Synthesis and Intramolecular Energy- and Electron-Transfer of 3D-Conformeric Tris(fluorenyl-[60]fullerenylfluorene) Derivatives

New 3D conformers were synthesized to show a nanomolecular configuration with geometrically branched 2-diphenylaminofluorene (DPAF-C2M) chromophores using a symmetrical 1,3,5-triaminobenzene ring as the center core for the connection of three fused DPAF-C2M moieties. The design led to a class of cis-cup-tris[(DPAF-C2M)-C60(>DPAF-C9)] 3D conformers with three bisadduct-analogous  cages per nanomolecule facing at the same side of the geometrical molecular cis-cup-shape structure. A sequential synthetic route was described to afford this 3D configurated conformer in a high yield with various spectroscopic characterizations. In principle, a nanostructure with a non-coplanar 3D configuration in design should minimize the direct contact or π-stacking of fluorene rings with each other during molecular packing to the formation of fullerosome array. It may also prevent the self-quenching effect of its photoexcited states in solids. Photophysical properties of this cis-cup-conformer were also investigated.


Introduction
Photoinduced intramolecular energy and electron transfer phenomena in organo [60]fullerene derivatives having a covalent molecular composition of both an electron donor and a [60]fullerenyl or nanocarbon acceptor components were demonstrated over a number of years [1][2][3]. The energy process may involve the facile triplet state of fullerene and other chromophores [4][5][6][7][8]. This type of nanomolecular system was used in many technological applications [9,10], including photovoltaic devices [11,12], sensors and switches [13], and photodynamic therapy [14,15]. Fullerene-based nanostructures with multiple C 60 cages [16] in the structure were found to be suitable for the applications of nanocars [17][18][19], photoswitches [20], molecular heterojunctions [21], and catalysts [22]. Unusual molecular properties of multi-cage fullerene objects were theoretically predicted [23][24][25][26]. Recently, similar photophysical chemistry was also simulated in the modulation of photoswitchable dielectric properties to observe a large amplification of dielectric constants in a material combination form of multi-layered core-shell nanoparticles (NPs) [27][28][29]. The latter system was based on photoinduced intramolecular electronic charge-polarization of light-harvesting chromophoric nano[60]fullerenyl conjugates, such as 9,9-di(3,5,5-trimethylhexyl)-2-diphenylaminofluorenyl-methano[60]fullerene C 60 (>DPAF-C 9 ) (1-C 9 , Figure 1). The polarization provided detectable dielectric property enhancement in a layered [60]fullerosome membrane structure on gold-shelled nanoparticles. The phenomena were based on the high electronegativity of C 60 > cage making it possible to rapidly shift an electron from light-harvesting DPAF (diphenylaminofluorene) donor moiety within the molecular structure to the C 60 > moiety. This resulted in the formation of the corresponding charge-separated (CS) state C 60 − ·(>DPAF-C 9 ) + as the source of polarized charges. In fact, ultrafast concurrent intramolecular energy and electron transfer kinetics within 1-C 9 were substantiated previously by femtosecond transient absorption measurements (pump-probe) [30][31][32]. When these negative charges were distributed, delocalized, and stabilized in the fullerosome membrane array at the shell layer on core-shell NPs, it resulted in a CS state with a lifetime prolonged enough for the detection of dielectric characteristics. The process involved interlayer photoinduced plasmonic energy transfer from the Au shell layer to the outer shell layer of C 60 (>DPAF-C 9 ) in addition to the fact that 1-C 9 itself is also photoresponsive and excitable under light irradiation.
One crucial parameter to consider is the method of molecular packing within the fullerosome shell layer. In this regard, strong tendency of light-harvesting chromophores to aggregate among π-conjugated planar aromatic moieties can cause either concentration-dependent self-quenching effects of excited states or luminescence in the solid phase, including fullerosome. The π-π stacking may result in significant reduction of many photophysical properties. This type of packing aggregation can be partially minimized by the use of highly bulky and geometrically hindered π-conjugated chromophores in a structural design to restrict or distort intramolecular rotation bonding units with steric hindrance [33]. In the case of DPAF-C 9 , we recently developed and synthesized highly restricted 3D conformers based on inter-connected three DPAF-C 9 chromophore units giving a structure of tris(DPAF-C 9 ) (2-C 9 ) [34] to prevent and minimize the tendency of planar DPAF units to undergo aromatic-aromatic stacking, overlapping, and aggregation via intermolecular hydrophobic-hydrophobic interactions in solid thin-films. This structural modification led the enhancement of photophysical properties, including the intensity of photoluminescence (PL) and electroluminescence (EL) emissions [35].
Molecules 2019, 24, x FOR PEER REVIEW 2 of 16 structure on gold-shelled nanoparticles. The phenomena were based on the high electronegativity of C60> cage making it possible to rapidly shift an electron from light-harvesting DPAF (diphenylaminofluorene) donor moiety within the molecular structure to the C60> moiety. This resulted in the formation of the corresponding charge-separated (CS) state C60 -·(>DPAF-C9) + as the source of polarized charges. In fact, ultrafast concurrent intramolecular energy and electron transfer kinetics within 1-C9 were substantiated previously by femtosecond transient absorption measurements (pump-probe) [30][31][32]. When these negative charges were distributed, delocalized, and stabilized in the fullerosome membrane array at the shell layer on core-shell NPs, it resulted in a CS state with a lifetime prolonged enough for the detection of dielectric characteristics. The process involved interlayer photoinduced plasmonic energy transfer from the Au shell layer to the outer shell layer of C60(>DPAF-C9) in addition to the fact that 1-C9 itself is also photoresponsive and excitable under light irradiation.
One crucial parameter to consider is the method of molecular packing within the fullerosome shell layer. In this regard, strong tendency of light-harvesting chromophores to aggregate among πconjugated planar aromatic moieties can cause either concentration-dependent self-quenching effects of excited states or luminescence in the solid phase, including fullerosome. The π-π stacking may result in significant reduction of many photophysical properties. This type of packing aggregation can be partially minimized by the use of highly bulky and geometrically hindered π-conjugated chromophores in a structural design to restrict or distort intramolecular rotation bonding units with steric hindrance [33]. In the case of DPAF-C9, we recently developed and synthesized highly restricted 3D conformers based on inter-connected three DPAF-C9 chromophore units giving a structure of tris(DPAF-C9) (2-C9) [34] to prevent and minimize the tendency of planar DPAF units to undergo aromatic-aromatic stacking, overlapping, and aggregation via intermolecular hydrophobichydrophobic interactions in solid thin-films. This structural modification led the enhancement of photophysical properties, including the intensity of photoluminescence (PL) and electroluminescence (EL) emissions [35]. Accordingly, we extended the similar structural strategy to design new 3D conformers tris[(DPAF-C2M)-C60(>DPAF-C9)] (4-C2M-9), as shown in Figure 1, for the study. The stereochemical Accordingly, we extended the similar structural strategy to design new 3D conformers tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] (4-C 2M-9 ), as shown in Figure 1, for the study. The stereochemical modification was based on the construction of 3D geometrically branched tris[(DPAF-C 2M ) (2-C 2M ) chromophore having a shared central benzene unit among three 2-diphenylaminofluorenes. Highly steric hindrance at the corresponding 1,3,5-phenylfluorenylaminobenzene moiety forced three fluorene ring moieties to twist either upward or downward away from the central benzene plane with a large torsional angle on the nitrogen atom. This resulted in the formation of three to four possible stereoisomeric configurations, such as cis-cup-2-C 2M , trans-chair-2-C 2M , and cis/or trans-propeller-2-C 2M , as shown in Figure 1. All of these conformer forms were proposed to be capable of fully eliminating the tendency of 4-C 2M-9 in inducing π-π aromatic-aromatic type stacking packing and to allow all C 60 > cages to interact with each other via strong hydrophobic-hydrophobic interaction forces between (C 60 >)−(C 60 >) fullerenyl cages, forming the nano-layer array of fullerosome membrane.

Results and Discussion
Rapidly responsive nanophotonic physical properties of 3D conformers tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] (4-C 2M−9 ) are achievable by specifically associating a donor-acceptor type chemical structure to a photomechanism having the ability to create a largely enhanced intramolecular energy and electron transfer efficiency. This mechanism occurs between C 60 > acceptor and DPAF-C 2M /and DPAF-C 9 donor moieties bonded on fullerenes. In our functional group design of 4-C 2M−9 , a methanoketo bridging unit was used to trigger the keto-enol isomerization tautomerism that is capable of inducing π-periconjugation between C 60 > and DFAP to provide a partial conjugation pathway for enhancing the π-electron mobility around the conjugated system of molecular nanostructures [36,37]. In addition, the new molecular design of stereoisomeric tris(fluorenylphenylamino)benzene [tris(DPAF-C 9 )] analogous was proven to act as a fluorophore showing high intensity of photoluminescence and electroluminescence emission efficiency [34,35]. This revealed the successful utilization of stereochemistry to allow hindered and branched 3 ,5 ,5 -trimethylhexyl (C 9 ) arms to maintain the space-separation of three planar DPAF moieties intramolecularly within the nanostructure. It also behaved similarly in intermolecular packing that improved light-harvesting efficiency.
Based on the molecular formation energy based density functional theory (DFT) calculations of three plausible stereoisomers of tris(DPAF-C 9 ) (2-C 9 ) via B3LYP/6-31G* level of theory using SPARTAN08 [34,35], the results revealed high stability of the cis-cup-form with other forms in stability order of cup > chair > propeller, as shown in Figure 1. This agreed well for the alkyl n-C 6 , n-C 7 , and n-C 8 substituents owing to the influence by strong dispersion interactions within the alkyl chains. In the case of the methyl and the ethyl substituents, trans-chair-form may have been more stable than cis-cup-form. Accordingly, three C 4 -analogous substituents of tris(DPAF-C 2M ) (2-C 2M ) facing toward each other in the 3D molecular space above the central benzene ring should have brought in the minimum alkyl-alkyl interaction forces required to keep a slight favor of the cis-cup-form over either trans-chairor cis/trans-propeller form.
The compound 4-C 2M-9 exhibited good solubility in common organic solvents owing to its possession of three DPAF-C 9 (with a total of six branched C 9 -alkyl groups) and three DPAF-C 2M (with a total of six methoxyethyl groups) moieties (Figure 1), making three C 60 > cages become encapsulated in the center of the 3D molecular configuration. The 3D configuration of cis-cup form resulted in these alkyl groups being the main structural moieties interacting with the solvent. Accordingly, the compound had solubility (20 mg/mL in CHCl 3 or CH 2 Cl 2 ) over 10 times higher than that of C 60 itself in toluene (1.4 mg/mL).

Spectroscopic Characterization of Synthetic 3D Configurated Fullerenyl Nanomaterials
All chemical conversions of intermediate chemicals to the corresponding products at each step of the reactions were characterized by various spectroscopic techniques. The functional attachment of three α-bromoaceto groups (3.0 equiv.) to tris(DPAF-C 2M ) given the product of tris(BrDPAF-C 2M ) was verified by both infrared (FT-IR) and 1 H-NMR spectra. The former showed a new strong carbonyl (-C=O) stretching absorption band centered at 1673 cm −1 , indicating each of the three carbonyl groups being bonded on a phenyl moiety, such as that of 2-C 2M . This absorption wavelength was in clear contrast to the strong band absorption at 1725 cm −1 normally detectable for an alkyl carbonyl group. In the case of 1 H-NMR spectrum, tris(BrDPAF-C 2M ) displayed characteristic new peak signals of three methylene protons (H α ) next to the carbonyl group of the α-bromoaceto moiety at δ 4.49 ( Figure 2Ab) as compared with that of 2-C 2M (Figure 2Aa). Subsequent attachment of three C 60 (>DPAF-C 9 ) moieties to each of the three DPAF moieties of 3-C 2M with a cylopropylaceto bridging unit on each C 60 > cage of three 1-C 9 (applied as a reagent) showed evidence of changing solubility characteristics of the product 4-C 2M-9 matching with those of 1-C 9 . Its FTIR spectrum displayed a slight shift of cyclopropyl keto group absorption band to υ max 1679 cm −1 (Figure 3d), which was assigned to the carbonyl (C=O) stretching band. It was also accompanied by an olefinic (C=C) absorption band centered at υ max 1591 cm −1 . Both C=O and C=C bands were correlated to those of C 60 (>DPAF-C 9 ) ( Figure 3b) and tris[C 60 (>DPAF-C 9 )] (Figure 3c), showing a nearly identical absorption wavenumber. Most importantly, we were able to detect two typical fullerenyl cage bands at υ max 574 (w) and 524 (s) cm −1 (Figure 3d). These two bands were corresponding characteristic absorptions used to provide evidence of (C 60 >)-related monoadducts and bisadducts with absorption wavenumbers and relative intensity ratios differentiable from those of C 60 ( Figure 3a) and 1-C 9 substituents ( Figure 3b). Accordingly, we applied this IR technique for the product structure verification during the chemical conversion from 3-C 2M to 4-C 2M−9 . Upon conversion of C 60 to its monoadducts, such as those of Figure 3b,c, the remaining cage structure of C 60 > exhibited the same two bands with a reduced peak intensity for the 574 cm −1 band. The intensity of this band was further reduced in the structure of <C 60 >-like bisadduct, such as 4-C 2M−9 ( Figure 3d). Furthermore, the latter band at 524 cm −1 still remaining strong was indicative of successful attachment of C 60 (>DPAF-C 9 ) moieties on 3-C 2M due to the possibility of having the second malonate bridging unit being attached at or near the equator region of the C 60 > cage. This would have led to the retention of a C 60 half-cage that enabled absorption at 524 cm −1 .
Molecules 2019, 24, x FOR PEER REVIEW 5 of 16   reported chemical shift values of the α-proton (H α ') in C 60 (>DPAF-C 9 ) (Figure 2Ad) [20] and the related α-proton (H α ") in tris(DPAF-C 9 ) (2-C 9 , Figure 2Ae) [23] as the reference, we assigned these proton peaks to a combination of H α and H α '. A large downfield shift of the H α chemical shift from those of 3-C 2M at δ 4.49 to δ 5.25-5.78 for 4-C 2M-9 provided clear evidence of three C 60 (>DPAF-C 9 ) moieties being attached on the corresponding α-bromoaceto bridging units of 3-C 2M . The characteristics of the multipeaks for H α and H α ' revealed a less symmetrical environment among these six protons of 4-C 2M-9 as the geometric shape of the nanostructure extended to a 3D configuration. It is worthwhile to mention that a large down-fielded chemical shift value of either H α ' or H α " away from the normal value of δ 2.1-2.5 for an alkyl aceto-α-proton was caused by the influence of strong [60]fullerenyl ring current in close vicinity. In addition, the alkyl proton regions over δ 2.93 (methoxy proton, 18H) and δ 2.75-2.64 (ethylenoxy proton, 12H) of Figure 2Ac (marked by beige) were correlated well to those of 2-C 2M at δ 2.91 (18H) and δ 2.64 (12H) (Figure 2Aa), respectively, indicating good retention of central tris(DPAF-C 2M ) core region without any structural change of methoxy groups during the Friedel-Crafts acylation reaction. It also showed a new group of methyl proton peaks at δ 0.30-2.07 having an integration ratio value of 113.28, which represented 114 fluorenyl protons of C 9 -alkyl proton (19Hs for each of the two C 9 -alkyls of DPAF-C 9 ) and was indicative of six C 9 -alkyl groups in the structure, consistent with the product structure. Additional 1 H-NMR spectroscopic data analyses on proton integrations of all proton peaks to substantiate and count for the molar quantity ratio among fluorene, methoxyethyl, and C 9 alkyl moieties to prove the molecular formulation of 4-C 2M-9 are provided in supporting information.
Most importantly, characteristics of central benzene protons at the core region could be used for the analysis of the relatively geometric configuration of three fluorenyl rings with respect to each other. With a symmetrical structure of TPAB, three benzene protons should have displayed a singlet peak in its 1 H-NMR spectrum. Upon attachment of a bulky fluorenyl moiety at each diphenylamino group, it induced high torsional stress and steric hindrance at the nitrogen atom that forced each 9,9 -di(methoxyethyl)fluorene moiety to twist or rotate either upward or downward from the central benzene plane. The action resulted in two main 3D conformers: cis-cup-2-C 2M and trans-chair-2-C 2M . The former with three C 60 (>DPAF-C 9 ) moieties facing upward on the same side in the structure gave a singlet H a peak (Figure 1). The latter with one facing downward and two C 60 (>DPAF-C 9 ) moieties facing upward in the structure resulted in two proton peaks for trans-H a ' (1H) and trans-H b ' (2H) (Figure 2Ba). By analyzing Figure 2Aa of tris(DPAF-C 2M ), a sharp singlet proton peak at δ 6.53 was assigned to the chemical shift of central benzene proton cis-H a . This peak was compared with that of the H a proton peak of cis-cup-tris(DPAF-C 9 ) (cis-cup-2-C 9 , Figure 2Bb) showing even better resolution of the peak profile, indicating a high purity of one 3D conformer fraction in a cis-cup-2-C 2M form. Surprisingly, this conformer fraction was, in fact, the major product. Apparently, the hydrophobic-hydrophobic dispersion interaction forces derived from three methoxyethyl chains and heteroatoms were stronger than those among all C 4 -alkyl groups, which led to higher tendency in formation of the cis-cup-form. Accordingly, subsequent attachment of three C 60 (>DPAF-C 9 ) moieties on cis-cup-2-C 2M led to a similar formation of corresponding cis-cup-tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] (cis-cup-4-C 2M-9 ), all having 1-C 9 moieties facing upward from the central benzene core at the same side with respect to each other. Additional structural analyses and discussions are provided in supporting information.
In the case of the potential formation of regio-isomers of 4-C 2M−9 at the <C 60 > moiety, since the monoadduct structure of C 60 (>DPAF-C 9 ) was well-defined, with the assistance of X-ray single crystal structural analysis of C 60 (>DPAF-C 2 ) [31,36], its attachment on tris(BrDPAF-C 2M ) (3-C 2M ) was believed to be governed by the bulkiness and the steric hindrance of both relatively large entities to result in only a limited number of region isomers on the C 60 cage. This was proven by the 1 H-NMR spectrum of 4-C 2M−9 showing only several H α and H α ' proton peaks at roughly δ 5.2-5.8 (Figure 2Ac) instead of the broad band normally seen for the existence of a large number of region isomers. To our surprise, a peak at 524 cm −1 assigned the characteristic infrared absorption band of a half-C 60 cage (as stated above) showed close resemblance to those of the monoadduct C 60 (>DPAF-C 9 ) (Figure 3b) and tris[C 60 (>DPAF-C 9 )] (Figure 3c) at an identical wavenumber. This implied the structure of the major regio-isomeric products had both addend moieties located at the same half-sphere of a C 60 cage that left the other half-sphere of C 60 untouched.

Photophysical and Physical Properties of 3D Conformeric Fullerenyl Nanomaterials
Photophysical properties of the 3D conformer cis-cup-4-C 2M-9 were compared with those of precursor intermediates using the UV-vis spectroscopic technique. They were governed by two photoresponsive moieties, including three electron (e − )-accepting fullerene cages and six light-harvesting DPAF antenna units as electron (e − )-donors. The use of the latter was to enhance the optical absorption capability at longer visible wavelengths. The absorption wavelength could be varied and modulated by the appropriate chemical modification of functional substituents on fluorenyl moiety to affect electron-pushing (donating) and pulling (accepting toward the molecular edge of C 60 > cage moiety) mobility across the molecular π-conjugation system. As shown in Figure 4Ae of cis-cup-4-C 2M-9 , optical absorption of C 60 > cage moieties appeared mainly at the broad band centered at 296 nm (1.82 × 10 5 L mol −1 cm −1 ), whereas the band centered at 411 nm (1.11 × 10 5 L mol −1 cm −1 ) was attributed to the absorption of DPAF moieties. Characteristics of the latter band were compared with those of tris(DPAF-C 2M ) (Figure 4Aa Figure 4A) with a slightly higher extinction coefficient for the monoadduct 1-C 9 than the bisadduct cis-cup-4-C 2M-9 , which was also consistent with the photophysical property discussion above and provided further confirmation of a conjugated fullerenyl nanostructure.  Figure 4A) with a slightly higher extinction coefficient for the monoadduct 1-C9 than the bisadduct cis-cup-4-C2M-9, which was also consistent with the photophysical property discussion above and provided further confirmation of a conjugated fullerenyl nanostructure. In addition, a roughly 2.1-fold higher ε value (1.11 × 10 5 L mol −1 cm −1 ) of the 411 nm peak in Figure 4Ae compared to that of 404 nm band of cis-cup-tris[C60(>DPAF-C9)] was consistent with a double number of DPAF arms per molecule for the former. Furthermore, very efficient intramolecular energy transfer from the excited singlet state of DPAF-C9 antenna to C60> was detected, which nearly eliminated the fluorescence of C60(>DPAF-C9) (λex: 406 nm, Figure 4Bb). In high contrast, without any C60> cage in the structure, the compound of tris(DPAF-C2M) showed a strong intensity of fluorescence emission (λex: 352 nm, Figure 4Ba) that clearly indicated the loss of photoexcited energy being double number of DPAF arms per molecule for the former. Furthermore, very efficient intramolecular energy transfer from the excited singlet state of DPAF-C 9 antenna to C 60 > was detected, which nearly eliminated the fluorescence of C 60 (>DPAF-C 9 ) (λ ex : 406 nm, Figure 4Bb). In high contrast, without any C 60 > cage in the structure, the compound of tris(DPAF-C 2M ) showed a strong intensity of fluorescence emission (λ ex : 352 nm, Figure 4Ba) that clearly indicated the loss of photoexcited energy being associated with the influence of [60]fullerene. With an additional DPAF-C 9 antenna in the structure of 4-C 2M-9 , it began to experience a slightly excessive fluorescence emission (λ ex : 410 nm, Figure 4Bc) after the majority of photoexcited DPAF-C 2M energy underwent direct intramolecular energy transfer to the closely bonded [60]fullerene cage.
In investigating the plausibility of photoinduced intramolecular electron (e − )-transfer capability within the nanostructure of the 3D conformer cis-cup-tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] (cis-cup-4-C 2M-9 ), we first investigated the unit character of redox potentials among all structural components, including the bisadduct-based <C 60 > cage and the DPAF-C 9 moieties for comparison using the cyclic voltammetric (CV) technique. Several CV measurements were performed on the sample of cis-cup-4-C 2M-9 in a solution of CH 2 Cl 2 containing (n-butyl) 4 N + -PF 6 − as the electrolyte and Pt as both the working and the counter electrodes and with Ag/AgCl as the reference electrode.
To deliver appropriate redox potential analyses and data interpretation, related CV characteristics of a C 60 -bisadduct of C 60 (>t-Bu-malonate) 2 with a <C 60 > cage attached by two t-butylmalonate groups and the precursor compound C 60 (>DPAF-C 9 ) were collected. They were performed under the same CV condition over cyclic oxidation and reduction voltages versus Ag/Ag + from −2.0 to 2.0 V as those for cis-cup-4-C 2M-9 , as shown in Figure 5. As a result, it displayed one reversible oxidation ( 1 E ox of 1.51 V) reduction ( 1 E red of 1.32 V) cyclic wave with the first half wave oxidation potential ( 1 E 1/2,ox ) of 1.42 V for the DPAF moieties of cis-cup-4-C 2M-9 at positive voltages (Figure 5Ab,Bb). In the negative voltage region, its CV diagram displayed three reversible reductions at −0.34 ( 1 E red ), −0.82 ( 2 E red ), and −1.29 V ( 3 E red ) with the corresponding cyclic oxidation waves at −0.15 ( 1 E ox ), −0.44 ( 2 E ox ), and −0.96 ( 3 E ox ), respectively. These data corresponded to the first to the third half wave reduction potentials of −0.25 ( 1 E 1/2 , red ), −0.63 ( 2 E 1/2 , red ), and −1.12 V ( 3 E 1/2 , red ), respectively. By comparison of these values to those of C 60 (>t-Bu-malonate) 2 (Figure 5Aa,Ba) for the fullerene cage moiety and those of C 60 (>DPAF-C 9 ) (Figure 5Ac,Bc) for both DPAF and fullerene cage moieties, highly consistent and reproducible redox potential characteristics among structural components were found that also substantiated the structural derivatization of tris[(BrDPAF-C 2M ) (3-C 2M ) with triple C 60 (>DPAF-C 9 ) to form cis-cup-4-C 2M-9 . Accordingly, the latter exhibited combined CV characteristics of <C 60 > and DPAF-C 9 . These CV characteristics were reproducible for four repeated redox cycles with the reductive C 60 > and the oxidative DPAF potential profiles showing only slight changes at the potential range of −2.0 to 2.0 V. This implied good stability of the material under CV conditions that led to possible reuse of 4-C 2M-9 . structural components were found that also substantiated the structural derivatization of tris[(BrDPAF-C2M) (3-C2M) with triple C60(>DPAF-C9) to form cis-cup-4-C2M-9. Accordingly, the latter exhibited combined CV characteristics of <C60> and DPAF-C9. These CV characteristics were reproducible for four repeated redox cycles with the reductive C60> and the oxidative DPAF potential profiles showing only slight changes at the potential range of −2.0 to 2.0 V. This implied good stability of the material under CV conditions that led to possible reuse of 4-C2M-9.

Evidence of Intramolecular Energy-and Electron-Transfer Events within cis-cup-4-C2M-9 by Detection of Corresponding Reactive Oxygen Species (ROS).
There is an appropriate approach to substantiate intramolecular energy and electron transfer events within the nanomolecular structure of cis-cup-4-C2M-9 by directly detecting the photoinduced production of reactive oxygen species (ROS). In general, the most common ROS includes singlet --
Accordingly, by the direct detection of ROS on either 1 O 2 and/or O 2 − · upon irradiation on cis-cup-4-C 2M-9 at either <C 60 > or DPAF-C 2M /DPAF-C 9 moiety, we were able to provide the evidence of intramolecular energy and electron transfer processes happening within this 3D-conformer. We selected two reliable fluorescent (FL) probes for the detection of either 1 O 2 or O 2 − ·separately in the solution of cis-cup-4-C 2M-9 with high selectivity and specificity as a crucial measure. To detect the former ROS 1 O 2 , a synthetic highly fluorescent compound α,α'-(anthracene-9,10-diyl)-bis(methylmalonic acid) (ABMA) was used as the probe in the experiment. Its UV-vis absorption and fluorescence emission spectra are given in Figure 6a,b, respectively. In the probe reaction, chemical trapping of 1 O 2 by highly fluorescent ABMA resulted in the formation of non-fluorescent 9,10-endoperoxide product ABMA-O 2 ( Figure 7A). This chemical conversion allowed us to follow the intensity loss of fluoresce emission upon photoexcitation. The loss could be associated with the proportional quantity of 1 O 2 produced. The correlation was valid owing to a higher reaction kinetic rate of the trapping process in solution than the internal decay of 1 O 2 in the same solvent system of a DMF-CHCl 3 (1:9, v/v) mixture. Experimentally, the quantity of 1 O 2 generated was monitored and counted by the relative intensity decrease of fluorescence emission (λ em ) of ABMA at 428 nm under excitation wavelength (λ ex ) of 380 nm. At this excitation wavelength, it matched partially with the optical absorption band of DPAF moieties of cis-cup-4-C 2M-9 that led to a slight fluorescence emission (Figure 6c) after the intramolecular energy and the e − -transfer processes. It gave a slightly higher count in the overall FL intensity during the experiment (Figure 6d). In a typical probe reaction, a master solution of ABMA in DMF was diluted by CHCl 3 prior to the addition of cis-cup-4-C 2M-9 . It was followed by periodical illumination using a light emitting diode (LED) lamp of white light (a power output of >2.0 W) operated at two major emission peak maxima (λ max ) centered at 451 and 530 nm. The former light emission spectrum exhibited a sufficient bandwidth covering the 410-470-nm region for photoexcitation of DPAF moieties with optical absorption bands covering 380-500-nm (Figure 4Ae). As a result, we were able to detect rapid production of 1 O 2 by cis-cup-4-C 2M-9 upon irradiation in a decreasing curve profile over a period of more than 120 min (Figure 7Ab).
selected two reliable fluorescent (FL) probes for the detection of either 1 O2 or O2 − ·separately in the solution of cis-cup-4-C2M-9 with high selectivity and specificity as a crucial measure. To detect the former ROS 1 O2, a synthetic highly fluorescent compound α,α'-(anthracene-9,10-diyl)bis(methylmalonic acid) (ABMA) was used as the probe in the experiment. Its UV-vis absorption and fluorescence emission spectra are given in Figures 6a and 6b, respectively. In the probe reaction, chemical trapping of 1 O2 by highly fluorescent ABMA resulted in the formation of non-fluorescent 9,10-endoperoxide product ABMA-O2 ( Figure 7A). This chemical conversion allowed us to follow the intensity loss of fluoresce emission upon photoexcitation. The loss could be associated with the proportional quantity of 1 O2 produced. The correlation was valid owing to a higher reaction kinetic rate of the trapping process in solution than the internal decay of 1 O2 in the same solvent system of a DMF-CHCl3 (1:9, v/v) mixture. Experimentally, the quantity of 1 O2 generated was monitored and counted by the relative intensity decrease of fluorescence emission (λem) of ABMA at 428 nm under excitation wavelength (λex) of 380 nm. At this excitation wavelength, it matched partially with the optical absorption band of DPAF moieties of cis-cup-4-C2M-9 that led to a slight fluorescence emission (Figure 6c) after the intramolecular energy and the e --transfer processes. It gave a slightly higher count in the overall FL intensity during the experiment (Figure 6d). In a typical probe reaction, a master solution of ABMA in DMF was diluted by CHCl3 prior to the addition of cis-cup-4-C2M-9. It was followed by periodical illumination using a light emitting diode (LED) lamp of white light (a power output of >2.0 W) operated at two major emission peak maxima (λmax) centered at 451 and 530 nm. The former light emission spectrum exhibited a sufficient bandwidth covering the 410-470-nm region for photoexcitation of DPAF moieties with optical absorption bands covering 380-500-nm ( Figure 4e). As a result, we were able to detect rapid production of 1 O2 by cis-cup-4-C2M-9 upon irradiation in a decreasing curve profile over a period of more than 120 min (Figure 7Ab). The FL probe experiments were calibrated by a blank control run using the same probe concentration of ABMA alone and an illumination time scale in the absence of cis-cup-4-C 2M-9 (Figure 7Aa). Apparently, we observed slight photodegradation of ABMA itself. This may have implied the existence of a photoinduced triplet state of ABMA in a low quantity due to exposure to short wavelength regions of the light emission bandwidth covering over~380 nm of ABMA absorption bands.  The FL probe experiments were calibrated by a blank control run using the same probe concentration of ABMA alone and an illumination time scale in the absence of cis-cup-4-C2M -9 ( Figure  7Aa). Apparently, we observed slight photodegradation of ABMA itself. This may have implied the existence of a photoinduced triplet state of ABMA in a low quantity due to exposure to short wavelength regions of the light emission bandwidth covering over ~380 nm of ABMA absorption bands.
In the case of detecting superoxide radical as the second ROS, a synthetic O2 -·-reactive fluorescent probe precursor molecule, non-fluorescent potassium bis(2,4-dinitrobenzenesulfonyl)-2′,4′,5′,7′-tetrafluorofluorescein-10′ (or 11′)-carboxylate (DNBs-TFFC), was applied for the experiment. Its molecular structure was synthetically modified from that reported previously [38], showing good reaction selectivity with a high O2 − ·/ 1 O2 sensitivity ratio. Since DNBs-TFFC itself is photodegradable, a dialysis film with the molecular weight cut-off (MWCO) of 100−500 Daltons was applied to hold the solution of cis-cup-4-C2M-9 in toluene-DMSO (9:1, v/v). The sack bag was separated from the solution of the probe DNBs-TFFC. The latter was kept in a cuvette with stirring during the fluorescence emission measurement. Only the solution of cis-cup-4-C2M-9 in the dialysis membrane sack was subjected to the white LED light exposure. Any superoxide radical produced was allowed to rapidly diffuse into the probe solution through the dialysis membrane and initiate the desulfonylation of DNBs-TFFC. The O2 − ·-trapping reaction led to the elimination of two dinitrobenzenesulfonyl moieties and yielded the corresponding bisphenol intermediate, as shown in Figure 7B. Rearrangement of the bisphenol intermediate to the ring-opening of lactone afforded highly fluorescent potassium 2′,4′,5′,7′-tetrafluorofluorescein-10′ (or 11′)-carboxylate regiosisomers (TFFC). The latter compound gave the fluorescence emission at 530 nm (λem) with the excitation at In the case of detecting superoxide radical as the second ROS, a synthetic O 2 − ·-reactive fluorescent probe precursor molecule, non-fluorescent potassium bis(2,4-dinitrobenzenesulfonyl)-2 ,4 ,5 ,7 -tetrafluorofluorescein-10 (or 11 )-carboxylate (DNBs-TFFC), was applied for the experiment. Its molecular structure was synthetically modified from that reported previously [38], showing good reaction selectivity with a high O 2 − ·/ 1 O 2 sensitivity ratio. Since DNBs-TFFC itself is photodegradable, a dialysis film with the molecular weight cut-off (MWCO) of 100−500 Daltons was applied to hold the solution of cis-cup-4-C 2M-9 in toluene-DMSO (9:1, v/v). The sack bag was separated from the solution of the probe DNBs-TFFC. The latter was kept in a cuvette with stirring during the fluorescence emission measurement. Only the solution of cis-cup-4-C 2M-9 in the dialysis membrane sack was subjected to the white LED light exposure. Any superoxide radical produced was allowed to rapidly diffuse into the probe solution through the dialysis membrane and initiate the desulfonylation of DNBs-TFFC. The O 2 − ·-trapping reaction led to the elimination of two dinitrobenzenesulfonyl moieties and yielded the corresponding bisphenol intermediate, as shown in Figure 7B. Rearrangement of the bisphenol intermediate to the ring-opening of lactone afforded highly fluorescent potassium 2 ,4 ,5 ,7 -tetrafluorofluorescein-10 (or 11 )-carboxylate regiosisomers (TFFC). The latter compound gave the fluorescence emission at 530 nm (λ em ) with the excitation at 484 nm (λ ex ). As the probe DNBs-TFFC was not a fluorescent compound, detected emission photon counts were fully associated with the quantity of TFFC produced. Measured total emission intensity counts were then correlated to the relative quantity of O 2 − · generated. As shown in Figure 7Bb, nearly linear progressive increase of fluorescence intensity counts over the full irradiation period was observed that revealed the constant production of O 2 − · from the photoexcited cis-cup-4-C 2M-9 . As discussed above, continuous irradiation on six DPAF moieties of cis-cup-4-C 2M-9 by white LED light (2.0 W) stimulated photoexcitation from the ground to the singlet excited state. Subsequent intramolecular e − -transfer from 1 (DPAF-C n ) * to <C 60 > moieties resulted in the formation of anionic provided the evidence of photoinduced intramolecular e − -transfer mechanism within the 3D-conformer cis-cup-4-C 2M-9 . Furthermore, one of the main crucial criterion for the use of these types of C 60 -(light-harvesting antenna) n conjugates, such as 4-C 2M-9 , as the nano-photosensitizers for antibacterial inactivation (aPDI) is their high photostability. Unlike the conventional organic chromophore-based photosensitizers suffering rapid photodegradation, C 60 -(light-harvesting antenna) n based nano-drugs were found to be capable of a single dose with multiple aPDI/PDT (photodynamic therapy) treatments [39][40][41][42].

Instruments for Spectroscopic Measurements
1 H-NMR spectra were recorded on either Bruker and Spectrospin Avance 500 or Bruker AC-300 spectrometer. UV-vis spectra were recorded on a PerkinElmer Lambda 750 UV spectrometer. Fluorometric traces were collected using a PTI QuantaMaster TM 40 Fluorescence Spectrofluorometer. The light source used in this experiment included a collimated white LED light with an output power of 2.0 W (Prizmatix, Southfield, MI, USA). Infrared spectra were recorded as KBr pellets on a Thermo Nicolet AVATAR 370 FTIR spectrometer. Cyclic voltammetry (CV) was record on EG&G Princeton Applied Research 263A Potentiostat/Galvanostat using Pt metal as the working electrode, Ag/AgCl as the reference electrode, and Pt wire as the counter electrode at a scan rate of 10 mV/s. The solution for CV measurements was prepared in a concentration of 1.0-5.0 × 10 −3 M in appropriate solvents containing the electrolyte Bu 4 N + -PF 6 − (0.1 M).
3.3. Synthesis of N 1 ,N 3 ,N 5 -Tris(9,9-di(methoxyethyl)fluoren-2-yl)-1",3",5"-tris(phenylamino)-benzene as Tris(DPAF-C 2M ) (2-C 2M ) Synthetic procedure for the preparation of tris(DPAF-C 2M ) was slightly modified from those methods reported recently [34]. In general, a mixture of BrF-C 2M (7.33 g, 20.3 mmol, excess), TPAB (1.16 g, 3.3 mmol), and sodium t-butoxide (1.94 g, 20.3 mmol) was dissolved in anhydrous toluene (75 mL) and stirred for 1 h to give a homogeneous solution. The catalyst Pd 2 (dba) 3 (0) (0.023 g, 0.25 mol%) and rac-BINAP (0.046 g, 0.75 mol%) were added to the solution, followed by heating to refluxing temperature under nitrogen for a period of 72 h. After cooling the resulting mixture to room temperature, it was washed with water three times by extraction, the organic layer was separated, and it was dried over sodium sulfate. After solvent evaporation, a small quantity of crude paste was tested on the TLC plate to show the major product at R f = 0.6 using hexane-ethylacetate (1:1, v/v) as the eluent. This product spot had a dense yellow-brown color on the top accompanied by a light visible tail. The tail portion was assumed to be the product in the trans-chair form. This tail portion was subsequently separated from the main top portion of the cis-cup form via column chromatography, followed by the TLC plate purification using silica gel as the stationary phase and hexane-ethylacetate from methanol (100 mL) to afford the crude product mixture, which was isolated by centrifugation. Further purification by column chromatography (silica gel) using toluene to a solvent mixture of toluene-EtOAc (7:3, v/v) as the eluent with sequential increments of increasing solvent polarity afforded cis-cup-tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )]. It was further purified by TLC with isolation of only the narrow dense color fraction band to give brown solids of cis-cup-4-C 2M-9 in 79% yield (0.74 g) (at

ROS Measurements Using singlet oxygen ( 1 O 2 )-Sensitive Fluorescent Probe
The compound α,α'-(anthracene-9,10-diyl)bis(methylmalonic acid) (ABMA) was used as a fluorescent probe for singlet oxygen ( 1 O 2 ) trapping. The quantity of 1 O 2 generated was monitored and counted by the relative intensity decrease of fluorescence emission of ABMA at 428 nm under excitation wavelengths of 380 nm (λ ex ). A typical probe solution was prepared by diluting a master solution of ABMA (1.0 × 10 −5 M in DMF, 0.4 mL) with an amount 9-fold in volume of CHCl 3 (3.2 mL) in a cuvette (10 × 10 × 45 mm). The solution was added by a pre-defined volume of tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] in CHCl 3 (1.0 × 10 −5 M, 0.4 mL), followed by periodic illumination using an ultrahigh power white light LED lamp (Prizmatix, operated at the emission peak maxima centered at 451 and 530 nm with the collimated optical power output of >2.0 W in a diameter of 5.2 cm). Progressive fluorescent spectra were taken on the PTI QuantaMaster TM 40 Fluorescence Spectrofluorometer. was counted in association with its reaction with DNBs-TFFC that resulted in the product of highly fluorescent potassium 2 ,4 ,5 ,7 -tetrafluorofluorescein-10 (or 11 )-carboxylate regioisomers (TFFC) with fluorescence emission at 530 nm (λ em ) upon excitation at 484 nm (λ ex ). The detected intensity increase of fluorescence emission was then correlated to the relative quantity of O 2 − produced.

Conclusions
Previous studies on detected photoswitchable dielectric amplification phenomena by the simulated ferroelectric-like capacitor design using the construction and the fabrication of a fullerosome shell layer on core-shell γ-FeO x @AuNPs were based on the photoinduced intramolecular charge-polarization of (C 60 > acceptor)-(DPAF-C n donor) conjugates [16][17][18]. The corresponding formation of dielectric ion-radical components (C 60 >) − and DPAF + ·-C n within a fullerosome array layer was the basis of observed dielectric properties. Accordingly, several such conjugates were developed by the extension from the initial C 60 (>DPAF-C 9 ) to demonstrate the correlation of the structural relationship and the chemical modifications to the enhanced dielectric properties, as stated above. Two interesting modifications both involved 3D-conformeric C 60 (>DPAF-C 9 ) derivatives by fusing three phenyl rings of three diphenylamino groups together to form a shared central benzene moiety as the base of 3D configuration design. Specifically, the successful synthesis of cis-cup-tris[(DPAF-C 2M )-C 60 (>DPAF-C 9 )] stereoisomer may be beneficial for use as positive (DPAF-C n ) + and negative charge (<C 60 >) − carriers in enhancing photoinduced dielectric characteristics [43]. It can also be applied as the precursor building block in the synthesis of several C 60 -and C 70 -based ultrafast photoresponsive nonlinear two-photon absorptive nanomaterials. Accordingly, we demonstrated efficient intramolecular energy and electron transfer capabilities of 3D conformer cis-cup-4-C 2M-9 using photophysical measurements and its effective production of singlet oxygen (via the energy transfer mechanism) and superoxide radicals (via the electron transfer mechanism). They can be applied as nano-photosensitizers [39][40][41][42] and nonlinear photonic agents [30][31][32]44].