Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2020 Nov 30;27(5):1655–1669. doi: 10.1002/chem.202003232

Smart Shell‐by‐Shell Nanoparticles with Tunable Perylene Fluorescence in the Organic Interlayer

Lisa M S Stiegler 1, Stefanie Klein 2, Carola Kryschi 2, Winfried Neuhuber 3, Andreas Hirsch 1,✉
PMCID: PMC7898710  PMID: 33459437

Abstract

A new series of shell‐by‐shell (SbS)‐functionalized Al2O3 nanoparticles (NPs) containing a perylene core in the organic interlayer as a fluorescence marker is introduced. Initially, the NPs were functionalized with both, a fluorescent perylene phosphonic acid derivative, together with the lipophilic hexadecylphosphonic acid or the fluorophilic (1 H,1 H,2 H,2H‐perfluorodecyl)phosphonic acid. The lipophilic first‐shell functionalized NPs were further implemented with amphiphiles built of aliphatic chains and polar head‐groups. However, the fluorophilic NPs were combined with amphiphiles consisting of fluorocarbon tails and polar head‐groups. Depending on the nature of the combined phosphonic acids and the amphiphiles, tuning of the perylene fluorescence can be accomplished due variations of supramolecular organization with the shell interface. Because the SbS‐functionalized NPs dispose excellent dispersibility in water and in biological media, two sorts of NPs with different surface properties were tested with respect to biological fluorescent imaging applications. Depending on the agglomeration of the NPs, the cellular uptake differs. The uptake of larger agglomerates is facilitated by endocytosis, whereas individualized NPs cross directly the cellular membrane. Also, the larger agglomerates were preferentially incorporated by all tested cells.

Keywords: amphiphiles; fluorescent imaging; hydro-, lipo-, fluoro-philic/phobic; perylene chromophore; shell-by-shell (SbS)-functionalization

Adjustable fluorescence features of a perylene unit in the organic interlayer of SbS‐functionalized Al2O3 NPs were described. A perylene terminated phosphonic acid is attached covalently onto the NP surface, amphiphiles generated the second ligand shell. Depending on the type of amphiphile, changes in the initial fluorescence properties of the perylene were observed, that could be matched to aggregation, disaggregation, or electronic effects.

graphic file with name CHEM-27-1655-g024.jpg

Introduction

Over the last years, our group established for the first time a highly hierarchical structured functionalization concept for metal oxide NPs (Scheme 1), which is based on a combination of two functionalization steps. [1] In the first step, a wet‐chemical functionalization process takes place, where a metal oxide NP surface is treated with phosphonic acid derivatives, resulting in covalent binding between the organic molecules and the inorganic core. [2] In the next step, an attachment of amphiphiles around first‐shell functionalized NPs follows, which is based on the self‐assembly provided by solvophobic interactions of the apolar moiety of the amphiphiles interdigitating into the first ligand shell of the NPs. The polar moieties pointed outwards and create water‐dispersible character of the corresponding hybrids. [3]

Scheme 1.

Scheme 1

Open in a new tab

General structure of SbS‐functionalized NPs.

This field of NP functionalization was extended recently, with new implications and applications. First, this functionalization principle of metal oxide NPs was used to generate a switchable dispersibility behavior of NPs in fluorocarbons, hydrocarbons, and water and further allows the tuning of surface energies of NP surfaces, for example, from hydrophilic to superhydrophobic, from lipophilic to lipophobic, from fluorophilic to fluorophobic. [4] We have also established a method, in which the fluorescence of a pyrene core coupled to a phosphonic acid attached covalently to a Al2O3 NP surface, could be quenched and turned on and off. [5] Therefore, the spacing phosphonic acids and amphiphiles were carefully selected in order to control the intermolecular interaction of the NP surface. By introducing an amphiphile with a pyridinium head‐group, the fluorescence of the pyrene moiety was turned off and by removing the amphiphile from the SbS‐structure, the fluorescence was switched on, again. Compounds involving pyridinium units cause a decrease in the fluorescence of alternant polycyclic aromatic hydrocarbons (PAHs), like pyrene or perylene, in polar solvents such as water and acetonitrile. [6] Responsible for the quenching is a photo‐induced electron‐transfer (PET) from the LUMO of the PAHs excited state towards the LUMO of the positively charged nitrogen atom of the pyridinium unit, that has a lower energy level. [7] Recently, we demonstrated that the type and nature of spacing phosphonic acids influences the aggregation of perylene bisimides (PBIs) onto Al2O3 NP surfaces and thus the optical properties of the PBIs. [8] Whereas the rigid fluorocarbon phosphonic acids separate the PBI units very effectively, interactions of PBIs with hydrocarbon phosphonic acids are less hindered. The highly flexible glycol phosphonic acids lead to the weakest hindered interaction causing strong aggregation. We also reported on SbS‐functionalized NPs with electronic communication between the first and second‐layer interface. [9] In a lego‐type principle, electron‐donors were combined with electron‐acceptors, which resulted on one hand in electronical communications between the chromophores with different electron demand and on the other hand, the optical properties of the NPs changed. Because of the excellent water dispersibility and stability of SbS‐functionalized NPs with interdigitating amphiphiles containing polar head‐groups, also some biological applications were already reported. [10] Initially coated hydrophobic Au‐Fe3O4 nanoheterodimers (NHDs) that were capped with oleic acid, were encapsulated in a self‐assembled bilayer shell formation with 1‐octadecylpyridinium, or 4‐dodecylbenzenesulfonate, to provide a positive or negative surface charge onto the NP surface. We recognized, that the surface charge as well as the surface architecture of the NHDs influenced the mechanism and the efficiency of the cellular uptake pathway, cellular localization, and the toxicity of human tumor (MCF‐7) and healthy epithelial (MCF‐10 A) cells. [10]

Herein, we report on a series of SbS‐functionalized Al2O3 NPs, involving a perylene core as a fluorescence marker in the first ligand shell in combination with spacing phosphonic acids and amphiphiles of different nature resulting in diverse mutations of the initial perylene fluorescence. We created a hydrocarbon as well as a fluorocarbon analogue of Al2O3 NPs, by combining 10 % PAR1 with 90 % PAR2, resulting in Al2O3‐(PAR1 10 % PAR2 90 % ) NPs, or in the case of combining 10 % PAR1 with 90 % PAR3, Al2O3‐(PAR1 10 % PAR3 90 % ) NPs were formed. The hydrocarbon analogy was further implemented with amphiphiles built of aliphatic chains and polar head‐groups (SDBS, SDS, DDPB, DTAB, HC‐PEG), generating lipophilic environment around the perylene units. The fluorocarbon analogy was further implemented with amphiphiles consisting of fluorocarbon tails and polar head‐groups (PFUnDA, FC‐PB, FC‐PEG), generating fluorophilic pockets. This allowed for tailoring the perylene fluorescence, from drastic increase of fluorescence by disaggregation of perylene excimers [11] onto the NP surface up to fluorescence quenching by forcing the perylene units in its excimeric form [12]. Drastic decrease of fluorescence was observed by combining the perylene functionalized NPs with amphiphiles containing a pyridinium head‐group. The hydrocarbon analogy of the particles that were functionalized with SDBS or DTAB (contrary surface charge), namely [(Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS and [(Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB, were further used in fluorescence microscopy of the non‐cancerous human umbilical vein endothelial cells (HUVEC) and breast epithelial cells (MCF‐10 A) as well of the cancerous A549 (lung cancer) and MDA‐MB‐231 (breast cancer) cells. Further their biocompatibility, cellular uptake pathway and the intracellular NP content was studied.

Results and Discussion

Synthesis of the parent compounds

The stepwise synthesis of the so far unknown phosphonic acid PAR1 is outlined in Scheme 4. Whereas compounds 1 and 2 were known, [5] synthetic access to 3 and PAR1 had be established. The precursor molecule 1 was synthesized by a Michaelis–Arbuzov reaction of 1,12‐dibromododecane with triethylphosphite and isolated as a pure compound in 26 % yield. Compound 2 was generated by a nucleophilic substitution of the bromide in 1 with sodium azide in 96 % yield. The azide functionality of the phosphonate‐ester was subsequently subjected to a “click” reaction with 3‐ethynyl perylene leading to the formation of 3 in 34 % yield.[ 5 , 9 , 13 ] Final deprotection of 3 afforded the target molecule PAR1 in 80 % yield.

Scheme 4.

Scheme 4

Open in a new tab

Synthesis of the phosphonic acid PAR1.

Whereas the amphiphiles SDBS, SDS, DTAB, and PFUnDA are commercially available, we synthesized DDPB, HC‐PEG, FC‐PB, and FC‐PEG according to literature procedures.[ 4a , 5 , 14 ] The syntheses of the amphiphiles are outlined in Scheme 5. DDPB was synthesized by implementation of n‐dodecyl bromide with pyridine and was received as a pure compound in 57 % yield. [14a] FC‐PB was generated via a two‐step synthesis. In the first step, a Steglich esterification of 11‐bromoundecanoic acid with heptadecafluorodecan‐1‐ol was performed, resulting in molecule 4 in 43 % yield. Subsequently, 4 was refluxed in pyridine, resulting in 40 % yield of FC‐PB. [5] HC‐PEG was afforded by implementation of lauric acid chloride with triethylene glycol, the reaction was catalyzed by triethylamine. The product was obtained in 61 % yield. Towards FC‐PEG, firstly, perfluoro‐n‐undecanoic acid was converted to the corresponding perfluoro‐n‐undecanoic acid chloride 5 by the help of thionyl chloride. Subsequently, 5 reacted with triethylene glycol, catalyzed by triethylamine, resulting in pure FC‐PEG in 13 % yield. [4a]

Scheme 5.

Scheme 5

Open in a new tab

Synthesis of the amphiphiles DDPB, FC‐PB, HC‐PEG, and FC‐PEG.

Assembly of the hydrocarbon SbS‐functionalized NPs

The optical properties of the hydrocarbon SbS‐functionalized Al2O3‐(PAR1 10 % PAR2 90 % ) NPs during the formation of a lipophilic pocket among the perylene units caused by the assembly of amphiphiles, were studied (Scheme 2, Scheme 3). The Al2O3 NPs are commercially available. As obtained from BET measurements they have an average diameter size of 14 nm. DLS measurements of the pristine NPs in IPA show a hydrodynamic diameter of <50 nm whereas the SbS‐functionalized NPs in deionized water reach a hydrodynamic diameter of around 100 nm. Additional information on the Al2O3 NP core is given in the supporting information. For the functionalization procedure, 7.5 mL 0.15 wt % Al2O3 NP dispersions in isopropanol were mixed with 5 mL of a phosphonic acid solution in MeOH/toluene (1:1), resulting in a total concentration of the phosphonic acid mixture of 5 mm. Subsequently, 30 min. of ultrasonication and repeated washing with isopropanol/toluene/isopropanol followed, to achieve homogeneous surface functionalization and to remove excess of phosphonic acid. The value of 5 mm has already been proven for providing a full surface coverage of such metal oxide NPs with phosphonic acids.[ 9 , 15 ] With the results of TGA measurements, grafting densities of 4.29 nm−2 for the Al2O3‐(PAR1 10 % PAR2 90 % ) hybrids and 3.02 nm−2 for the Al2O3‐(PAR1 10 % PAR2 90 % ) hybrids were obtained (see Figure S3). The fact, that phosphonic acid mixtures can be stoichiometrically displayed onto metal oxide surfaces, was already corroborated in recent studies.[ 4b , 5 , 8 , 9 , 15 , 16 ] TGA measurements of Al2O3 NPs functionalized with different ratios of PAR1 and PAR2/PAR3 were shown in the supporting information (Figure S3) and displayed the differences in the grafting densities as well as in the DTG curves. The coverage with PAR1 (Scheme 4) and PAR2 provides a lipophilic surface.[ 4a , 17 ] The second ligand shell was assembled through the non‐covalent interdigitation of the amphiphiles fused of a hydrocarbon moiety and a polar head‐group, namely sodium dodecylbenzenesulfonate (SDBS), sodium dodecylsulfate (SDS), 1‐dodecylpyridinium bromide (DDPB), (1‐dodecyl)trimethylammonium bromide (DTAB), or 2‐(2‐(2‐hydroxyethoxy)ethoxy)ethyl dodecanoate (HC‐PEG, Scheme 5). The amphiphiles comprising the second ligand shell caused a polarity umpolung from lipophilic (PAR2) to hydrophilic causing water‐dispersibility. [4a] The formation of the second shell was monitored via fluorescence measurements. Therefore, 2 mL of 0.015 wt % Al2O3‐(PAR1 10 % PAR2 90 % ) were filled in a cuvette and were titrated with 0.015 wt % [Al2O3‐(PAR1 10 % PAR2 90 % )]20 mM amphiphile dispersions, resulting in a concentration range from zero up to 10 mm amphiphile with a constant 0.015 wt % concentration of first‐shell functionalized NPs. The constant concentration of first‐shell functionalized NPs is crucial for excluding dilution effects during titration, so that the changes in the fluorescence spectra can be assigned exclusively to the addition of amphiphile. Exactly the same titration procedure was applied for all other examples (Scheme 2, Scheme 3).

Scheme 2.

Scheme 2

Open in a new tab

Hydrocarbon and fluorocarbon based SbS‐architectures consisting of aluminum oxide NP cores, the fluorescence marker PAR1, the spacing phosphonic acids PAR2 and PAR3, as well as the hydrocarbon‐polar fused amphiphiles SDBS, SDS, DTAB, HC‐PEG and the fluorocarbon‐polar fused amphiphiles PFUnDA, FC‐PB, FC‐PEG (see also Scheme 3).

Scheme 3.

Scheme 3

Open in a new tab

Building blocks for the SbS‐functionalization of aluminum oxide NPs with I) a mixture of PAR1 and PAR2 with the amphiphiles SDBS, SDS, DDPB, DTAB, HC‐PEG (left column) and II) PAR1 and PAR3 with PFUnDA, FC‐PB, FC‐PEG (right column).

Figure 1 a, b, c shows the fluorescence and UV/Vis spectra of Al2O3‐(PAR1 10 % PAR2 90 % ) NP dispersions, that were titrated with [Al2O3‐(PAR1 10 % PAR2 90 % )]20 mM SDBS dispersions (Scheme 6). The NPs Al2O3‐(PAR1 10 % PAR2 90 % ) without SDBS give raise to weak fluorescence signals with weak and undefined 0–0* and 0–1* transition bands and a missing 0–2* transition band. During titration with SDBS, the overall fluorescence intensity of the perylene unit increased. The intensity of the 0–0* transition at 469 nm increased strongest, then the intensity of the 0–1* transition at 499 nm, and also the signal of the 0–2* transition at 537 nm appeared. By titrating SDBS to the Al2O3‐(PAR1 10 % PAR2 90 % ) NPs, of course, the intensities of the absorption bands of the benzene moieties of SDBS at 255 and 261 nm increased, whereas the characteristic absorption signals of the perylene core at 392, 421, and 448 nm stayed unaffected. Because of an unchanged concentration of perylene during titration, the molecular density of the perylene core stayed constant (see Figure 1 c). The same trend of intensifying fluorescence was observed for Al2O3‐(PAR1 10 % PAR2 90 % ) NPs, that were titrated with SDS or DTAB (Figure S5 and S8, Supporting Information). The UV/Vis spectra of those two systems stayed unaffected during the titration process, because of the optical inactive amphiphiles.

Figure 1.

Figure 1

Open in a new tab

(a) Fluorescence spectra, (b) trend of the fluorescence spectra, and (c) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR2 90 % ) titrated with SDBS in DIW (λ ext=408 nm).

Scheme 6.

Scheme 6

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs.

Further experiments on the optical properties of Al2O3‐(PAR1 10 % PAR2 90 % ) NPs focussing the perylene fluorescence, were done by using DDPB as a second‐shell amphiphile (see Scheme 7). Figure 2 a and b show that already after the first titration step with DDPB, resulting in a concentration of 0.10 mm, the fluorescence intensity of the perylene core drop. After the second titration step, the fluorescence intensity decreased more moderate and stayed constant at the same level during further titration with DDPB (see Figure 2 b). The UV/Vis spectra in Figure 2 c show an increase of the pyridinium absorption signal at 259 nm and constant perylene absorption bands during titration of Al2O3‐(PAR1 10 % PAR2 90 % ) with DDPB.

Scheme 7.

Scheme 7

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR2 90 % )]DDPB NPs with insights into the electronical behavior of the structure.

Figure 2.

Figure 2

Open in a new tab

(a) Fluorescence spectra, (b) trend of the fluorescence spectra, and (c) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR2 90 % ) titrated with DDPB in DIW (λ ext=408 nm).

The differences in the optical properties of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS and [Al2O3‐(PAR1 10 % PAR2 90 % )]DDPB NPs can also be seen from the visual appearance of the corresponding dispersions themselves. Figure 3 a show the NP dispersions [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS (right side) and [Al2O3‐(PAR1 10 % PAR2 90 % )]DDPB (left side) under day‐light conditions, both dispersions dispose of equivalent yellow colors. Under UV‐light conditions (see Figure 3 b), the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs show a strong yellowish‐green fluorescence color, whereas the [Al2O3‐(PAR1 10 % PAR2 90 % )]DDPB NPs show a quenched fluorescence with a slightly orange fluorescence color.

Figure 3.

Figure 3

Open in a new tab

Optical appearance of [Al2O3‐(PAR2 10 % PAR1 90 % )]DDPB (left side) and [Al2O3‐(PAR2 10 % PAR1 90 % )]SDBS (right side) under (a) day‐light and (b) UV‐light conditions.

In a next step, the nonionic amphiphile HC‐PEG, was used to generate another family of SbS‐functionalized NPs, namely [Al2O3‐(PAR1 10 % PAR2 90 % )]HC‐PEG (see Scheme 8). Figure 4 a show the fluorescence measurements of Al2O3‐(PAR1 10 % PAR2 90 % ) NPs that were titrated with HC‐PEG and in Figure 4 b the trend of the fluorescence intensity of the Al2O3‐(PAR1 10 % PAR2 90 % ) NPs in dependence on the HC‐PEG concentration is shown. After the first titration step, resulting in a concentration of 0.10 mm HC‐PEG, the fluorescence band of the 0–0* transition (392 nm) strongly increased, also the fluorescence band of the 0–1* transition (421 nm) increased, but more moderately compared to the 0–0* transition. Additionally, the fluorescence signal of the 0–2* transition (448 nm) arise. Extra amount of HC‐PEG had no further influence on the fluorescence properties of the NPs, the fluorescence signals of the perylene core stayed constant after the first titration step. In Figure 4 c, the UV/Vis spectra of [Al2O3‐(PAR1 10 % PAR2 90 % )]HC‐PEG NPs during increasing amount of HC‐PEG, were shown. From a concentration range from 0 mm to 0.20 mm HC‐PEG, the initial UV/Vis spectrum stayed unaffected by increasing the amount of HC‐PEG in the NP dispersion. During further titration with amphiphile, beginning at 0.29 mm HC‐PEG, the overall absorption of the NPs increased and cloudy dispersions were formed.

Scheme 8.

Scheme 8

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR2 90 % )]HC‐PG NPs with excess of the HC‐PG amphiphile forming NP‐independent micelles.

Figure 4.

Figure 4

Open in a new tab

(a) Fluorescence spectra, (b) trend of the fluorescence spectra, and (c) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR2 90 % ) titrated with HC‐PEG in DIW (λ ext=408 nm).

Assembly of the fluorocarbon SbS‐functionalized NPs

As an alternative to Al2O3‐(PAR1 10 % PAR2 90 % ) NPs, the fluorinated pendants were manufactured. The functionalization procedure was the same, but instead of using 90 % PAR2, 90 % PAR3 was used as a spacing phosphonic acid, resulting in Al2O3‐(PAR1 10 % PAR3 90 % ) NP hybrids. The predominant coverage with PAR3 provides a fluorophilic surface (Scheme 2, Scheme 3).[ 4a , 17 ] For the second functionalization step, amphiphiles with a fluorocarbon tail were used, for providing a successful assembly of the second layer around first‐shell functionalized NPs. For this purpose we selected perfluoroundecanoic acid (PFUnDA), 1‐(11‐((3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10‐heptadecafluorodecyl)oxy)‐11‐oxoundecyl)pyridin‐1‐ium bromide (FC‐PB), or 2‐(2‐(2‐hydroxyethoxy)ethoxy)ethyl 2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11‐henicosafluoroundecanoate (FC‐PEG).

In comparison to the hydrocarbon analogy Al2O3‐(PAR1 10 % PAR2 90 % ), the Al2O3‐(PAR1 10 % PAR3 90 % ) NPs showed a better refined fluorescence spectrum with a more intense 0–0* transition (458 nm) compared to the 0–1* transition (484 nm). But the 0–2* transition band is also hidden. Figure 5 a and b represent the fluorescence and UV/Vis measurements of Al2O3‐(PAR1 10 % PAR3 90 % ) NPs that were titrated with PFUnDA (Scheme 9). After the first titration step of Al2O3‐(PAR1 10 % PAR3 90 % ) with PFUnDA, resulting in a concentration of 0.10 mm, the perylene fluorescence was quenched and during further titration, the fluorescence band resulted in a broad undefined band that was shifted to higher wavelength. Starting from a concentration of 0.58 mm PFUnDA, the fluorescence spectrum did not further change. In the UV/Vis spectra of the [Al2O3‐(PAR1 10 % PAR3 90 % )]PFUnDA NPs, shown in Figure 5 b, it can be seen, that the optical density of the perylene core was unaffected during the titration experiment with PFUnDA.

Figure 5.

Figure 5

Open in a new tab

(a) Fluorescence spectra, (b) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR3 90 % ) titrated with PFUnDA in DIW (λ ext=408 nm).

Scheme 9.

Scheme 9

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR3 90 % )]PFUnDA NPs, whereas PFUnDA formed a fluorophilic pocket around the perylene units.

Next, [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PB NPs were manufactured involving FC‐PB as a special type of amphiphile within the second shell (see Scheme 10). FC‐PB consists of a fluorocarbon moiety on one molecular end, the other end consists of a polar pyridinium bromide head‐group generating the water‐dispersibility. The fluorocarbon moiety and the polar head‐group were bridged through a hydrocarbon spacer. That extraordinary type of amphiphile caused special fluorescence variations of Al2O3‐(PAR1 10 % PAR3 90 % ) NPs during the titration course with FC‐PB. Figure 6 a, b, and c show detailed fluorescence measurements. The first titration step with the amphiphile applied in a concentration of 0.10 mm FC‐PB, lead to a decreased perylene fluorescence. In the later course of the titration, the fluorescence of the perylene unit increased stepwise with increasing amount of FC‐PB. Noticeable is also the fact, that the fluorescence band of the 0‐0* transition of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PB is shifted around 12 nm to higher wavelength, compared to the system Al2O3‐(PAR1 10 % PAR3 90 % ) without amphiphile. Figure 6 d represent the UV/Vis spectra of the NP system Al2O3‐(PAR1 10 % PAR3 90 % ) during titration with amphiphile FC‐PB. The absorption signal of the pyridinium unit at 259 nm increased, whereas the optical density of the perylene core was constant during the titration experiment.

Scheme 10.

Scheme 10

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PB NPs with insights into the electronical behavior of the system.

Figure 6.

Figure 6

Open in a new tab

(a), (b) Fluorescence spectra, (c) trend of the fluorescence spectra, and (d) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR3 90 % ) titrated with FC‐PB in DIW (λ ext=408 nm).

As a third example to study the optical properties of Al2O3‐(PAR1 10 % PAR3 90 % ) NPs during assembly of a second shell, the nonionic amphiphile FC‐PEG was used (see Scheme 11). In Figure 7 a and b the monitoring of the fluorescence and UV/Vis spectra of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PEG NPs in dependence on the amphiphile concentration is displayed. During titration with FC‐PEG, the overall perylene fluorescence decreased. The fluorescence band of the 0–0* transition (458 nm) decreased stronger than that of the 0–1* transition (484 nm) and beginning with a concentration of 1.48 mm FC‐PEG, there is just the leftover of a broad undefined fluorescence band. By looking at the UV/Vis spectra (see Figure 7 b), no changes in the optical density of the perylene core were observed, because of a stable concentration of Al2O3‐(PAR1 10 % PAR3 90 % ) during titration with the optical inactive amphiphile FC‐PEG.

Scheme 11.

Scheme 11

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PEG NPs, whereas FC‐PEG formed a fluorophilic pocket around the perylene core.

Figure 7.

Figure 7

Open in a new tab

(a) Fluorescence spectra, and (b) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR3 90 % ) titrated with FC‐PEG in DIW (λ ext=408 nm).

Assembly of hydrocarbon SbS‐functionalized NPs including a mixed second ligand shell

In a next step of complexity, a hydrocarbon SbS‐architecture was generated including a mixed second ligand shell of DDPB and SDBS. That system is of high importance for comparative studies, because theoretically, it combines a fluorescence quencher and a fluorescence intensifier in one system. For this purpose, the first‐shell functionalized hydrocarbon analogue Al2O3‐(PAR1 10 % PAR2 90 % ) was mixed with the second‐shell ligands DDPB and SDBS in a ratio of 1:1 (Scheme 12), 1:2, or 2:1). It can be assumed, that the amphiphiles were attached roughly stoichiometrically onto the first‐shell functionalized nanoparticles, as indicated by zeta‐potential measurements of the mixed second‐shell systems (see Supporting Information, Table S4). The fluorescence measurements showed in all three cases the same trend. There is one significant quenching step which could be matched to DDPB (Figure 8 a, b) and in the further course of the titration, the fluorescence increased, initiated by SDBS (see Figure 8 a and c, S6 and S7). Beginning at a total concentration of 1.13 mm DDPB and SDBS (1:1), the NPs formed flaky dispersions indicating that the electrosteric stabilization of the NPs could no longer be provided. Regarding the UV/Vis spectra, starting from a concentration of 0.86 mm DDPB‐SDBS mixture, the absorption signal at 260 nm stayed constant.

Scheme 12.

Scheme 12

Open in a new tab

Schematic representation of the formation of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS&DDPB NPs including a mixed second layer consisting of SDBS and DDPB in a one‐to‐one ratio.

Figure 8.

Figure 8

Open in a new tab

(a), (b) Fluorescence spectra, (c) trend of the fluorescence spectra, and (d) UV/Vis spectra of Al2O3‐(PAR1 10 % PAR2 90 % ) titrated with DDPB & SDBS (1:1) in DIW (λ ext=408 nm).

Comparative analysis and theoretical interpretation of the obtained fluorescence studies during SbS‐functionalization

First of all, the fact, that both starting materials, the lipophilic Al2O3‐(PAR1 10 % PAR2 90 % ) and the fluorophilic Al2O3‐(PAR1 10 % PAR3 90 % ) NPs vary in their fluorescence spectra, is striking. Whereas Al2O3‐(PAR1 10 % PAR2 90 % ) NPs give rise to undefined transition bands of the perylene core, Al2O3‐(PAR1 10 % PAR3 90 % ) NPs displayed more defined fluorescence signals. The undefined transition bands of Al2O3‐(PAR1 10 % PAR2 90 % ) NPs are characteristic for excimer formation of the perylene [12] units. Fluorocarbon moieties are more rigid than hydrocarbons [18] and are thus more suitable for separating the perylene cores onto the NP surface and prevent excimer formation, resulting in more structured fluorescence signals of the Al2O3‐(PAR1 10 % PAR3 90 % ) NPs. This has already been seen for pyrene and perylene bisimide functionalized Al2O3 NPs by using either hydrocarbon or fluorocarbon spacing phosphonic acids.[ 5 , 8 ]

The higher hierarchically assembled SbS‐system [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS displayed an increasing fluorescence intensity of Al2O3‐(PAR1 10 % PAR2 90 % ) during addition of SDBS, with refined 0–0*, 0–1* and 0–2* transition bands (Figure 1 a and b). Very similar results were observed for the titrations of Al2O3‐(PAR1 10 % PAR2 90 % ) with SDS and DTAB (see Figure S5 and S8, Supporting Information). During all three titration experiments, the form of the fluorescence spectra aligned to the reference fluorescence spectrum of the monomeric compound 3 in THF (Figure S4a). This is a proof, that the lipophilic moiety of SDBS (or SDS/DTAB) was incorporated into the first ligand shell, and lead to disaggregation of the perylene excimers [11] onto the NP surface.

By doing the same experiment with DDPB instead of SDBS, the fluorescence of the Al2O3‐(PAR1 10 % PAR2 90 % ) NPs was strongly quenched (Figure 2 a and b). Reason for that phenomenon is an electronic communication between the pyridinium and the perylene unit. Pyridinium quenches the fluorescence of the perylene through a photo‐induced electron‐transfer (PET) from the excited state of the perylene, that was formed during excitation with light of 408 nm, towards the nitrogen atom of the pyridinium, that has a lowered LUMO energy level compared to that of the perylene.[ 6a , 6b , 6d , 6e , 19 ]

By creating mixed second ligand shells, a combination of the amphiphiles SDBS and DDPB in a ratio of 1:1, 1:2, or 2:1, was titrated to Al2O3‐(PAR1 10 % PAR2 90 % ) NPs, while a new combination of effects has been arisen (Figure 8 a and b, S6 and S7). After the first titration step, the fluorescence was quenched which could be matched to the DDPB that generated a PET from the perylene core to the pyridinium unit. By further titration, the effect of SDBS was predominant and the perylene fluorescence increased again, initiated by disaggregation of the perylene units onto the NP surface.

Manufacturing of [Al2O3‐(PAR1 10 % PAR2 90 % )]HC‐PEG NPs lead to an increased fluorescence intensity with well resolved fluorescence signals (Figure 3 a and b), merely after the first titration step of Al2O3‐(PAR1 10 % PAR2 90 % ) with HC‐PEG, resulting in a HC‐PEG concentration of 0.10 mm. This fluorescence spectrum is already comparable to the monomeric fluorescence spectrum of reference compound 3 in THF (Figure S4a). Further addition of HC‐PEG had no further influence on that fluorescence spectrum. These observations proof the successful formation of a lipophilic pocket among the perylene units and following separation of excimers. The single‐step increase of the fluorescence and the formation of cloudy dispersions by further addition of HC‐PEG indicate that pure micelles of the amphiphile were formed without interaction to the NP surface. This explanation could be strengthened by taking into account, that tetraethylene glycol monododecyl ether (C12EO4), an amphiphile with comparable molecular length and polarity to HC‐PEG, has also a relatively low critical micelle concentration of around cmc=0.046 mm. [20]

Coming to the fluorocarbon system, the first titration experiment was performed with Al2O3‐(PAR1 10 % PAR3 90 %) and PFUnDA. During the formation of the SbS‐system [Al2O3‐(PAR1 10 % PAR3 90 % )]PFUnDA, it could be observed that the perylene fluorescence was quenched, already beginning at a concentration of 0.10 mm PFUnDA (see Figure 5 a, b). Further addition of PFUnDA resulted in a broad undefined fluorescence band that was shifted to higher wavelengths, characteristic for excimers.[ 12a , 21 ] At a concentration of 0.58 mm PFUnDA, the fluorescence spectrum stayed constant, at this point, the surface seems fully covered with amphiphile. Reason for the changes in the fluorescence spectra was a formation of a fluorophilic pocket around the perylene cores through incorporation of the fluorocarbon moiety of PFUnDA into the mainly fluorinated first‐shell of the NPs (see Scheme 9). During this process, a phase separation of the hydrocarbons (PAR1) from the fluorocarbons (PAR3/PFUnDA) [22] onto the NP surface took place, which leads to a stronger interaction of the perylene units forming stacked complexes, so called excimers of the perylene, [12] resulting in a reduced fluorescence intensity with a limited resolution of the characteristic fluorescence bands of the perylene.

During the formation of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PB, the fluorescence of Al2O3‐(PAR1 10 % PAR3 90 % ) was quenched after addition of 0.10 mm FC‐PB, while the fluorescence band of the 0–0* transition of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PB is shifted around 12 nm to higher wavelength, compared to the system Al2O3‐(PAR1 10 % PAR3 90 % ) without amphiphile. The quenching effect as well as the shift of the fluorescence band are signals for an electronical communication between the perylene core and the pyridinium unit. The forthgoing titration displayed a stepwise increase of the fluorescence with increasing amount of FC‐PB. It is thus comparable to the hydrocarbon SbS‐functionalized system including a mixed second ligand shell of SDBS and DDPB. When the fluorocarbon moiety of FC‐PB was incorporated into the fluorinated first‐shell of the Al2O3‐(PAR1 10 % PAR3 90 % ) NPs, the hydrocarbon spacing unit of the amphiphile FC‐PB was positioned around the perylene core and separated the perylene units onto the NP surface which goes hand‐in‐hand with an increased fluorescence intensity of the perylene signals. The absent formation of a broad undefined excimer band signals that in this case no fluorophilic pocket was formed among the perylene core.

During the SbS‐formation of [Al2O3‐(PAR1 10 % PAR3 90 % )]FC‐PEG, the fluorescence signals changed into a broad undefined excimeric fluorescence band. These results are very similar to those of [Al2O3‐(PAR1 10 % PAR3 90 % )]PFUnDA. This is also a successful proof for the formation of fluorophilic pockets around the perylene units causing stack formation.

Biological experiments

Fluorescent dyes and proteins are used in immunofluorescent assay, live cell or in vivo animal imaging. Long observation times and cell autofluorescence in the visible spectrum leads to the search for new fluorescent markers with high quantum yields, large saturation intensities and good biocompatibility. Because of their excellent fluorescence properties, the NPs [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS, and [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB, functionalized with PAR1 and PAR2 in the first shell and amphiphiles fused of hydrocarbons and polar head‐groups with contrary surface charge, were tested for their application in fluorescence imaging in different cell lines (A549‐lung cancer cells, HUVEC‐endothelial cells, MCF‐10 A‐breast epithelial cells, and MDA‐MB‐231‐breast cancer cells). These NPs are excellent dispersible in water. Both kinds of SbS‐functionalized nanoparticles exhibit a hydrodynamic diameter of 100 nm (Table 1, Figure S12 and S15) in water. Compared to the diameter size of 14 nm obtained by BET and TEM measurements (Figure S1 and S2), the size is quite enlarged. The hydrodynamic diameter obtained by DLS measurements of the pristine NPs are below 50 nm (Figure S11). This indicating already some clustering of the pristine NPs. Therefore, it is possible that polynuclear clusters form the center of the SbS‐functionalized NPs. However, biological fluids have a different ionic strength compared to the ultrapure water and is additionally enriched with proteins and other biological substances that can form the so‐called protein corona on the surface of the NPs. Supplementary to the DLS measurements of a 0.015 wt % NP dispersion in water, the same measurements were conducted with NPs in cell culture medium enriched with 10 % fetal calf serum (FCS) as a protein source. The hydrodynamic diameter of the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS decreases in cell culture medium, whereas the diameter of the [(Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs increases (Table 1). With hydrodynamic size of around 46 nm in cell medium, the attached biomolecules provide an even better dispersibility of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs in medium than in water. The [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs with the protein shell become single particles or very small aggregates (Figure S20). After incubation of the NPs in serum enriched buffer solution, proteins or other biomolecules of the solution became attached to the surface. To determine the change in the surface potential, the NPs were collected by centrifugation and re‐dispersed in water. The zeta potential is reduced but still negative. The hydrodynamic diameter of [(Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs is around 149 nm with a broad size distribution in the DLS measurements (Figure S21). Since the diameter of the NPs are much smaller and it is very unlikely, that the attachment of the coating and protein corona expands a single particle to this size, these NPs form larger agglomerates than the [(Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs. The zeta potential turns from positive to negative charge, which indicates that the surface charge is determined by the charge of proteins and biomolecules attached. Since the zeta potential of both particles after protein attachment is similar, only the surface attached proteins of the serum define now the zeta potential of the NPs. In a next step the amount of protein at the NPs surface was measured. Both NPs were incubated in a PBS solution with 1 % FCS overnight and afterwards the amount of proteins on the surface was determined. The proteins are getting adsorbed on the surface of single particles, smaller or larger aggregates forming a small shell around these. Since both NP solutions contain the same concentration, the larger agglomerates of the [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs form fewer particles than the smaller sized aggregates of the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs and therefore a lesser amount of proteins is needed to surround these NPs as can be seen in Table 1.

Table 1.

DLS (intensity in number) and zeta‐potential measurements of the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS and [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs in water and cell culture medium with 10 % serum and amount of proteins attached onto the particle surface.

 

DLS

(water)

ζ

(water)

DLS

(medium)

ζ

(water)

Mass of

protein/1 mg NP

[Al2O3‐(PAR1 10 % PAR2 90 %)]SDBS

100.89±6.80 nm

−58.0±2.46 mV

46.13±4.00 nm

−26.4±1.89 mV

0.80 mg protein/mg NP

[Al2O3‐(PAR1 10 % PAR2 90 %)]DTAB

100.89±6.80 nm

+33.8±2.87 mV

149.27±12.93 nm

−32.8±0.651 mV

0.38 mg protein/mg NP
Open in a new tab

The size of the NP aggregates also determines the way NPs were taken up by cells and their localization inside the cells [23] (Figure 9 a–c). The large agglomerates of the NPs enter the cells via endocytosis most likely macro‐pinocytosis. Macro‐pinocytosis involves the formation of actin filament driven plasma membrane protrusion, which finally covers the large agglomerates of the NPs (see Figure 9 a). Finally a large vesicle, the micropinosome, loaded with these NPs is formed (see Figure 9 b). [24] The well dispersed [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS can be found in small NP clusters or single particles distributed in the whole cytoplasm (Figure 9 c). These particles are not found inside of vesicles like the [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB. The absence of vesicles, which are built during endocytosis, indicates a different uptake mechanism like a direct diffusion through the plasma membrane. [7] Both NPs were tested for their biocompatibility via the neutral red assay in all different cell lines. The [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs show nearly no cytotoxic effects on the non‐cancerous as well as the cancerous cell lines, when cultivated for 24 h in cell culture medium containing 50 μg mL−1 NPs (Figure 9 d). In contrast, the [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs are more toxic for the A549, MDA‐MB‐231 and MCF‐10 A cells than the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs.

Figure 9.

Figure 9

Open in a new tab

TEM images of the A549 cells with (a, b) [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB, (c) [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS and (d) biocompatibility of the NPs to the different used cell lines.

The different uptake mechanisms are also reflected in the fluorescent images of the cells. To ensure, that the attached protein shell and the retention time of the NPs in cell culture medium, does not change the absorption and fluorescence of the NPs, long term stability UV/Vis and fluorescence measurements were conducted (Figure S10). After 58 h no change in the UV/Vis absorption or fluorescence can be detected. The fluorescent [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs build large green spots, which surround the blue stained cell nucleus of the A549 cells (Figure 10 a) and HUVECs (Figure 10 c). The large green spots represent the agglomerates of NPs concentrated inside the vesicles. The [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs appear as small green dots or a blurry shell around the cell nucleus of the A549 cells (Figure 10 b). These NPs are evenly distributed throughout the whole cytoplasm. The concentration of the [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs inside of the macropinosomes lead to a higher fluorescent signal inside the cells than the evenly distributed [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs. No green fluorescent can be seen inside the cytoplasm of the HUVECs (Figure 10 d). A similar behavior can are recognized for the MDA‐MB‐231 and MCF‐10 A cells (Figure S22).

Figure 10.

Figure 10

Open in a new tab

Fluorescent images of the A549 cells with (a) [Al2O3‐(PAR1 10 % PAR2 90 %)]DTAB, (b) [Al2O3‐(PAR1 10 % PAR2 90 %)]SDBS and HUVEC cells with (c) [Al2O3‐(PAR1 10 % PAR2 90 %)]DTAB and (d) the [Al2O3‐(PAR1 10 % PAR2 90 %)]SDBS NPs.

To test if the fluorescent signal of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS is to low inside the cells or the concentration of these NPs is really much smaller inside the HUVECs, the concentration of the NPs inside the cells was detected. The fluorescent intensity was correlated with the concentration of the different NPs and normalized to the cell number. For the fluorescent images the concentration of [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs is clearly greater inside all cell lines than that of the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs (Figure 11 a), but compared to the non‐cancerous HUVECs and MCF‐10 A cells the amount of [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS is still much higher inside the cancerous A549 and MDA‐MB‐231 cells. To test if these results reflect the actual concentration of the NPs inside the cells, the amount of aluminum was determined via ICP‐AAS and also normalized to the cell number. For a better comparability the aluminum concentration of the NPs was measured and so the NP concentration per cell was stated (Figure 11 b). The determination of the mass of NPs via ICP‐AAS shows very similar results to the measurements via the fluorescent imaging. The smaller aggregates or single NPs of the [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS dispersion are hardly penetrate the cellular membranes of the non‐cancerous cells, in contrast to the cancerous cells. The structure and composition of the plasma membrane as well as its biophysical status regulates the intracellular uptake of different substances. The rapid proliferation of cancer cells leads to a frequently need of membrane lipids and proteins. Therefore, their biosynthetic pathways are modified and up‐regulated in cancer cells, which is followed by a continuous production of alterations in lipids or proteins. The nature of these changes determines the properties and uptake mechanics of the cancer cell membrane, which differs from the uptake behavior of healthy cells like HUVEC and MCF‐10 A. [25]

Figure 11.

Figure 11

Open in a new tab

Determination of the intracellular NP content either by (a) fluorescent intensity or by (b) ICP‐AAS measurements.

Conclusions

We reported on a series of SbS‐functionalized Al2O3 NPs involving a perylene terminated phosphonic acid as a fluorescence marker in the first ligand shell in combination with a hydrocarbon or a fluorocarbon terminated phosphonic acid as well as amphiphiles in the second ligand shell. Already the combination of the perylene phosphonic acid with the matrix phosphonic acids leads to different fluorescence features of the perylene core. For the second‐shell functionalization, amphiphiles consisting of a hydrocarbon or fluorocarbon moiety were used. The interdigitation of the amphiphiles into the first ligand shell leads to unique changes in the fluorescence properties of the NP hybrids. The changes in the fluorescence are associated to disaggregation or aggregation of the perylene units through formation of a lipophilic or fluorophilic pocked among the perylene units. An increase in fluorescence and creation of more refined fluorescence signals was observed for disaggregation of the perylene units, whereas a decrease of fluorescence sets in when the perylene units are forced to aggregate. Furthermore, the perylene fluorescence of first‐shell functionalized NPs was reduced by implementation of amphiphiles involving a pyridinium head‐group, due to the establishment of a photo‐induced electron‐transfer from the perylene towards the pyridinium building block. After studying a series of examples of water‐dispersible NPs, we selected two systems with excellent fluorescence features and opposite surface charges, namely [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS and [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB, for biological applications. These NPs differ in their dispersibility in the cell culture medium. The [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs form large agglomerates surrounded by a protein corona that change their surface charge. These large particles were taken up by all used cells to a large extend via macro‐pinocytosis. The [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs are well dispersible forming individualized NPs or very small aggregates surrounded by a still negatively charged protein shell. These NPs cross the membranes of the cancer cells by penetration and are afterwards evenly distributed in the cytoplasm. In fluorescent imaging the large agglomerates of the [Al2O3‐(PAR1 10 % PAR2 90 % )]DTAB NPs inside vesicles can be easier spotted than the single, widespread [Al2O3‐(PAR1 10 % PAR2 90 % )]SDBS NPs. In upcoming studies of our laboratory these excellent fluorescent probes will be developed for their use in multicellular tumor spheroids. With the help of the fluorescent signal, the pathway and penetration depth of NPs inside this several micrometers large tumor spheroids will be followed.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supplementary

Click here for additional data file. (3.4MB, pdf)

Acknowledgements

We thank the Cluster of Excellence “Engineering of Advanced Materials” (EAM) andthe SFB 953 “Synthetic Carbon Allotropes” funded by the German Research Council (DFG) as well as the Graduate School Advanced Materials and Processes (GS AMP) for financial support. We also thank the Chair of Chemical Reaction Engineering (CRT) in Erlangen for the BET measurement, Tobias Weißenberger (Institute of Chemical Reaction Engineering, University Erlangen‐Nuremberg) for the ICP‐AAS measurements, and Tobias Luchs for the Cover Design. Open access funding enabled and organized by Projekt DEAL.

L. M. S. Stiegler, S. Klein, C. Kryschi, W. Neuhuber, A. Hirsch, Chem. Eur. J. 2021, 27, 1655.

References

  • 1. 
  • 1a. Zeininger L., Petzi S., Schönamsgruber J., Portilla L., Halik M., Hirsch A., Chem. Eur. J. 2015, 21, 14030–14035; [DOI] [PubMed] [Google Scholar]
  • 1b. Stiegler L. M. S., Luchs T., Hirsch A., Chem. Eur. J. 2020, 26, 8483–8498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. 
  • 2a. Pujari S. P., Scheres L., Marcelis A. T., Zuilhof H., Angew. Chem. Int. Ed. 2014, 53, 6322–6356; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2014, 126, 6438–6474; [Google Scholar]
  • 2b. Dietrich H., Schmaltz T., Halik M., Zahn D., Phys. Chem. Chem. Phys. 2017, 19, 5137–5144. [DOI] [PubMed] [Google Scholar]
  • 3. Pellegrino T., Manna L., Kudera S., Liedl T., Koktysh D., Rogach A. L., Keller S., Raedler J., Natile G., Parak W. J., Nano. Lett. 2004, 4, 703–707. [Google Scholar]
  • 4. 
  • 4a. Zeininger L., Stiegler L. M. S., Portilla L., Halik M., Hirsch A., ChemistryOpen 2018, 7, 282–287; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4b. Portilla L., Halik M., ACS Appl. Mater. Interfaces 2014, 6, 5977–5982; [DOI] [PubMed] [Google Scholar]
  • 4c. Paniagua S. A., Hotchkiss P. J., Jones S. C., Marder S. R., Mudalige A., Marrikar F. S., Pemberton J. E., Armstrong N. R., J. Phys. Chem. C 2008, 112, 7809–7817. [Google Scholar]
  • 5. Wittmann J. E., Stiegler L. M. S., Henkel C., Träg J., Götz K., Unruh T., Zahn D., Hirsch A., Guldi D., Halik M., Adv. Mater. Interfaces 2019, 6, 1801930. [Google Scholar]
  • 6. 
  • 6a. Wade D. A., Mao C., Hollenbeck A. C., Tucker S. A., Frenzius J. Anal. Chem. 2001, 369, 378–384; [DOI] [PubMed] [Google Scholar]
  • 6b. Wade D. A., Tucker S. A., Talanta 2000, 53, 571–578; [DOI] [PubMed] [Google Scholar]
  • 6c. Pandey S., Roy L. E., Fetzer J. C., Acree W. E., Talanta 1999, 48, 1103–1110; [DOI] [PubMed] [Google Scholar]
  • 6d. Pandey S., Acree W. E., Fetzer J. C., Talanta 1998, 47, 769–778; [DOI] [PubMed] [Google Scholar]
  • 6e. Pandey S., Acree W. E., Fetzer J. C., Talanta 1997, 45, 39–45. [DOI] [PubMed] [Google Scholar]
  • 7. Breymann U., Dreeskamp H., Koch E., Zander M., Chem. Phys. Lett. 1978, 59, 68–70. [Google Scholar]
  • 8. Henkel C., Wittmann J. E., Trag J., Will J., Stiegler L. M. S., Strohriegl P., Hirsch A., Unruh T., Zahn D., Halik M., Guldi D. M., Small 2020, 16, 1903729. [DOI] [PubMed] [Google Scholar]
  • 9. Stiegler L. M. S., Hirsch A., Chem. Eur. J. 2019, 25, 11864–11875. [DOI] [PubMed] [Google Scholar]
  • 10. Klein S., Stiegler L. M. S., Harreiss C., Distel L. V. R., Neuhuber W., Spiecker E., Hirsch A., Kryschi C., ACS Appl. Biol. Mater. 2018, 1, 2002–2011. [DOI] [PubMed] [Google Scholar]
  • 11. Zhai D., Xu W., Zhang L., Chang Y. T., Chem. Soc. Rev. 2014, 43, 2402–2411. [DOI] [PubMed] [Google Scholar]
  • 12. 
  • 12a. Birks J. B., Rep. Prog. Phys. 1975, 38, 903–974; [Google Scholar]
  • 12b. Katoh R., Sinha S., Murata S., Tachiy M., J. Photoch. Photobio. A 2001, 145, 23–34. [Google Scholar]
  • 13. Etschel S. H., Portilla L., Kirschner J., Drost M., Tu F., Marbach H., Tykwinski R. R., Halik M., Angew. Chem. Int. Ed. 2015, 54, 9235–9238; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2015, 127, 9367–9370. [Google Scholar]
  • 14. 
  • 14a. Jordan D., Tan E., Hegh D., J. Surfactants Deterg. 2012, 15, 587–592; [Google Scholar]
  • 14b. Luchs T., Sarcletti M., Zeininger L., Portilla L., Fischer C., Harder S., Halik M., Hirsch A., Chem. Eur. J. 2018, 24, 13589–13595. [DOI] [PubMed] [Google Scholar]
  • 15. Zeininger L., Portilla L., Halik M., Hirsch A., Chem. Eur. J. 2016, 22, 13506–13512. [DOI] [PubMed] [Google Scholar]
  • 16. Jäger C. M., Schmaltz T., Novak M., Khassanov A., Vorobiev A., Hennemann M., Krause A., Dietrich H., Zahn D., Hirsch A., Halik M., Clark T., J. Am. Chem. Soc. 2013, 135, 4893–4900. [DOI] [PubMed] [Google Scholar]
  • 17. Hotchkiss P. J., Jones S. C., Paniagua S. A., Sharma A., Kippelen B., Armstrong N. R., Marder S. R., Acc. Chem. Res. 2012, 45, 337–346. [DOI] [PubMed] [Google Scholar]
  • 18. 
  • 18a. Wilson L. M., Griffin A. C., Macromolecules 1993, 26, 6312–6314; [Google Scholar]
  • 18b. Wilson L. M., Griffin A. C., Macromolecules 1994, 27, 1928–1931; [Google Scholar]
  • 18c. Träg J., Zahn D., J. Mol. Model. 2019, 25, 39. [DOI] [PubMed] [Google Scholar]
  • 19. Pandey S., Acree W. E., Fetzer J. C., Mikrochim. Acta 1998, 129, 41–45. [Google Scholar]
  • 20. 
  • 20a. Rodenas E., Valiente M., d. S. Villafruela M., J. Phys. Chem. B 1999, 103, 4549–4554; [Google Scholar]
  • 20b. Haftka J. J., Scherpenisse P., Oetter G., Hodges G., Eadsforth C. V., Kotthoff M., Hermens J. L., Environ. Toxicol. Chem. 2016, 35, 2173–2181; [DOI] [PubMed] [Google Scholar]
  • 20c. Yuan S., Cai Z., Xu G., Jiang Y., Colloid. Polym. Sci. 2002, 280, 630–636. [Google Scholar]
  • 21. 
  • 21a. Birks J. B., Christophorou L. G., Nature 1963, 197, 1064–1065; [Google Scholar]
  • 21b. Birks J. B., J. Phys. Chem. 1963, 67, 2199–2200; [Google Scholar]
  • 21c. Birks J. B., Christophorou L. G., Spechtrochim. Acta 1963, 19, 401–410; [Google Scholar]
  • 21d. Birks J. B., Christophorou L. G., Nature 1962, 194, 442–444. [Google Scholar]
  • 22. 
  • 22a. Mukerjee P., Yang A. Y. S., J. Phys. Chem. 1976, 80, 1388–1390; [Google Scholar]
  • 22b. Lo Nostro P., Adv. Colloid Interface Sci. 1995, 56, 245–287; [Google Scholar]
  • 22c. Shinoda K., Nomura T., J. Phys. Chem. 1980, 84, 365–369; [Google Scholar]
  • 22d. Kunitake T., Tawaki S.-i., Nakashima N., Bull. Chem. Soc. Jpn. 1983, 56, 3235–3242. [Google Scholar]
  • 23. 
  • 23a. Treuel L., Jiang X., Nienhaus G. U., J. R. Soc. Interface 2013, 10, 20120939; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23b. Ritz S., Schöttler S., Kotman N., Baier G., Musyanovych A., Kuharev J., Landfester K., Schild H., Jahn O., Tenzer S., Mailänder V., Biomacromolecules 2015, 16, 1311–1321; [DOI] [PubMed] [Google Scholar]
  • 23c. Llop J., Estrela-Lopis I., Ziolo R. F., González A., Fleddermann J., Dorn M., Vallejo V. G., Simon-Vazquez R., Donath E., Mao Z., Gao C., Moya S. E., Part. Part. Syst. Charact. 2014, 31, 24–35; [Google Scholar]
  • 23d. Ma N., Ma C., Li C., Wang T., Tang Y., Wang H., Mou X., Chen Z., He N., J. Nanosci. Nanotechnol. 2013, 13, 6485–6498. [DOI] [PubMed] [Google Scholar]
  • 24. Stern S. T., Adiseshaiah P. P., Crist R. M., Part. Fibre Toxicol. 2012, 9, 20–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. 
  • 25a. Bernardes N., Fialho A. M., Int. J. Mol. Sci. 2018, 19, 3871–3890; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25b. Galeotti T., Borello S., Minotti G., Masotti L., N. Y. Acad. Sci. 1986, 488, 468–480. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supplementary

Click here for additional data file. (3.4MB, pdf)

Articles from Chemistry (Weinheim an Der Bergstrasse, Germany) are provided here courtesy of Wiley

RESOURCES