Unveiling size‐fluorescence correlation of organic nanoparticles and its use in nanoparticle size determination

Quantitatively establishing the correlation between nanoparticle size and fluorescence is essential for understanding the behavior and functionality of fluorescent nanoparticles (FNPs). However, such exploration focusing on organic FNPs has not been achieved to date. Herein, we employ the use of supramolecular polymeric FNPs prepared from tetraphenylethylene‐based bis‐ureidopyrimidinone monomers (bis‐UPys) to relate the size to the fluorescence of organic nanoparticles. At an equal concentration of bis‐UPys, a logarithmic relationship between them is built with a correlation coefficient higher than 0.96. Theoretical calculations indicate that variations in fluorescence intensity among FNPs of different sizes are attributed to the distinct molecular packing environments at the surface and within the interior of the nanoparticles. This leads to different nonradiative decay rates of the embedded and exposed bis‐UPys and thereby changes the overall fluorescence quantum yield of nanoparticles due to their different specific surface areas. The established fluorescence intensity‐size correlation possesses fine universality and reliability, and it is successfully utilized to estimate the sizes of other nanoparticles, including those in highly diluted dispersions of FNPs. This work paves a new way for the simple and real‐time determination of nanoparticle sizes and offers an attractive paradigm to optimize nanoparticle functionalities by the size effect.

][11][12] In this aspect, unraveling the correlation between nanoparticle size and fluorescence is invaluable for interpreting behaviors and exploring the structure−function relationships of FNPs.For instance, the size and emission of quantum dots have been quantitatively correlated for multiplexed optical coding, and so on. [13]However, to the best of our knowledge, no such studies focusing on organic nanoparticles have been reported to date.The paucity of them is presumably as a result of the following factors: (i) unlike semiconductor quantum dots, the product of preparation is often prone to generate polydispersed organic nanoparticles with a broad distribution of shapes and defects.Such reproducibility issues hinder the exploration of standard size-fluorescence correlations due to the difficulty of controlling variables; (ii) the formation of organic FNPs is associated with aggregation of chromophores, which usually leads to the notorious aggregation-caused quenching (ACQ) effect and thereby causes diminished fluorescence intensity and lowered signal sensitivity.
Herein, we achieve a quantitative investigation of the correlation between nanoparticle size and fluorescence based on supramolecular polymeric FNPs.As a proof-ofconcept, three bis-ureidopyrimidinone (UPy) monomers (bis-UPys, named TPE-UPy, TPEPy-UPy, and TPEDC-UPy) containing tetraphenylethylene (TPE) derivatives were synthesized to fabricate organic nanoparticles with different sizes (Scheme 1).[20][21][22][23] On this basis, a logarithmic correlation between the fluorescence intensity and nanoparticle size is successfully established for nanoparticles of TPE-UPy, TPEPy-UPy, and TPEDC-UPy with a correlation coefficient higher than 0.96.Theoretical calculations demonstrate that the bis-UPys at the surface and within the interior of the nanoparticles have different molecular pack-ing environments, which greatly influence the nonradiative decay rates of the TPE derivatives.The embedded bis-UPy molecules exhibit higher fluorescence quantum yield than the exposed ones in the nanoparticles.As nanoparticle size increases, the specific surface area diminishes, leading to a gradually elevated proportion of the embedded TPE derivatives, which in turn amplifies the AIE signals.The correlation between nanoparticle size and fluorescence intensity is independent of the emulsifier types and the intramolecular charge-transfer effects.Particularly, the reliability of this correlation was verified by estimating the sizes of dispersed FNPs through measuring their fluorescence intensity and fitting results to the established nonlinear correlations.
Their structures were fully characterized by 1 H NMR, 13 C NMR, and mass spectroscopies (Figures S1-S9).The absorption spectra of bis-UPys display two main absorption peaks at around 280 and 350 nm due to the π-π* transition of TPE derivatives (Figure S10).To investigate the AIE characteristics, we studied the emission behaviors of TPE-UPy, TPEPy-UPy, and TPEDC-UPy in tetrahydrofuran/water mixtures (Figure S11).When the poor solvent, that is, water was added, their fluorescence in tetrahydrofuran solutions was gradually intensified, suggesting the typical AIE features.According to previous literature reports, [17,24] supramolecular polymers can be formed in chloroform from TPE-UPy, TPEPy-UPy, and TPEDC-UPy, respectively.The supramolecular polymeric FNPs of bis-UPys were obtained by the miniemulsion method (see Experimental Section for details).This preparation process consists of two key steps: (i) the self-complementary hydrogen bonding of UPy units drives the formation of supramolecular polymers; and (ii) the organic emulsified droplets provide a hydrophobic local environment for the aggregation of supramolecular polymers to form stable nanoparticles with desired sizes.[27][28] In this context, nine groups of FNPs (FNP1-FNP9) for each type of bis-UPy monomer were prepared by emulsifying their chloroform solution (200 μL, 3.75-60.00mg/mL) in the aqueous solution of the cetyltrimethyl ammonium bromide (CTAB) surfactant (15 mL, 1.1 mM).The formation of nanoparticles was tentatively verified by the uniformly distributed fluorescence and Tyndall effect of the obtained water dispersions (Figure 1A and Figure S12).
To precisely establish the correlation between nanoparticle size and fluorescence signal, we investigated the shape and size of the nanoparticles by transmission electron microscopy (TEM) and dynamic light scattering (DLS).As shown in Figure 1B and Figure S13, TEM images reveal that the obtained nanoparticles appear as regular nanospheres.Meanwhile, the sizes of the nanoparticles observed by TEM show gradually larger diameters from FNP5 to FNP9.Nevertheless, the utilization of TEM for examining FNP1 to FNP4 proved to be challenging due to the difficulty in locating nanoparticle samples within the selected images.Consequently, the dimensions of the FNPs were determined utilizing DLS.As anticipated, the hydrodynamic diameters demonstrate a progressive increase from FNP1 to FNP9 for each type of bis-UPy-based sample (Figures 1C-E and Figures S14-S16).
Shuai and Zheng et al. calculated the fluorescence quantum yield (Φ f ) differences between the exposed and embedded fluorophores in amorphous aggregates, providing a theoretical basis for monitoring the nanoparticle size by fluorescence detection. [29]We further studied the emission behaviors of the three sets of FNPs prepared from TPE-UPy, TPEPy-UPy, and TPEDC-UPy, respectively (see Experimental Section for testing details).Excitingly, under the identical concentration of bis-UPy in dispersions (8 μM), the fluorescence intensity from FNP1 to FNP9 for each set of nanoparticles increases progressively, demonstrating the underlying relationships between nanoparticle size and fluorescence signals (Figures 2A-C).
Based on the hydrodynamic diameter and emission behavior, the correlation between the size and fluorescence intensity of FNPs is established by using a logarithmic fitting method (Table S1).As shown in Figures 2D-F, the logarithmic relationship for FNPs of TPE-UPy, TPEPy-UPy, and TPEDC-UPy has fitting correlation coefficients as high as 0.965, 0.989, and 0.993, respectively.This result indicates that the correlation between nanoparticle size and fluorescence intensity is suitable for different types of AIE chromophores.Accordingly, the correlation between specific surface area and fluorescence intensity of those organic FNPs is also built up, that is, the gradually reduced specific surface area leads to progressively enhanced fluorescence intensity through decreasing the proportion of the exposed bis-UPy molecules in the dispersed nanoparticles.This straightforward size-fluorescence correlation enables direct access to determine the nanoparticle size in general labs by measuring the fluorescence signals.
To gain a deeper insight into the origin of the sizefluorescence intensity correlation of nanoparticles, theoretical calculations combining molecular dynamics (MD) simulations and quantum mechanics/molecular mechanics (QM/MM) calculations were carried out.The electronic distributions of the major transition orbitals (HOMO and LUMO) indicate that the emission behavior of TPE-UPy is mainly determined by the TPE moiety (Figure S17).Considering that irregular molecular packing and diverse molecular conformations may cause distinct surrounding environments for molecules at the surface and within the interior of nanoparticles, the QM/MM models based on the conformations of TPE-UPy in dimer and aggregate of 60 TPE-UPy molecules obtained by MD simulations were set up for investigating photophysical processes of the exposed and embedded molecules of amorphous aggregates (Figure 3A).
The key structural parameters (bond lengths, bond angles, and dihedral angles) of the optimized ground (S 0 ) and excited (S 1 ) states, as well as the corresponding structural changes |Δ(S 0 −S 1 )| of TPE-UPy in all cases are examined (Figure 3B, and Figures S18, S19 and Table S2).It is clear that the values of |Δ(S 0 −S 1 )| in the dimerized and exposed QM/MM models are close, confirming that the local environments of TPE-UPy exposed at the surface of amorphous aggregate are similar to the dimerized molecules in solution.However, |Δ(S 0 −S 1 )| of the embedded QM/MM models is obviously smaller than those of the exposed ones.This implies that the molecular motions (i.e., twisted vibration of C=C and rotational motions of phenyl rings) of the embedded TPE-UPy are more tightly restricted than those of the exposed molecules in the amorphous aggregate, which is beneficial to activate the AIE effect and reach the higher fluorescence quantum yield.On the other hand, the fluorescence quantum yield can be estimated by Φ f = k r /(k r + k nr ), where k r represents the radiative decay rate and k nr denotes the nonradiative decay rate.Accordingly, k r and reorganization energy (λ) were calculated to further investigate the Φ f variations of TPE-UPy in distinct surrounding environments.
As shown in Figure 3C, D and Tables S3-S4, the k r values of TPE-UPy in dimer and aggregate are at the same order of magnitude.As known from previous work, [30,31] the decrease in reorganization energy can sharply retard the electronvibration coupling that causes the nonradiative decay.The λ of the embedded TPE-UPy is considerably smaller than those of the molecules at the aggregate surface and in the dimer.This suggests a much smaller k nr , thereby implying the much higher Φ f value of the embedded TPE-UPy than those of the exposed or dimerized ones.Taken together, the different molecular packing environments at the surface and within the interior of nanoparticles lead to distinct fluorescence quantum yields of bis-UPy molecules.Consequently, with the nanoparticle size increasing, the specific surface area decreases, which results in an elevated proportion of the embedded bis-UPys in dispersed FNPs that in turn enhances the fluorescence intensity.The validity of the theoretical demonstration is further supported by the observed logarithmic relationship between fluorescence quantum yield and nanoparticle size in FNPs of bis-UPys (Table S5 and Figure S20).
The universality of this quantitative correlation was further investigated by selecting sodium dodecyl sulfate (SDS, an anionic surfactant) as an emulsifier to construct the nanoparticles of bis-UPys.The results are similar to those based on the FNPs prepared with the aid of CTAB, suggesting that the size-fluorescence correlations are independent of the surface charges (Figures S21-S26, and Table S6).Moreover, because TPEPy-UPy and TPEDC-UPy have typical donor-acceptor structures, their nanoparticles of different sizes can exhibit emission shifts to some extent due to the twisted intramolecular charge transfer effect.However, this phenomenon does not affect the nonlinear correlations between nanoparticle size and fluorescence intensity at a specific wavelength (Figures 2D-F and Figures S26g-i).
To corroborate the reliability of the established correlation between nanoparticle size and fluorescence intensity, we  3E-G, the fluorescence intensity of FNPx at 520 nm was 164, corresponding to a nanoparticle size of 41.7 nm.This value is very close to the hydrodynamic diameter of 40.4 nm determined by DLS.These quantitative correlations demonstrate the viability of evaluating nanoparticle size via fluorescence signal changes.Given that when a nanostructure is too small and/or its dispersion is too dilute to be resolved in a light microscope, the emission light may remain detectable, we fabricated small FNPy and FNPz of TPE-UPy, TPEPy-UPy, and TPEDC-UPy, respectively, by using their chloroform solutions of 2.50 and 1.25 mg/mL.As shown in Figure S27 and Table S7, the sizes of FNPy and FNPz can be effectively estimated in real-time through the adaptation of fluorescence intensity to the corresponding fluorescence-size correlation curves.

CONCLUSION
In this study, we successfully explore the quantitative correlation between the size and fluorescence intensity of organic nanoparticles using supramolecular polymeric FNPs and AIE.The dynamic nature of supramolecular polymers guarantees reproducibility of the nanoparticles with a well-defined shape, while the AIE feature ensures bright fluorescence signals.The logarithmic fluorescence-size correlation for organic FNPs originates from the distinct molecular packing environments of bis-UPys at the surface and within the interior of nanoparticles, which cause different nonradiative decay rates of the TPE derivatives.
The embedded bis-UPys in nanoparticles have a tighter molecular packing than that of the exposed molecules, thus exhibiting a higher fluorescence quantum yield due to the reduced nonradiative decay rates.For this reason, FNPs with larger dimensions in dispersion can display higher fluorescence intensity as a consequence of reducing specific surface areas that increase the ratio of embedded molecules to exposed ones.Utilizing the size-fluorescence correlation, we successfully estimated the sizes of other nanoparticles, including those in highly diluted dispersions of FNPs, through the measurement of their fluorescence intensity.This work offers an attractive paradigm to optimize organic nanoparticle functionalities by the size effect and paves a new way for the cost-effective determination of nanoparticle sizes.

S C H E M E 1
Chemical structures of TPE-UPy, TPEPy-UPy, and TPEDC-UPy, and schematic illustration showing the fluorescence intensity variations depending on the size of supramolecular polymeric FNPs.

F
I G U R E 1 (A) Fluorescent photographs of the nanoparticle dispersions of bis-UPys upon light irradiation at 365 nm.The FNPs from left to right correspond to FNP1 through FNP9.(B) TEM images of FNP6, FNP7, and FNP8 of bis-UPys.(C-E) Hydrodynamic diameter distributions of the FNPs for (C) TPE-UPy, (D) TPEPy-UPy, and (E) TPEDC-UPy.

F
I G U R E 3 (A) The QM/MM models of the dimer and aggregate including exposed and embedded TPE-UPy.(B) The changes of key dihedral angles (∆|S 0 −S 1 |) between the optimized ground (S 0 ) and excited (S 1 ) states for TPE-UPy in all cases.(C) The radiative decay rate (k r ) and (D) reorganization energy (λ) of TPE-UPy in different surrounding environments.(E) The fluorescence spectra and (F) hydrodynamic diameter distribution of FNPx of TPEPy-UPy (8 μM, λ ex = 335 nm).(G) The estimated nanoparticle size of FNPx based on the size-fluorescence intensity correlation.prepared an additional dispersion of FNPx using a 15.00 mg/mL chloroform solution of TPEPy-UPy.The fluorescence signal was measured (c TPEPy-UPy = 8 μM) for fitting the plot of relative fluorescence intensity versus the size of the TPEPy-UPy-based nanoparticles.As shown in Figures