu+to M2+exchange was sluggish, resulting in undefined aggregations. In a typical Zn2+exchange, Cu26Se13 cluster (2 μmol) dissolved in CH2Cl2 was injected into a solution containing ZnI2(tmeda) complex (60 μmol) and PPr3(120 μmol). The reaction mixture gradually changed from red to colorless within two minutes (Fig. 20), and an absorption feature at 292 nm appeared in half an hour (Fig. 15C, dashed line). A pair of distinct excitonic transitions at 288 nm (E1) and 270 nm (E2) were further resolved after purification and re-dissolving the clusters in methanol, which has a shorter solvent cut (Fig. 15C, solid line). Colorless rod-shaped crystals were obtained by layering heptane on a CH2Cl2solution Attorney Ref. No.10457-559PC0 of purified clusters together with sodium tetraphenylborate (NaBPh4, added as the bulkier counterion to facilitate crystal growth). Sc-XRD unveiled a [Zn14Se13(tmeda)6]2+cluster co- crystallized with two [BPh4]–(Fig. 21). When CdI2(tmeda) complex was used in the transformation, the red color of Cu26Se13cluster disappeared within one second, and two sharp peaks at 356 nm and 329 nm were observed (Fig. 15D, dashed line), which remained unchanged throughout the reaction (Fig. 20). The faster Cd2+exchange compared with Zn2+could be explained by the weaker Cd-tmeda bond, leading to easier dissociation and release of Cd2+. The peak positions are reminiscent of previously reported (CdSe)13 and Cd14Se13 species (17, 20). Rod-shaped crystals formed after storing the reaction mixture at –20oC for about two to three weeks, and a Cd14Se13(tmeda)6I2was resolved by sc-XRD (Fig. 21). Redissolving the crystals in toluene with the assistance of tmeda results in identical absorption peaks as the reaction mixture (Fig. 15D, solid line). However, applying similar reaction conditions to Hg2+exchange resulted in black precipitates immediately. The aggregation is consistent with previous observations for mercury chalcogenide QDs (37–39). It is hypothesized that the weaker Hg-tmeda bond led to instant ligand dissociation and aggregation of exchange clusters. Increasing tmeda concentration could reinforce their coordination with Hg2+, thus stabilizing the HgSe cluster. Indeed, when using excess tmeda as the solvent, the injection of Cu26Se13 into HgI2(tmeda) resulted in a clear yellow solution in about one minute, and a pair of peaks at 457 nm and 423 nm appeared (Fig. 15E, dashed line), which were preserved over one day (Fig.20). By layering heptane on the reaction mixture, yellow crystalline rods of Hg14Se13(tmeda)6I2cluster were obtained. The crystals can be redissolved in toluene and tmeda, which shows nearly identical peaks at 456 nm and 423 nm (Fig. 15E, solid line). Characterization of Atomic structures of the child II14VI13 clusters The total structures of parent and child clusters are compared in Fig. 16. Significantly, all the child clusters are constructed based on a common A14B13 skeleton, with 13 anions packed into a highly symmetric icosahedron, 6 edge bridging cations (eM), and the remaining 8 cations distributed as core (cM) or face capping (fM) cations. The Se13 icosahedral lattices of child clusters are nearly identical to that of the parent Cu26Se13 cluster with slight variation in sizes (Fig. 16A, B). Based on the Se-Se distance measurements in Cu26Se13(4.31 ± 0.17 Å), contraction was found for Zn14Se13(4.12 ± 0.11 Å), while expansion was noted for Cd14Se13(4.43 ± 0.17 Å) and Hg14Se13 Attorney Ref. No.10457-559PC0 (4.47 ± 0.16 Å, Fig.22). The trend is comparable to that observed in bulk materials, which can be attributed to the size difference of cations. In addition, all the cluster frameworks are marginally expanded by ~1% to ~ 4% compared to their bulk lattices. While the cation positions are ostensibly intricate, careful analysis shows a shell-by-shell assembly of cations into symmetric polyhedrons (Fig.16C). In general, cations are bonded to Se13 icosahedron (Ih) in three modes: (i) core cation inside the Ih shell (cM, Fig.16D); (ii) face-capping cation connected to three Se on Ihfacet (fM-Se3, Fig. 16E); and (iii) edge-bridging cation bonded to two Se at the Ih edge (eM-Se2, Fig. 16F). In Cu26Se13 cluster, twelvecCu form an inner Ih shell orthogonal to the anion Ih; eightfCu form a distorted cube; and sixeCu are arranged into a distorted octahedron (Fig. 2C-F, blue). EachfCu8andeCu6cation is further coordinated to a PEt2Ph ligand to form a complete cluster. In all child clusters, six cations are located at the same edge-bridging sites (eM) as the parent cluster (Fig. 16F). EacheM is further chelated by a tmeda ligand to form aeMSe2(tmeda) surface motif. The saturated tetrahedral coordination resembles that of parent MI2(tmeda) complexes, which further verifies cation exchange is accompanied by surface ligand exchange and that the surface structures of the child clusters are dictated by the parent coordination complex (30). The positions of the remaining eight cations vary slightly among child clusters (Fig. 16D, E). In Hg14Se13, the eight cations are distributed as four corecHg, two face-cappingfHgSe3I with saturated coordination, and two unsaturatedf’HgSe3. The unsaturatedf’Hg is closer to the central Se (3.47 ± 0.03 Å) compared to the saturatedfHg (4.03 ± 0.01 Å). The Cd14Se13cluster resembles Hg14Se13, except the two unsaturatedf’Cd are pushed further inward (Fig. 16E). ThecCd4,f’Cd2, andfCd2 distances to central Se are 2.75 ± 0.01, 3.01 ± 0.01, 4.17 ± 0.02 Å, respectively. In Zn14Se13, thefZn4 andcZn4 are equivalent, forming a cubiccZn8 shell surrounding the central Se (Fig. 16C-E). They are closer to the shell Se (2.42 ± 0.01 Å) than the central Se (2.80 ± 0.05 Å). The absence of iodides on ZnSe cluster could be attributed to the unfavored bonding between hard Zn2+and soft I–(32). In all clusters, the cations bind more tightly to the shell Se than the central Se. The bond lengths with shell Se follow the trend ofcM >fM >f’M >eM. Structural diversity in II14VI13 clusters While the Se13framework andeM6surface motifs are consistent, the remaining eight cations are quite dynamic and vary with synthetic and crystallization conditions (Fig. 17). For Attorney Ref. No.10457-559PC0 example, when ZnCl2was used in the transformation (Fig. 17A and Fig. 23), a neutral Zn14Se13(tmeda)6Cl2 cluster was obtained, similar to the previously reported structure (15). The cluster shows the same symmetry as Cd14Se13(tmeda)6I2, with two chlorides pulling two Zn out of the Se shell to form face-cappingfZnSe3Cl sites. The use of CdCl2results in a highly symmetric network of Cd14Se13(tmeda)6Cl2 cluster (Fig. 17B and Fig.24). The eight Cd split evenly to form innercCd4 and outerfCd4 tetrahedral shells. Such cation arrangement is nearly identical to the Cd14Se13fragment in the Cd26Se17cluster (30). ThefCd is each bonded to a Cl–. Through the linear Cd-Cl-Cd bridging bond, each cluster is further connected to four neighboring ones, forming infinite diamond superlattices (11, 12). When HgCl2 was used, a polymeric chain of Hg14Se13(tmeda)6Cl2cluster was obtained (Fig. 17C and Fig. 25). ThreefHg are coordinated to Cl–, and onef’Hg is undercoordinated. The clusters are linked into helical chains via the two bridging chlorides (Fig. 25). Solvents also play a critical role in positioning the eight cations. When CH3CN was used in the crystallization of ZnSe clusters, two CH3CN molecules were found to coordinate and pull out thefZn in the body diagonal direction (Fig. 26). The occupancy of thefZn-CH3CN moieties is refined to 50% in XRD analysis, resulting in an overall formula of [Zn14Se13(tmeda)6(CH3CN)]2+(Fig.27). When the iodide-capped CdSe cluster was synthesized in CH2Cl2instead of toluene, the resulting crystal can be viewed as a mixture of two clusters with 75% occupancy for [Cd14Se13(tmeda)6]2+and 25% occupancy for [Cd14Se13(tmeda)6I8]6–, comprising the overall composition of Cd14Se13(tmeda)6I2(Fig. 27, 28). The polar CH2Cl2could help shuffle the iodides among clusters during synthesis and crystallization. Significantly, despite the differences in cation positions, cluster charges, and coordinating halides, the absorption spectra remain nearly the same (Fig. 28). The peaks of neutral Zn14Se13(tmeda)6Cl2cluster (289 nm / 268 nm) are close to the positively charged cluster at 288 nm / 270 nm (Fig. 17D). Minimal blueshifts of both peaks were observed in chloride-capped Cd14Se13 and Hg14Se13 cluster compared to the iodide-coordinated ones (Fig. 17E, F). Correlation of optical properties and electronic structures of II14VI13 artificial atoms The sequence of homolog II-VI clusters enables the study of the precise structure-property correlation and the “periodic” rules in these artificial atoms. Notably, the Zn14Se13, Cd14Se13, and Hg14Se13clusters all feature a “doublet” of absorption peaks with similar energy of 0.287, 0.286, and 0.212 eV, respectively (Table 1). The peaks redshift as the atomic number of the cation Attorney Ref. No.10457-559PC0 increases (Fig. 18A). A nearly linear fitting curve was observed when plotting E1(or E2) peaks against the optical gaps (Eg) of corresponding bulk materials (Fig. 18B). (41, 42) Similar linear relationship of E1 ~ Eg was documented for a series of II-VI MSCs (22, 43). While the atomic structures of those MSCs are unknown, the resemblance in fittings indicates those clusters could share the same core structures with II14VI13 clusters (Fig. 29). Table 1. The optical characteristics of the II14VI13 clusters (1-3) Zn14Se13 Cd14Se13 Hg14Se13 9 1 3 1 Conclusion In conclusion, Example 5 builds on the findings in Examples 1-4 and validates the generalization of transformation synthesis to access a library of atomically defined II-VI artificial atoms with atomic engineering over their cations and surface ligands. By using universal MX2(tmeda) coordination complexes as precise cation carriers, a family of homologous Zn14Se13, Cd14Se13, and Hg14Se13 clusters were obtained from the parent Cu26Se13 cluster. The shared A13B14 skeleton with nearly spherical shape, ensemble icosahedral packing and local tetrahedral coordination could be a general blueprint for constructing various semiconductor artificial atoms, akin to the zincblende and wurtzite as the common phases for bulk compound semiconductors. All the child clusters are characterized by a “doublet” of excitonic transitions which redshifts as the cation size increases. The generalization of the transformative synthesis is an important step towards atomic engineering of artificial atoms and expanding their diversity for various applications in energy, information, and biomedical nanotechnologies. Attorney Ref. No.10457-559PC0 References for Example 5 1. R. C. Ashoori, Electrons in artificial atoms. Nature 379, 413–419 (1996). 2. M. A. Kastner, Artificial Atoms. Phys Today 46, 24–31 (1993). 3. A. L. Efros, L. E. Brus, Nanocrystal Quantum Dots: From Discovery to Modern Development. ACS Nano15, 6192–6210 (2021).4. C. B. Murray, C. R. Kagan, M. G. Bawendi, Self-organization of CdSe nanocrystallites into three- dimensional quantum dot superlattices. Science (1979) 270, 1335 (1995). 5. E. V. Shevchenko, D. V. Talapin, N. A. Kotov, S. O’Brien, C. B. Murray, Structural diversity in binary nanoparticle superlattices. Nature 439, 55–59 (2006). 6. C. R. Kagan, E. Lifshitz, E. H. Sargent, D. V. Talapin, Building devices from colloidal quantum dots. Science 353, 885–892 (2016). 7. C. Zeng, Y. Chen, K. Kirschbaum, K. J. Lambright, R. Jin, Emergence of hierarchical structural complexities in nanoparticles and their assembly. Science 354, 1580–1584 (2016). 8. C. Zeng, Precision at the nanoscale: on the structure and property evolution of gold nanoclusters. Pure Appl.Chem. 90, 1409–1428 (2018).9. I. G. Dance, A. Choy, M. L. Scudder, Syntheses, properties, and molecular and crystal structures of (Me4N)4[E4M10(SPh)16] (E = sulfur or selenium; M = zinc or cadmium): molecular supertetrahedral fragments of the cubic metal chalcogenide lattice. J Am Chem Soc 106, 6285–6295 (1984). 10. N. Herron, J. C. Calabrese, W. E. Farneth, Y. Wang, Crystal structure and optical properties of Cd32S14(SC6H5)36. DMF4, a cluster with a 15 angstrom CdS core. Science 259, 1426–1428 (1993). 11. T. Vossmeyer, G. Reck, L. Katsikas, E. T. K. Haupt, B. Schulz, H. Weller, A “Double-Diamond Superlattice” Built Up of Cd 17 S 4 (SCH 2 CH 2 OH) 26 Clusters. Science 267, 1476–1479 (1995). 12. N. Zheng, X. Bu, H. Lu, Q. Zhang, P. Feng, Crystalline Superlattices from Single-Sized Quantum Dots. J Am Chem Soc 127, 11963–11965 (2005). 13. V. N. Soloviev, A. Eichhöfer, D. Fenske, U. Banin, Size-dependent optical spectroscopy of a homologous series of CdSe cluster molecules. J. Am. Chem. Soc.123, 2354–2364 (2001). 14. A. N. Beecher, X. Yang, J. H. Palmer, A. L. Lagrassa, P. Juhas, S. J. L. Billinge, J. S. Owen, Atomic structures and gram scale synthesis of three tetrahedral quantum dots. J. Am. Chem. Soc.136, 10645–10653 (2014). 15. C. B. Khadka, A. Eichhöfer, F. Weigend, J. F. Corrigan, Zinc Chalcogenolate Complexes as Precursors to ZnE and Mn / ZnE (E = S, Se) Clusters. Inorg. Chem.51, 2747–2756 (2012). 16. D. C. Gary, S. E. Flowers, W. Kaminsky, A. Petrone, X. Li, B. M. Cossairt, Single-crystal and electronic structure of a 1.3 nm indium phosphide nanocluster. J. Am. Chem. Soc.138, 1510–1513 (2016). 17. M. S. Bootharaju, W. Baek, G. Deng, K. Singh, O. Voznyy, N. Zheng, T. Hyeon, Structure of a subnanometer-sized semiconductor Cd14Se13 cluster. Chem 8, 2978–2989 (2022).18. A. Kasuya, R. Sivamohan, Y. A. Barnakov, I. M. Dmitruk, T. Nirasawa, V. R. Romanyuk, V. Kumar, S. VMamykin, K. Tohji, B. Jeyadevan, K. Shinoda, T. Kudo, O. Terasaki, Z. Liu, R. V Belosludov, V. Sundararajan, Y. Kawazoe, Ultra-stable nanoparticles of CdSe revealed from mass spectrometry. Nat. Mater.3, 99–102 (2004). 19. S. Kudera, M. Zanella, C. Giannini, A. Rizzo, Y. Li, G. Gigli, R. Cingolani, G. Ciccarella, W. Spahl, W. J. Parak, L. Manna, Sequential growth of magic-size CdSe nanocrystals. Adv. Mater.19, 548–552 (2007). 20. Y. Wang, Y.-H. Liu, Y. Zhang, F. Wang, P. J. Kowalski, H. W. Rohrs, R. A. Loomis, M. L. Gross, W. E. Buhro, Isolation of the Magic-Size CdSe Nanoclusters [(CdSe)13(n-octylamine)13] and [(CdSe)13(oleylamine)13]. Angew. Chem. Int. Ed.51, 6154–6157 (2012). 21. A. S. Mule, S. Mazzotti, A. A. Rossinelli, M. Aellen, P. T. Prins, J. C. van der Bok, S. F. Solari, Y. M. Glauser, P. V Kumar, A. Riedinger, D. J. Norris, Unraveling the Growth Mechanism of Magic-Sized SemiconductorNanocrystals. J. Am. Chem. Soc. 143, 2037–2048 (2021).22. Y. Zhou, R. Jiang, Y. Wang, H. W. Rohrs, N. P. Rath, W. E. Buhro, Isolation of Amine Derivatives of (ZnSe)34 and (CdTe)34. Spectroscopic Comparisons of the (II–VI)13 and (II–VI)34 Magic-Size Nanoclusters. Inorg. Chem.58, 1815–1825 (2019). 23. C. B. Williamson, D. R. Nevers, A. Nelson, I. Hadar, U. Banin, T. Hanrath, R. D. Robinson, Chemically reversible isomerization of inorganic clusters. Science 363, 731–735 (2019). 24. C. B. Murray, D. J. Norris, M. G. Bawendi, Synthesis and characterization of nearly monodisperse CdE (E = sulfur, selenium, tellurium) semiconductor nanocrystallites. J. Am. Chem. Soc.115, 8706–8715 (1993). 25. Z. A. Peng, X. Peng, Formation of high-quality CdTe, CdSe, and CdS nanocrystals using CdO as precursor. J. Am. Chem. Soc.123, 183–184 (2001). Attorney Ref. No.10457-559PC0 26. R. García-Rodríguez, M. P. Hendricks, B. M. Cossairt, H. Liu, J. S. Owen, Conversion reactions of cadmium chalcogenide nanocrystal precursors. Chem. Mater.25, 1233–1249 (2013). 27. D. H. Son, S. M. Hughes, Y. Yin, A. P. Alivisatos, Cation exchange reactions in ionic nanocrystals. Science 306, 1009–1012 (2004). 28. L. De Trizio, L. Manna, Forging colloidal nanostructures via cation exchange reactions. Chem. Rev. 116, 10852–10887 (2016). 29. J. L. Fenton, B. C. Steimle, R. E. Schaak, Tunable intraparticle frameworks for creating complex heterostructured nanoparticle libraries. Science 360, 513–517 (2018). 30. F. Ma, K. A. Abboud, C. Zeng, Precision synthesis of a CdSe semiconductor nanocluster via cation exchange. Nat. Synth.2, 949–959 (2023). 31. S. Htoon, M. F. C. Ladd, Crystallographic studies of dihalogenotetramethylethylenediamine complexes of zinc, cadmium, and mercury. IV: General crystallography and molecular geometry. J. Cryst. Mol. Struct. 6, 55–58 (1976). 32. R. G. Pearson, Absolute Electronegativity and Hardness: Application to Inorganic Chemistry. Inorg Chem 27, 734–740 (1988). 33. P. Paoletti, International union of pure and applied chemistry analytical chemistry division commission on equilibrium data formation of metal complexes with ethylenediamine: A critical survey of equilibrium constants, enthalpy and entropy values. Pure and Applied Chemistry 56, 491–522 (1984). 34. A. Deveson, S. Dehnen, D. Fenske, Syntheses and structures of four new copper(I)-selenium clusters: size dependence of the cluster on the reaction conditions. J. chem. Soc., Dalton Trans.23, 4491–4497 (1997). 35. J. Gui, M. Ji, J. Liu, M. Xu, J. Zhang, H. Zhu, Phosphine-initiated cation exchange for precisely tailoring composition and properties of semiconductor nanostructures: old concept, new applications. Angew. Chem. Int. Ed. 54, 3683–3687 (2015). 36. L. R. Snyder, J. J. Kirkland, J. W. Dolan, “Appendix I: Properties of HPLC Solvents” in Introduction to Modern Liquid Chromatography (2010 John Wiley & Sons, Inc., 2009), pp.879–886. 37. C. Gréboval, A. Chu, N. Goubet, C. Livache, S. Ithurria, E. Lhuillier, Mercury Chalcogenide Quantum Dots: Material Perspective for Device Integration. Chem. Rev.121, 3627–3700 (2021). 38. A. M. Smith, S. Nie, Bright and Compact Alloyed Quantum Dots with Broadly Tunable Near-Infrared Absorption and Fluorescence Spectra through Mercury Cation Exchange. J. Am. Chem. Soc.133, 24–26 (2011). 39. S. Keuleyan, E. Lhuillier, V. Brajuskovic, P. Guyot-Sionnest, Mid-infrared HgTe colloidal quantum dot photodetectors. Nat. Photonics 5, 489–493 (2011). 40. J. A. Dean, Ed., Lange’s Handbook of Chemistry (McGraw-Hill, Inc., New York, ed.15th, 1999). 41. O. Madelung, Ed., Semiconductors — Basic Data (Springer Berlin Heidelberg, Berlin, Heidelberg, 1996). 42. K.-U. Gawlik, L. Kipp, M. Skibowski, N. Orłowski, R. Manzke, HgSe: Metal or Semiconductor? Phys. Rev. Lett.78, 3165–3168 (1997). 43. Y. Wang, Y. Zhou, Y. Zhang, W. E. Buhro, Magic-Size II–VI Nanoclusters as Synthons for Flat Colloidal Nanocrystals. Inorg. Chem.54, 1165–1177 (2015).

[0002] Attorney Ref. No.10457-559PC0 General References 1. Kagan, C. R., Bassett, L. C., Murray, C. B. & Thompson, S. M. Colloidal quantum dots as platforms for quantum information science. Chem. Rev.121, 3186–3233 (2021).2. Kagan, C. R., Lifshitz, E., Sargent, E. H. & Talapin, D. v. Building devices from colloidal quantum dots.Science 353, 885–892 (2016). 3. Murray, C. B., Norris, D. J. & Bawendi, M. G. Synthesis and characterization of nearly monodisperse CdE (E = sulfur, selenium, tellurium) semiconductor nanocrystallites. J. Am. Chem. Soc.115, 8706–8715 (1993). 4. Peng, Z. A. & Peng, X. Formation of high-quality CdTe, CdSe, and CdS nanocrystals using CdO as precursor. J. Am. Chem. Soc.123, 183–184 (2001). 5. Efros, A. L. & Brus, L. E. Nanocrystal Quantum Dots: From Discovery to Modern Development. ACS Nano 15, 6192–6210 (2021). 6. Boles, M. A., Ling, D., Hyeon, T. & Talapin, D. v. The surface science of nanocrystals. Nat. Mater.15, 141– 153 (2016). 7. Kagan, C. R. & Murray, C. B. Charge transport in strongly coupled quantum dot solids. Nat. Nanotechnol.10, 1013–1026 (2015). 8. Yu, W. W., Qu, L., Guo, W. & Peng, X. Experimental Determination of the Extinction Coefficient of CdTe, CdSe, and CdS Nanocrystals. Chem. Mater.15, 2854–2860 (2003). 9. Jin, R., Zeng, C., Zhou, M. & Chen, Y. Atomically precise colloidal metal nanoclusters and nanoparticles: fundamentals and opportunities. Chem. Rev.116, 10346–10413 (2016). 10. Zeng, C., Chen, Y., Kirschbaum, K., Lambright, K. J. & Jin, R. Emergence of hierarchical structural complexities in nanoparticles and their assembly. Science 354, 1580–1584 (2016). 11. Tran, N. T., Powell, D. R. & Dahl, L. F. Nanosized Pd145(CO)x(PEt3)30containing a capped three-shell 145- atom metal-core geometry of pseudo icosahedral symmetry. Angew. Chem. Int. Ed.39, 4121–4125 (2000).12. Jadzinsky, P. D., Calero, G., Ackerson, C. J., Bushnell, D. A. & Kornberg, R. D. Structure of a thiolmonolayer-protected gold nanoparticle at 1.1 Å resolution. Science 318, 430–433 (2007). 13. 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). 14. Owen, J. The coordination chemistry of nanocrystal surfaces. Science 347, 615–616 (2015). 15. Jasieniak, J. & Mulvaney, P. From Cd-Rich to Se-Rich − the Manipulation of CdSe Nanocrystal Surface Stoichiometry. J. Am. Chem. Soc.129, 2841–2848 (2007). 16. Wei, H. H.-Y. et al. Colloidal Semiconductor Quantum Dots with Tunable Surface Composition. Nano Lett. 12, 4465–4471 (2012). 17. B., W. C. et al. Chemically reversible isomerization of inorganic clusters. Science 363, 731–735 (2019).18. Lee, G. S. H. et al. [S4Cd17(SPh)28]2-, the first member of a third series of tetrahedral [SWMX(SR)y]z- clusters.J. Am. Chem. Soc.110, 4863–4864 (1988). 19. Herron, N., Calabrese, J. C., Farneth, W. E. & Wang, Y. Crystal structure and optical properties of Cd32S14(SC6H5)36. DMF4, a cluster with a 15 angstrom CdS core. Science 259, 1426–1428 (1993). 20. Vossmeyer, T., Reck, G., Schulz, B., Katsikas, L. & Weller, H. Double-layer superlattice structure built up of Cd32S14(SCH2CH(OH)CH3)36.4H2O clusters. J. Am. Chem. Soc.117, 12881–12882 (1995). 21. Soloviev, V. N., Eichhöfer, A., Fenske, D. & Banin, U. Size-dependent optical spectroscopy of a homologous series of CdSe cluster molecules. J. Am. Chem. Soc.123, 2354–2364 (2001). 22. Beecher, A. N. et al. Atomic structures and gram scale synthesis of three tetrahedral quantum dots. J. Am. Chem. Soc.136, 10645–10653 (2014).23. Gary, D. C. et al. Single-crystal and electronic structure of a 1.3 nm indium phosphide nanocluster. J. Am.Chem. Soc.138, 1510–1513 (2016). 24. Zeng, C. Precision at the nanoscale: on the structure and property evolution of gold nanoclusters. Pure Appl. Chem.90, 1409–1428 (2018). 25. Anderson, P. W. More is different. Science 177, 393–396 (1972). 26. Yin, Y. & Alivisatos, A. P. Colloidal nanocrystal synthesis and the organic-inorganic interface. Nature 437, 664–670 (2005). 27. Silvera Batista, C. A., Larson, R. G. & Kotov, N. A. Nonadditivity of nanoparticle interactions. Science 350, (2015). Attorney Ref. No.10457-559PC0 28. de Trizio, L. & Manna, L. Forging colloidal nanostructures via cation exchange reactions. Chem. Rev. 116, 10852–10887 (2016). 29. Rivest, J. B. & Jain, P. K. Cation exchange on the nanoscale: an emerging technique for new material synthesis, device fabrication, and chemical sensing. Chem. Soc. Rev.42, 89–96 (2013). 30. Beberwyck, B. J., Surendranath, Y. & Alivisatos, A. P. Cation exchange: A versatile tool for nanomaterials synthesis. J. Phys. Chem. C 117, 19759–19770 (2013). 31. Gupta, S., Kershaw, S. v. & Rogach, A. L.25th Anniversary Article: Ion Exchange in Colloidal Nanocrystals. Adv. Mater.25, 6923–6944 (2013). 32. Zeng, C., Chen, Y., Das, A. & Jin, R. Transformation Chemistry of Gold Nanoclusters: From One Stable Size to Another. J. Phys. Chem. Lett.6, 2976–2986 (2015). 33. Zeng, C., Liu, C., Pei, Y. & Jin, R. Thiol Ligand-Induced Transformation of Au38(SC2H4Ph)24to Au36(SPh-t- Bu)24. ACS Nano 7, 6138–6145 (2013). 34. Singer, J. & Faeth, P. A. The orientation of Cu2S formed on single crystal CdS. App. Phys. Lett.11, 130–132 (1967). 35. Cook, W. R., Shiozawa, L. & Augustine, F. Relationship of copper sulfide and cadmium sulfide phases. J. Appl. Phys.41, 3058–3063 (1970). 36. Szeto, W. & Somorjai, G. A. Optical Study of Copper Diffusion in Cadmium Sulfide Single Crystals. J. Chem. Phys.44, 3490–3495 (1966). 37. Son, D. H., Hughes, S. M., Yin, Y. & Alivisatos, A. P. Cation exchange reactions in ionic nanocrystals. Science 306, 1009–1012 (2004). 38. Mews, A., Eychmueller, A., Giersig, M., Schooss, D. & Weller, H. Preparation, characterization, and photophysics of the quantum dot quantum well system cadmium sulfide / mercury sulfide / cadmium sulfide. J. Phys. Chem.98, 934–941 (1994). 39. Camargo, P. H. C., Lee, Y. H., Jeong, U., Zou, Z. & Xia, Y. Cation Exchange: A Simple and Versatile Route to Inorganic Colloidal Spheres with the Same Size but Different Compositions and Properties. Langmuir 23, 2985–2992 (2007). 40. Luther, J. M., Zheng, H., Sadtler, B. & Alivisatos, A. P. Synthesis of PbS nanorods and other ionic nanocrystals of complex morphology by sequential cation exchange reactions. J. Am. Chem. Soc.131, 16851– 16857 (2009). 41. Li, H. et al. Sequential cation exchange in nanocrystals: Preservation of crystal phase and formation of metastable phases. Nano Lett.11, 4964–4970 (2011). 42. Ha, D.-H. et al. Solid–Solid phase transformations induced through cation exchange and strain in 2D heterostructured copper sulfide nanocrystals. Nano Lett.14, 7090–7099 (2014). 43. Mu, L., Wang, F., Sadtler, B., Loomis, R. A. & Buhro, W. E. Influence of the Nanoscale Kirkendall Effect on the Morphology of Copper Indium Disulfide Nanoplatelets Synthesized by Ion Exchange. ACS Nano 9, 7419– 7428 (2015). 44. Fenton, J. L., Steimle, B. C. & Schaak, R. E. Tunable intraparticle frameworks for creating complex heterostructured nanoparticle libraries. Science 360, 513–517 (2018). 45. Steimle, B. C., Fenton, J. L. & Schaak, R. E. Rational construction of a scalable heterostructured nanorod megalibrary. Science 367, 418–424 (2020). 46. Zhao, Q. et al. Enhanced carrier transport in strongly coupled, epitaxially fused CdSe nanocrystal solids. Nano Lett.21, 3318–3324 (2021). 47. Stein, J. L. et al. Cation Exchange Induced Transformation of InP Magic-Sized Clusters. Chem. Mater. 29, 7984–7992 (2017). 48. White, S. L., Banerjee, P., Chakraborty, I. & Jain, P. K. Ion exchange transformation of magic-sized clusters. Chem. Mater.28, 8391–8398 (2016). 49. Deveson, A., Dehnen, S. & Fenske, D. Syntheses and structures of four new copper(I)–selenium clusters: size dependence of the cluster on the reaction conditions. J. Chem. Soc., Dalton Trans.4491–4498 (1997). 50. Gui, J. et al. Phosphine-initiated cation exchange for precisely tailoring composition and properties of semiconductor nanostructures: old concept, new applications. Angew. Chem. Int. Ed.54, 3683–3687 (2015). 51. Fuhr, O., Dehnen, S. & Fenske, D. Chalcogenide clusters of copper and silver from silylated chalcogenide sources. Chem. Soc. Rev.42, 1871–1906 (2013). 52. Luther, J. M., Jain, P. K., Ewers, T. & Alivisatos, A. P. Localized surface plasmon resonances arising from free carriers in doped quantum dots. Nat. Mater.10, 361–366 (2011). Attorney Ref. No.10457-559PC0 53. Hagen, K. S., Stephan, D. W. & Holm, R. H. Metal ion exchange reactions in cage molecules: the systems [M4-nM’n(SC6H5)10]2-(M, M’ = Fe(II), Co(II), Zn(II), Cd(II)) with adamantane-like stereochemistry and the structure of [Fe4(SC6H5)10]2-. Inorg. Chem.21, 3928–3936 (1982). 54. Zeng, C. et al. Controlling magnetism of Au133(TBBT)52 nanoclusters at single electron level and implication for nonmetal to metal transition. Chem. Sci.10, 9684–9691 (2019). 55. García-Rodríguez, R. & Liu, H. Solution structure of cadmium carboxylate and its implications for the synthesis of cadmium chalcogenide nanocrystals. Chem. Commun.49, 7857 (2013). 56. Kessler, J. M. et al. Comparison of solid-state and solution structures of (R3P)2CdX2, (Et3P)2Cd2X4 and (Bu3P)3Cd2X4complexes. Magn. Reson. Chem.29, S94–S105 (1991). 57. Kudera, S. et al. Sequential growth of magic-size CdSe nanocrystals. Adv. Mater.19, 548–552 (2007). 58. Kasuya, A. et al. Ultra-stable nanoparticles of CdSe revealed from mass spectrometry. Nat. Mater.3, 99–102 (2004). 59. Chen, O. et al. Surface-Functionalization-Dependent Optical Properties of II–VI Semiconductor Nanocrystals. J. Am. Chem. Soc.133, 17504–17512 (2011). 60. Bootharaju, M. S. et al. Structure of a subnanometer-sized semiconductor Cd14Se13cluster. Chem 8, 2978– 2989 (2022). 61. Khadka, C. B., Eichhöfer, A., Weigend, F. & Corrigan, J. F. Zinc Chalcogenolate Complexes as Precursors to ZnE and Mn / ZnE (E = S, Se) Clusters. Inorg. Chem.51, 2747–2756 (2012). 62. Wang, Y. et al. Isolation of the Magic-Size CdSe Nanoclusters [(CdSe)13(n-octylamine)13] and [(CdSe)13(oleylamine)13]. Angew. Chem. Int. Ed.51, 6154–6157 (2012). 63. del Ben, M., Havenith, R. W. A., Broer, R. & Stener, M. Density Functional Study on the Morphology and Photoabsorption of CdSe Nanoclusters. J. Phys. Chem. C 115, 16782–16796 (2011). 64. Shiang, J. J., Kadavanich, A. v., Grubbs, R. K. & Alivisatos, A. P. Symmetry of annealed wurtzite CdSe nanocrystals: assignment to the C3v point group. J. Phys. Chem.99, 17417–17422 (1995). 65. Nevers, D. R., Williamson, C. B., Hanrath, T. & Robinson, R. D. Surface chemistry of cadmium sulfide magic- sized clusters: a window into ligand-nanoparticle interactions. Chem. Commun.53, 2866–2869 (2017). 66. Mocatta, D. et al. Heavily doped semiconductor nanocrystal quantum dots. Science 332, 77–81 (2011). 67. Smith, A. M., Lane, L. A. & Nie, S. Mapping the spatial distribution of charge carriers in quantum-confined heterostructures. Nat. Commun.5, 4506 (2014). 68. Lange’s Handbook of Chemistry. (McGraw-Hill, Inc., 1999). 69. Goodsell, D. S. The Machinery of Life. (Springer New York, 2009). 70. Detty, M. R. & Seidler, M. D. Bis(trialkylsilyl) chalcogenides. 1. Preparation and reduction of group VIA oxides. J. Org. Chem.47, 1354–1356 (1982). 71. J. A. Dean, Ed., Lange’s Handbook of Chemistry (McGraw-Hill, Inc., New York, ed.15th, 1999).

Claims

Attorney Ref. No.10457-559PC0 CLAIMS What is claimed is:

1. A method for precision synthesis of a reaction product nanocluster, wherein the reaction product is produced by reacting a template cluster and a coordination complex according to the following Cat1-An1-L1+ Cat2-X2-L2+ L3^ Cat2-An1-X2-L2wherein the First reagent, Cat1-An1-L1, is a template nanocluster wherein the Second reagent, Cat2-X2-L2, is a metal-ligand coordination complex wherein L3 is an optional reagent and is a soft-base ligand (such as phosphine), which is optionally included in the reaction to facilitate extraction of a soft cation (such as Cu+) when the soft-base ligand is not present in the Cat2-X2-L2coordination complex, and wherein Cat2-An1-X2-L2is a product nanocluster.

2. The method of claim 1, , wherein Cat1is Cu(I) or Ag(I), An1is selected from S, Se, or Te, and L1is a surface ligand, wherein Cat2is selected from Cd2+, Zn2+, Hg2+, Pb2+, or other transitional metal cations, X2is selected from halides (Cl–, Br–, I–), carboxylates (–O2CR, R is R being an alkyl, alkenyl or aryl group), or other anionic ligands, and L2is a neutral surface ligand.

3. The method of claim 1, wherein L1is the original ligand that is carried in the template nanocluster, wherein L1is, optionally, PEt2Ph.

4. The method of claim 1, wherein L2is selected from a phosphine, amine, carbene, alcohol, or other neutral ligands, wherein L2is, optionally, is PPr3, NR3(R is various CxHygroup, including H), multidentate ligands (R2N-CxHy-NR2), or optionally, (CH3)2NCH2CH2N(CH3)2.

5. The method of claim 1, wherein X2is selected from a halide, carboxylate, or other negatively charged ligands, wherein X2is, optionally, Cl–, Br–, I–,–O2CR,–SR (R is an alkyl, alkenyl or aryl group).Attorney Ref. No.10457-559PC0 6. The method of any of claims 1-5, wherein the template nanocluster is Cu26Se13(PEt2Ph)14.

7. The method of any of claims 1-6, wherein the coordination complex is CdI2(PPr3)2.

8. The method of claim 7, wherein the nanocluster reaction product is Cd26Se17I18(PPr3)10.

9. The method of any of claims 1-8, wherein the template nanocluster and / or coordination complex and / or nanocluster reaction product comprise at least one dopant.

10. The method of claim 8, wherein the dopant comprises Mn2+.

11. A nanocluster comprising Cat2-An1-X2-L2, wherein Cat2is selected from Cd2+, Zn2+, Hg2+, Pb2+or other transitional metal cations, wherein An1is selected from S, Se, or Te, wherein X2is selected from halides (Cl–, Br–, I–), carboxylates (–O2CR, R is various CxHy group), or other anionic ligands, and wherein L2is a neutral surface ligand.

12. The nanocluster of claim 11, wherein L2is selected from a phosphine, amine, carbene, alcohol, or other neutral ligands, wherein L2is, optionally, is PPr3, NR3(R is various CxHygroup, including H), multidentate ligands (R2N-CxHy-NR2).

13. The nanocluster of claims 11 or 12, wherein the nanocluster comprises a size of 0.5- 5 nm.

14. The nanocluster of any of claims 11-13, wherein the nanocluster comprises optionally a dopant cation.

15. The nanocluster of claim 14, wherein the dopant comprises Mn2+.

16. The nanocluster of any of claims 11-13, wherein the nanocluster is comprised of Cd26Se17I18(PPr3)10, Zn14Se13L6, Cd14Se13L6, Hg14Se13L6, where L is (CH3)2NCH2CH2N(CH3)2.Attorney Ref. No.10457-559PC0 17. The nanocluster of any of claims 11-16 comprising a shape selected from tetrahedron, sphere, ellipsoidal, or triangular plate.

18. A nanocluster produced by the method of any of claims 1-10.

19. A solar cell comprising a plurality of nanoclusters according to any of claims 11-17.

20. A light-emitting diode comprising a plurality of nanoclusters according to any of claims 11-17.

Citation Information

Patent Citations

  • Gold-copper alloy nanocluster and preparation method and application thereof

    CN116037951A

  • Preparation of Nanoparticle Materials

    US20190161677A1

  • KR20230034108A