Postassembly Modification of Interior Spaces Within M₁₂L₂₄ Nanospheres

Abstract

Post assembly modification (PAM) of coordination‐based supramolecular self‐assemblies has developed in the past 10 years as a tool that allows expansion of the chemical diversity available in supramolecular synthesis. In this contribution we apply PAM to enable the creation of M₁₂L₂₄ nanospheres with unique interior spaces. We introduce alkyl N‐pyridinium carboxy aldehydes (ANCA) as ideal orthogonal functional group for PAM. The herein presented nanospheres undergo quantitative PAM with a variety of aniline derivatives, allowing the creation of hydrophobic, chiral, catalytic, and other confined spaces. As the parent spheres can be easily characterized before PAM, this methodology not only allows for the preparation of highly desirable multifunctional confined spaces but also serves with its post modulation possibilities as a fast and efficient tool for screening secondary interactions and confined spaces on probes of interest.

We described a procedure for the quantitative postassembly modification of the confined space within M₁₂L₂₄ nanospheres. Nanospheres with covalent linked interior pyridinium‐carboxyaldehydes are prepared. These groups form imines with amine derivatives. The describe strategy establishes guidelines for postmodification of supramolecular systems, thus providing new opportunities for unique confined spaces.

Introduction

Self‐assembled coordination cages (SCCs) have been established as multifunctional and versatile pillars of supramolecular chemistry. Many fascinating structures, such as helixes, sandwiches, and spheres have been reported with unique properties derived from the 3‐D space created within and on the outside of these assemblies.^{[ 1} , ² , ³ , ⁴ , ⁵ , ⁶ , ^{7 ]} As such, during the last decades SCCs have provided plenty of fascinating examples of defined spaces which are able to tune the properties of encapsulated molecules. Examples include: modulation of catalytic activity,^{[ 8} , ⁹ , ^{10 ]} changes of optical,^{[ 11} , ¹² , ^{13 ]} electronic properties,^{[ 14} , ¹⁵ , ¹⁶ , ^{17 ]} creation of unique phases,^{[ 18} , ^{19 ]} and host–guest chemistry.^{[ 20} , ²¹ , ^{22 ]} Classically, covalent chemistry is used during the preparation of the building blocks, which are required to construct an architecture by noncovalent (reversible) coordination to a transition metal. The resulting functionalities of the formed structure are then identical to the functionalities which were installed on the applied building block prior to the self‐assembly. As certain functional groups are not tolerated during the assembly process, post‐assembly modification (PAM) strategies have been developed to allow placement of functionalities that are incompatible with the formation of coordination‐based assemblies.^{[ 23} , ^{24 ]}

The different accessible PAM techniques used for SCCs remind one of classical biorthogonal chemistry, such as protein, DNA, and peptide modifications.^{[ 25} , ²⁶ , ^{27 ]} Both modulation of SCCs and biological systems have limited compatibility windows which decide the size of applicable toolkit. As such, orthogonality with functional groups present in the parent structure and high selectivity of the intended modification has to be provided. As most SCCs are constructed from dynamic coordination bonds and are thus held together by multiple weak and dynamic interactions (in contrast to covalent bonds), they often lack stability when exposed to high temperature, strong nucleophiles, acids, base, reducing/oxidation agents, or ligands.^{[ 28} , ²⁹ , ³⁰ , ^{31 ]} As most SCCs cannot be purified by common organic techniques, high selectivity and yield of the applied PAM is required. These challenges related to PAM of SCCs were addressed by the implementation of “click” chemistry.^{[ 32} , ^{33 ]} Similar to many bio‐orthogonal approaches, high yielding and selective reactions have been employed, including copper catalyzed cycloaddition, ruthenium catalyzed metathesis, and Diels–Alder reactions. Whereas many different PAM have been reported on exterior of SCCs^{[ 34} , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ^{39 ]} or on ligand functionalization^{[ 23} , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ^{47 ]} (Figure 1),^{[ 23} , ^{24 ]} limited examples of interior modifications are reported as typically limited space is available.^{[ 48} , ⁴⁹ , ^{50 ]} As a result, the reported interior PAM methodologies are limited to only small sized functional groups. We were wondering if it would be possible to postmodify the interior the much larger M₁₂L₂₄ nanospheres in a quantitative manner, thereby expanding the potential application area of such systems. Here we demonstrate an unprecedented straightforward quantitative PAM approach of M₁₂L₂₄ nanospheres by using highly electrophilic alkyl N‐pyridinium carboxy aldehyde (ANCA) inside such a system. The M₁₂ L^α ₂₄ nanospheres undergo selective and quantitative post modification with aniline and derivatives thereof by forming the corresponding imine. Using PAM amino acids, chromophores and several other functional groups can be installed in the well‐defined interior of M₁₂L₂₄ assemblies by using modified anilines, leading to new functionalities of the M₁₂ L^α ₂₄ nanospheres, such as binding of aromatic guests and catalysis. This work presents the first general quantitative PAM approach of the interior space of large nanospheres, expanding the potential application area of such systems.

Figure 1.

Selected examples presenting two previously applied postassembly modification strategies. Simplified representation of the herein reported confined space functionalized by PAM. Functional groups required for PAM (red) and the post modified products (blue) depicted as nanospheres.

Results and Discussion

PAM requires the presence of a reactive functionality at the interior of the self‐assembled structure and as such it needs to be installed on the building block that is used for the self‐assembly process. The reactive functionality can be converted to the functional group by postmodification. We chose the versatile M₁₂L₂₄ nanospheres for our investigations as these have been used in different areas and these have sufficient space to accommodate multiple functional groups. These types of nanospheres are formed by square planar complexation of ditopic pyridine building blocks with palladium or platinum. The building blocks were shown to be useful for different interior (endo) and exterior (exo) functionalization. As the nanosphere offers enough space for endo placement of many functional groups, it represents a good candidate for PAM of the inner part of the nanosphere. For our investigations, we focused on imine bond formation, as this type of reaction is often high yielding and does not require any catalyst or reagent, thus is anticipated to be compatible with M₁₂L₂₄ nanospheres. Along these lines we designed a building block that consists of two functionalities (Figure 2a). A nanosphere forming unit with two pyridine donors aligned in a 120° arrangement and an ANCA group connected by an alkane linker on the endo site. The ANCA group was chosen as it represents a highly electrophilic aldehyde which undergoes rapid imine formation with a variety of amines at room temperature. The building blocks were prepared in a simple one step S_N2 reaction using precursors reported in literature (see Scheme S1).

Figure 2.

Preparation of [Pd₁₂ L^α3 ₂₄]⁴⁸⁺ and [Pt₁₂ L^α4 ₂₄]⁴⁸⁺nanospheres. a) Reaction conditions for formation of the nanocages. Molecular structure of the displayed cage was minimized at the PM3 level. Carbon displayed in beige, nitrogen in blue, palladium as yellow spheres, interior ANCA group as blue spheres. b) DOSY NMR of the [Pd₁₂ L^a3 ₂₄] nanosphere. c) ¹H‐NMR spectra of [Pd₁₂ L^α3 ₂₄] (and [Pt₁₂ L^α4 ₂₄]) nanosphere and the corresponding building block. d) ESI‐MS spectra of [Pd₁₂ L^α3 ₂₄]⁴⁸⁺.

With the building blocks in hand, we studied its applicability in nanosphere formation and suitability for postmodification of M₁₂L₂₄ nanospheres. Classically, palladium is used for self‐assembly with some recent reports of platinum‐based M₁₂L₂₄ nanospheres.^{[ 29} , ^{51 ]} The platinum‐nanospheres are more robust/inert, which is useful for some applications, but are slightly more difficult to prepare. Because different applications require different metal nodes of M₁₂L₂₄ nanospheres, we studied our methodology using both, palladium and platinum. The Pd₁₂ L^α3 ₂₄ nanosphere is prepared according to standard literature procedures. L^α3 (1 equiv) and [Pd(MeCN)₄](BF₄)₂ (0.55 equiv) are heated at 35 °C in acetonitrile‐d₃ for 1 day (Figure 2a). A sharp ¹H‐NMR spectrum indicates the formation of a highly symmetrical structure (Figure 2c). Downfield shift of the pyridine protons is in line with coordination to palladium (signal a and b; Figure 2c). DOSY NMR shows the appearance of a slow diffusing species (log D = −9.47 m²·s⁻¹), with a calculated hydrodynamic radius of 2.0 nm in line with the formation Pd₁₂ L^α3 ₂₄ assembly (Figure 2b). Mass analysis confirms selective formation of the desired assembly by showing signals corresponding to different charged states of the assembly with the general formula [Pd₁₂ L^α3 ₂₄(BF₄)_x]^(48‐x)+ with matching isotope pattern for 6 ≤ x ≤ 15 (Figure 2d). The nanosphere is obtained in high selectivity and is suitable as such for further investigations.

Platinum‐based nanospheres or generally nanospheres with inert metal–ligand nodes require typically harsher conditions for their preparation. Therefore, finding a functional group which can be selectively addressed for PAM and can withstand nanosphere preparation with kinetically inert metal–ligand bonds is challenging. To evaluate whether ANCA is also suitable for robust nanospheres, we attempted platinum‐based nanosphere formation in line with previously reported protocols for which a charged functional group at the interior is important to obtain pure nanosphere.^{[ 29} , ^{51 ]} First, we attempted formation of Pt₁₂ L^α3 ₂₄ using the same ligands as described for palladium. The nanosphere Pt₁₂ L^α3 ₂₄ was synthesized by mixing 0.6 equiv [Pt(MeCN)₄](BF₄)₂, 1 equiv L^α3 and 7 mol% of TBACl (as catalyst) in acetonitrile‐d₃ (Scheme S4). The solution is heated for 3 days at 150 °C before being analyzed by different techniques. ¹H‐NMR displays a shift of the pyridine protons in analogy to the palladium‐based assembly (Figure S14). However, the ESI‐MS spectrum displays next to the different charged states of the desired Pt₁₂ L^α3 ₂₄ as the main species, also peaks associated to the kinetically trapped Pt₈ L^α3 ₁₆ and Pt₉ L^α3 ₁₈ species (Figure S15). Since L^α3 turned out to be not a suitable candidate for the selective formation of the desired Pt₁₂L₂₄ nanosphere, we decided to increase the length of the spacer between the sphere‐forming building block and the positively charged ANCA group to enhance the electrostatic repulsion of the ANCA group during sphere formation, a strategy which has previously been reported to allow for selective Pt₁₂L₂₄ formation.^{[ 51 ]} The platinum nanosphere Pt₁₂ L^α4 ₂₄ was synthesized by identical conditions as used for L^α3 (Figure 2a). ¹H‐NMR displays a sharp spectrum with a shift of the pyridine protons in analogy to the palladium‐based assembly (Figure 2c). DOSY NMR displays one slow diffusing species with a diffusion coefficient in line with the formation of Pt₁₂L₂₄ nanospheres (Figure S18). Finally, ESI‐MS confirms the selective formation of Pt₁₂ L^α4 ₂₄ by displaying multiple charged species which correspond to different charged states of the nanosphere (Figure S19), with no sign of the kinetically trapped smaller intermediates in the spectra.

With the two nanospheres (Pd₁₂ L^α3 ₂₄ and Pt₁₂ L^α4 ₂₄) in hand, post assembly covalent modification was explored. To a solution of Pd₁₂ L^α3 ₂₄ in acetonitrile‐d₃, 28 equiv aniline was added, and the solution was stirred overnight and subsequently analyzed via different techniques. ¹H‐NMR displays a sharp set of signals, supporting a symmetrical structure being present in solution. The position of all signals corresponding to the nanosphere forming backbone remain at the same location as before post‐modification (signal a, b, h, and i; Figure 3b). Furthermore, DOSY NMR shows the presence of a single species in solution with the same diffusion coefficient as the parent nanosphere (Figure S23). Both, the ¹H‐NMR and DOSY support the presence of a Pd₁₂L₂₄ nanosphere after PAM is applied. In contrast to the characteristic spectroscopic nanosphere signals, signals of the ANCA groups are changed. The aldehyde CH signal completely disappears (signal c; Figure 3b). All other ANCA signals shift in comparison to the parent nanosphere (signals d, e, f, and g; Figure 3b). An additional signal appears at 8.6 ppm, in line with the anticipated imine bond that is formed. The characteristic disappearance of the carbonyl signal together with a shift of all ANCA signals suggest successful formation of the desired aniline imine (Figure 3; right). Characterization of the formed structure using ESI‐MS furthermore supports the selective modification. Importantly, ESI‐MS displays only signals which can be attributed to the adduct of aniline with the parent nanosphere (Pd₁₂ L^α3βaniline ₂₄). Mass analysis shows signals corresponding to different charged states of the post‐modified nanosphere with the general formula [Pd₁₂ L^α3βaniline ₂₄(BF₄)_x]^(48‐x)+ with matching isotope pattern for 6 ≤ x ≤ 14 (Figure 3c). No signals corresponding to another stoichiometry and no signals corresponding to the parent nanosphere are detected. The combination of ¹H‐NMR, DOSY NMR and ESI‐MS analysis support the selective quantitative PAM of Pd₁₂ L^α3 ₂₄ using aniline. Therefore, he nanosphere remains fully intact using the ANCA‐aniline combination (as supported by ¹H‐NMR over time, Figure S21), and the protocol leads to the selective and quantitative modification of all 24 linkers in a single step.

Figure 3.

Post assembly covalent modification of [Pd₁₂ L^α3 ₂₄]⁴⁸⁺ using aniline. a) Reaction conditions for PCM of the nanocage. Molecular structure of the nanocage: carbon displayed in beige, nitrogen in blue, palladium as yellow nanocages, interior ANCA group as blue nanocages; post modified interior group as red nanosphere. b) ¹H‐NMR spectra of [Pd₁₂ L^α3 ₂₄] nanocage and the postmodifies [Pd₁₂ L^α3βaniline ₂₄]. c) ESI‐MS spectrum of [Pd₁₂ L^α3βaniline ₂₄] (top) and [Pd₁₂ L^α3 ₂₄] (bottom).

With an efficient ANCA‐aniline post assembly modification tool in hand, the scope of the approach was studied to support the generality of our concept of interior PAM using ANCA. For our PAM, we were inspired by interiorly functionalized nanospheres developed by the groups of Fujita and Stang. Specifically, we performed PAM of our Pd₁₂ L^α3 ₂₄ nanosphere using a) an amino pyrene for the creation of an hydrophobic, π rich environment for guest encapsulation reminiscent of the coronene functionalized nanosphere reported in literature,^{[ 52 ]} b) tetraphenyl ethylene amine as a photoprobe reminiscent of the nanosphere reported in literature, ^{[ 11 ]} c) a azostilbene as molecular switch (see analogous nanosphere, ^{[ 53 ]}) d) a hydrazine derivative, and e) an amino acid to create an chiral environment at the inside of the nanosphere (see analogous nanosphere ^{[ 54 ]}) (Figure 4a–e).

Figure 4.

Small literature inspired scope of PAM of Pd₁₂ L^a3 ₂₄ yielding nanospheres with internal functionalization containing a) hydrophobic π rich environment for guest binding, b) a dye to study confinement effects on photophysical properties, c) a azastilbene as a switch to generate a stimuli‐responsive interior, d) a hydrazone, and e) a chiral amine. Full details can be found in the Supporting Information.

First, we briefly explored the application of the Pd₁₂ L^α3 ₂₄ in host–guest chemistry using amino pyrene PAM (Figure 5) as π rich and hydrophobic interior functionalization of nanospheres has been shown previously useful for binding of aromatic guests.^{[ 52 ] 1}H NMR shows a decrease of the aldehyde signal accompanied by broadening of all other signals after addition of 0.5 equiv amino pyrene to Pd₁₂ L^α3 ₂₄ (Figure 5b). After addition of extra 0.6 equiv amino pyrene (total 1.1 equiv amino pyrene per building block of the nanosphere), the aldehyde signal completely disappears and the nanosphere signals sharpen up slightly (Figure 5b). Overall, disappearance of the aldehyde signals of the ANCA group indicates successful PAM of Pd₁₂ L^α3 ₂₄ using amino pyrene. Although, the resulting ¹H NMR spectrum of the nanosphere is somewhat broader than typically expected for a highly symmetrical assembly, it is in very good agreement with the literature reported broad ¹H NMR spectra of interiorly coronene functionalized Pd₁₂L₂₄ nanosphere.^{[ 52 ]} We hypothesize that the crowded environment inside the nanosphere (see Figure 5a) together with the strong hydrophobic character of the aromatics inside the nanosphere and the two possible isomers of the formed imine (Figure 5b) cause a broadening of the ¹H NMR signals similar to literature reports.^{[ 52 ]} The successful PAM using amino pyrene is further supported by DOSY NMR (Figure S36) and HR‐ESI MS (taken after the addition of 1.3 equiv amino pyrene; Figure 5c). Interestingly, also signals were identified in ESI‐MS that indicate the noncovalent binding of amino pyrene as guest in the in situ generated Pd₁₂ L^α3βpyrene ₂₄ leading to n aminopyrene⊃Pd₁₂ L^α3βpyrene ₂₄ (for 0 ≤ n ≤ 4; Figure 5c). Despite a 1.3‐fold excess of amino pyrene is used for the postmodification no free amino pyrene is present, as DOSY NMR shows all peaks at the same diffusion coefficient (Figure S36), implying host–guest complexation as further supported by HR‐ESI MS. In contrast, when using a nanosphere which cannot be modified interiorly by amino pyrene (Pd₁₂ L^α0 ₂₄ having an acetal instead of the the ANCA aldehyde group) but bears similar electronic and steric properties to Pd₁₂ L^α3 ₂₄, no binding of pyrene is observed as supported by ¹H NMR, DOSY, and HR‐ESI MS (Figures S40–S42). This example demonstrates how changing the interior of the nanosphere by PAM affects the properties of a nanosphere. It shows a practical application of PAM to create host systems for guests (in this case aromatic π rich systems) after the nanosphere is assembled.

Figure 5.

Post assembly covalent modification of [Pd₁₂ L^α3 ₂₄]⁴⁸⁺ using pyrene amine to create a unique confined space. a) Molecular structure of the cage: carbon displayed in white, nitrogen in blue, oxygen in red, and nanocage framework in pink. Molecular structure of aniline derivatives. b) ¹H‐NMR spectra of [Pd₁₂ L^α3 ₂₄] nanosphere and the postmodified [Pd₁₂ L^α3βpyrene ₂₄]. c) ESI‐MS spectrum of [Pd₁₂ L^α3βpyrene ₂₄].

PAM of Pd₁₂ L^α3 ₂₄ using other relatively larger aromatic aniline derivatives, such as tetraphenyl ethylene amine, and amino azobenzene (Figure 4b,c) proceeds like the detailed described amino pyrene. The addition of either tetraphenyl ethylene amine or amino azobenzene to Pd₁₂ L^α3 ₂₄ leads to relatively broad ¹H NMR spectra after PAM (Figures S43 and S49) in line with broad ¹H NMR spectra reported for pre‐functionalized derivatives of these nanospheres.^{[ 11 ]} The successful PAM yielding Pd₁₂ L^α3βTPE ₂₄ and Pd₁₂ L^α3βAza ₂₄ was furthermore supported by DOSY NMR and HR‐ESI MS (Figures S46 and S51). However, due to the difference in electronics at the aniline position of tetraphenyl ethylene amine and amino azobenzene compared to aniline and amino pyrene, we observe more fragmentation in ESI‐MS for tetraphenyl ethylene amine and amino azobenzene (Figures S46 and S51). As such, we observe signals corresponding to the post assembly modified nanospheres with 21–24 TPE and 17–24 azobenzene amines (Figures S46 and S51).

PAM of Pd₁₂ L^α3 ₂₄ using a chiral aniline functionalized alanine amino acid shows characteristic disappearance of the ANCA aldehyde protons indicative of successful PAM of all 24 interior ANCA groups. In contrast to the large multi aromatic systems, PAM using aniline alanine yields a sharp ¹H NMR spectra (Figure 6) in line with previously reported nanospheres containing small amino acids.^{[ 54 ]} The hydrogen bonding character of the amino acid together with the positive charge of the ANCA group makes it not possible to characterize the alanine PAM nanosphere by HR‐ESI MS (only noise visible, Figure S33) likely due to clustering of the nanosphere with solvent and water. Next to the successful PAM using different interesting functional groups from well documented nanospheres with specific properties, we finally, also attempted one PAM of Pd₁₂ L^α3 ₂₄ using a hydrazine derivative (Figure 4d). All 24 aldehyde functional groups of the nanosphere were successfully modified by p‐toluenesulfonyl hydrazine as evidenced by ¹H NMR, DOSY, and HR‐ESI MS (although signal intensity was relatively weak due to cluster formation with solvent and water, Figures S25–S28). Noteworthy, a small portion of the Pd₁₂ L^α3 ₂₄ nanosphere decomposed to the corresponding free building block L^α3 as evidenced by ¹H NMR and DOSY (roughly 15% decomposition, Figures S26, S27), indicating limited stability of the palladium nanosphere against hydrazine derivatives.

Figure 6.

Post assembly covalent modification of [Pd₁₂ L^α3 ₂₄]⁴⁸⁺ using aniline alanine. ¹H‐NMR spectra of [Pd₁₂ L^α3 ₂₄] nanocage and the postmodified [Pd₁₂ L^α4βala ₂₄].

After our proof of concept of efficient PAM using aniline derivatives for interior functionalization of Pd₁₂L₂₄ nanospheres decorated with ANCA, we set out to investigate the analytically more challenging Pt analogous nanospheres. Although, Pt‐nanospheres are much more robust compared to the palladium analogue making them very attractive for catalytic and biological applications,^{[ 55} , ^{56 ]} their analysis via ESI‐MS is challenging due to low intensity signals and the tendency to cluster with solvent.^{[ 11} , ^{57 ]} Therefore, a PAM approach which can be validated using ¹H NMR and DOSY together with other characteristics of interest, would yield an ideal tool for further studies into these robust nanospheres. Aware of the potential mass analytical difficulties observed for platinum‐based nanospheres, we first set out postmodification of the Pt₁₂ L^α4 ₂₄ nanospheres using aniline under similar conditions as for Pd₁₂ L^α3 ₂₄. Upon addition of aniline to Pt₁₂ L^{α 4} ₂₄, ¹H NMR shows disappearance of the ANCA aldehyde signal (Figure 7b) accompanied by a shift of the rest of the ANCA protons. The characteristic signals of the nanosphere (such as the pyridine signals, a and b, Figure 7b) remain at the same position, indicating that the nanosphere remains intact. Although, all signals are broader for the platinum nanosphere, the PAM forming Pt₁₂ L^α4βaniline ₂₄ shows signals identical in position to the palladium analogue nanosphere Pd₁₂ L^α3βaniline ₂₄ (with broadening generally attributed to platinum nanospheres and the extended linker provides a more crowded environment which may cause broadening of signals, see Figure 5). The successful PAM of Pt₁₂ L^{α 4} ₂₄ is furthermore supported by DOSY NMR (Figure S54). HR ESI analysis of Pt₁₂ L^α4βaniline ₂₄ displays broad signals centered around different charged species. The broadening of the signals is caused by association of different solvent molecules (such as MeCN and water, see Figures S55–S64) and by loss of aniline due to harsher measuring conditions required to detect platinum nanospheres. The signals centers of Pt₁₂ L^α4βaniline ₂₄ are displayed in Figure 7c with their corresponding charges (Figures S55–S57). In comparison to the starting material, we observe a shift of all mass signals by 1800 Da. This shift agrees with a condensation of 24 anilines and loss of 24 water molecules from the parent nanosphere. Altogether, all analytical techniques support our hypothesis that also platinum‐based nanospheres can be successfully modified postassembly using interior ANCA functionalized parent nanospheres.

Figure 7.

Post assembly covalent modification of [Pt₁₂ L^α4 ₂₄]⁴⁸⁺ using aniline. a) Reaction conditions for PCM of the nanocage. b) ¹H‐NMR spectra of [Pt₁₂ L^α4 ₂₄] nanocage and the postmodified [Pt₁₂ L^α4βaniline ₂₄] compared to the the postmodified [Pd₁₂ L^α3βaniline ₂₄]. c) ESI‐MS spectrum of [Pt₁₂ L^α4βaniline ₂₄] (top) and [Pt₁₂ L^α4 ₂₄] (bottom) with the observed and expected molecular weight difference between the starting material and the PAM nanosphere.

Finally, we explored our PAM methodology for catalytic applications. To do so, we set out to use the platinum‐based nanospheres as they show greater stability^{[ 29} , ⁵⁵ , ^{56 ]} and hydrazine modified catalyst for PAM also due to better stability of the resulting hydrazide in comparison to imines (see stability studies Supporting Information section SI4, Figures S67–S69). We postmodified the platinum nanosphere with a transition‐metal complex (βAu). This gold complex is closely related to reported analogues and used to probe the effect of high local concentrations; it was established previously that monomeric gold complexes of this type are inactive in this transformation, while at high local concentration when placed inside a nanospheres they become active.^{[ 58} , ^{59 ]} As gold complexes typically do not withstand Pt nanosphere formation protocols (150 °C), PAM of a platinum nanosphere offers an elegant opportunity for incorporation of these type of transition metal‐complexes. The parent nanosphere Pt₁₂ L^α4 ₂₄, the parent complex βAu, and the postmodified nanosphere Pt₁₂ L^α4βAu ₂₄ were briefly studied in gold catalyzed cyclization of 1 (Table 1). As expected from previous results, both parent systems display no activity. For the gold complex, inactivity stems from the coordinated halide which typically results in inactive complexes. In contrast to that, the post modified Pt₁₂ L^α4βAu ₂₄ (with high local gold concentration within the nanosphere and aurophilic interactions), catalyzes the cyclisation with 65% conversion after 43 h in line with previous results.^{[ 58} , ^{59 ]} As such, we conclude that our methodology is applicable to both organic molecules and metal complexes.

Table 1.

Gold catalyzed cyclization of 1, using the nanosphere Pt₁₂ L^α4 ₂₄ that was postmodified with the gold chloride complex βAu.

Entry	Catalyst	Conversion b)
1	Pt12 Lα4 24	0%
2	βAu	0%
3	Pt12 Lα4βAu 24	65% (100%) c)

^a) Reactions performed at 25 mM substrate with 0.5 mol% nanosphere (13 mol% gold) in MeNO₂ at r.t. for 48 h.

^b) Conversion was determined with ¹HNMR using mesitylene as internal standard, averaged over three runs.

^c) after 80 h.

In conclusion, we wondered if the interior of M₁₂L₂₄ nanospheres can be functionalized by postmodification to allow the generation of functionalized nanosphere systems that have functional groups that are not compatible with the self‐assembly of such nanospheres. For this purpose, we have introduced the alkyl ANCA group at the inside of M₁₂L₂₄ nanospheres as a functional group for post assembly modification by imine formation. Covalently linked ANCA has been installed at the endo site of building blocks, which together with palladium or platinum yield selectively the desired M₁₂L₂₄ nanospheres. The highly electrophilic nature of ANCA allows quantitative imine formation of all 24 endohedrally placed functionalities with aniline, hydrazine, and amine derivatives. This provides a unique tool for confined space modification. The interior of a nanosphere can be decorated with hydrophilic, hydrophobic, or chiral groups. We demonstrate the application of a Pd‐nanosphere in host guest chemistry, after PAM, and a Pt‐nanosphere in catalysis. As the parent nanospheres/structures can be easily characterized before PAM, they offer an ideal platform for studies into confined space effects on molecular probes of desire and allow studies into combinatorial and more complex function rich constructs, which were before not possible before due to analysis or synthesis restrictions. The high yielding ANCA derivatization provides a platform for application in other supramolecular structures, as the PAM does not require any catalyst, proceeds at room temperature and does not require any purification.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting Information

Bobylev E. O., de Bruin B., Reek J. N. H., Angew. Chem. Int. Ed. 2025, 64, e202512402. 10.1002/anie.202512402

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.