Optical Properties and Kinetics: New Insights to the Porphyrin

Article pubs.acs.org/JPCB

Cite This: J. Phys. Chem. B 2018, 122, 8429−8438

Optical Properties and Kinetics: New Insights to the Porphyrin Assembly and Disassembly by Polarized Resonance Synchronous Spectroscopy Buddhini C. N. Vithanage,†,§ Joanna Xiuzhu Xu,†,§ and Dongmao Zhang*,†,‡ †

Department of Chemistry, Mississippi State University, Mississippi State, Mississippi 39762, United States Department of Chemistry, Xihua University, Chengdu 610039, China

‡

Downloaded via KAOHSIUNG MEDICAL UNIV on September 11, 2018 at 01:56:53 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: With their unique photochemical properties, porphyrins have remained for decades the most interested chemicals as photonic materials for applications ranging from chemistry, biology, medicine, to photovoltaic. Porphyrins can self-assemble into higher order structures. However, information has been scant on the kinetics and structural evolution during porphyrin assembly and disassembly. Furthermore, quantitative understanding of the porphyrin optical activities is complicated by the complex interplay of photon absorption, scattering, and fluorescence emission that can concurrently occur in porphyrin samples. Using meso-tetrakis(4sulfonatophenyl)porphyrin as the model molecule, reported herein is a combined UV−vis extinction, polarized Stokes-shifted fluorescence, and polarized resonance synchronous spectroscopic (PRS2) study of porphyrin assembly and disassembly in acidic solutions. Although porphyrin assembly and disassembly occur instantaneously upon the sample preparation, both processes last at least a few months before reaching their approximate equilibrium states. The two processes were monitored in situ by quantifying the porphyrin fluorescence and scattering depolarizations as well as its extinction, absorption, scattering, and fluorescence emission cross sections. In addition to a series of new insights to the porphyrin assembly and disassembly, the methodology described in this work opens the door for the in situ study of the structural and optical properties of photonic materials comprising molecular assembly.

■

copy (RSS) methods.17 Although UV−vis measurements provide information regarding its photon extinction activity, decomposing a UV−vis extinction spectrum into its absorption and scattering component spectrum is difficult. Even though it has become common to refer to porphyrin RSS spectrum as its resonance light scattering (RLS),17,24−27 this practice has unfortunately neglected the possibility of fluorescence contribution to the porphyrin RSS spectrum.14,24,25 A series of recent studies showed that fluorescence signal can dominate the RSS spectrum obtained with fluorescence samples.28−30 We have recently developed a polarized resonance synchronous spectroscopy (PRS2) method that has enabled for the first time the experimentally disentangling complicated interplay of photon absorption, scattering, and fluorescence emission that can concurrently occur in many realistic samples. The PRS2 data acquisition is very similar to that for the conventional RSS method and they both are obtained with common spectrofluorometers. The only difference is that the excitation and detection light are both linearly polarized in

INTRODUCTION Molecular self-assembling into higher order structures through noncovalent interactions is an important physical process responsible for the formation of many natural and synthetic functional materials.1−5 Many fluorescent molecules can spontaneously assemble into supramolecules at high concentrations and/or with external stimuli including DNA templating,6,7 solution pH,8,9 ionic strength,8,10−12 or solvent compositions.13 Optical activities of the assembled fluorophores can be drastically different from their isolated counterparts.14−18 Therefore, exploring the fluorophore assembly and disassembly and their impact on the fluorophore optical properties have been an active research topic in photonic material design and applications. Understanding the optical properties of the aggregated fluorophores is challenging. Aggregated fluorophores are usually simultaneous photon absorbers, scatters, and fluorescence emitters. In contrast, isolated fluorophores, because of their small sizes, can be approximated as simultaneous photon absorbers and emitters with no significant light scattering. The most popular spectroscopic methods for studying fluorophore aggregation include UV−vis spectroscopy,19−21 fluorescence spectroscopy,20,22,23 and the resonance synchronous spectros© 2018 American Chemical Society

Received: June 22, 2018 Revised: August 10, 2018 Published: August 13, 2018 8429

DOI: 10.1021/acs.jpcb.8b05965 J. Phys. Chem. B 2018, 122, 8429−8438

Article

The Journal of Physical Chemistry B

Figure 1. Comparison of TSPP spectra obtained with (first column) as-prepared aggregated TSPP, (second column) the filtrate of as-prepared TSPP, and (third column) the mathematical difference spectrum between the as-prepared and filtrated TSPP. Plots (A−C) are the UV−vis extinction spectra. Plots in row 2 and 3 are polarized SSF spectra of TSPP acquired with excitation wavelength of (D−F) 430 nm and (G−I) 490 nm, respectively. Plots (J−L) in row 4 are PRS2 spectra of TSPP. The dashed lines in plots (J) and (L) indicate instrument signal saturation limit of 2 × 106 cps. The concentration of the as-prepared TSPP is 5 μM, and the pH is 0.6. The as-acquired experimental SSF and PRS2 spectrum before the IFE- and solvent-background correction are shown in Supporting Information (Figure S1).

structure as revealed by cryoelectron microscopy, small-angle X-ray scattering, transmission electron microscopy, atomic force microscopy, and scanning electron microscopy measurements. 4,37−42 This enabled us to probe the possible correlations between the light scattering and fluorescence depolarization of the aggregated fluorophores and their geometric features. This combined UV−vis, SSF, and PRS2 study provides a series of new insights to the assembly and disassembly of porphyrin aggregates and their optical properties. The most critical learning is that the porphyrin assembly and disassembly are extraordinary lengthy processes, lasting for at least 4 months after the sample preparation. This is in spite of the onset of porphyrin assembly and disassembly is instantaneous upon the sample preparation. Another notable observation is that PRS2 is drastically more sensitive than UV−vis in detecting the TSPP aggregations.

PRS2 but are plane-polarized in RSS. By combining material UV−vis extinction and PRS2 measurements, one can quantify sample fluorescence and light-scattering depolarizations and cross sections, and consequently their absorption cross sections and absolute fluorescence quantum yield.28,29,31,32 Detailed theoretical background, data analysis procedures, and example applications ranging from small molecules to large plasmonic gold nanoparticles (AuNPs) are available from recent publications.28,29 Reported herein is the combined UV−vis, Stokes-shifted fluorescence (SSF), and PRS2 study of the fluorophore assembly and disassembly in water. meso-Tetrakis(4sulfonatophenyl)porphyrin (TSPP) was chosen as the model fluorophore because of its broad interests in both fundamental research as well as practical applications. Porphyrins have been used in photosynthesis,1,33 photovoltaic,3 and photodynamic therapy,34 as well as photonic device fabrication.35 Porphyrin optical properties and their physiological functionalities depend critically on the porphyrin aggregation states.14,16,36 Furthermore, the aggregated TSPP has a nanosized tubular 8430


Article


■

EXPERIMENTAL SECTION Materials and Instruments. The polystyrene NPs (PSNPs) used as the external standard in the PRS2 spectral measurements were obtained with Polysciences. All other chemicals were purchased from Sigma-Aldrich and used as received. Nanopure water (18.2 MΩ cm) was used in solution preparation. Polyethersulfone membrane filters with a pore size of 0.1 μm were obtained from Pall Corporation, UK. The UV− vis extinction spectra were acquired using a Thermo Scientific Evolution 300 UV−vis spectrophotometer with a slit width of 2 nm. The SSF and the PRS2 spectra were obtained with a HORIBA FluoroMax-4-spectrofluorometer that equipped with a computerized excitation and detection polarizers. Polarized Fluorescence, PRS2, and BVPRS2 Measurements. All SSF and the PRS2 spectra were measured with 1 cm × 1 cm cuvettes and a spectral integration time of 0.3 s. The effective excitation and detection path lengths needed in correcting the sample inner filter effect (IFE) in the fluorescence and PRS2 measurement is 0.49 and 0.52 cm, respectively, which were quantified on the basis of the sample IFE on the water Raman signal.43 In the ratiometric bandwidth-varied PRS2 (BVPRS2) spectral acquisitions, the excitation and detection monochromator bandwidths (slitwidth) were kept identical in specific PRS2 measurements, and the value of the bandwidth varies from 1.0 to 2.0 nm with an increment of 0.2 nm. The G-factor spectrum needed for correcting the instrument polarization bias is obtained with a G-factor sample set (Raminescent LLC). The polarized SSF and PRS2 spectra will be referred to as ISSF,VV and ISSF,VH, and IPRS2,VV and IPRS2,VH, respectively. VV and VH refer to the combination of the excitation and detection polarization used during the spectral acquisition in the polarized SSF and PRS2 measurements. “V” refers to the linear polarization perpendicular to the plane defined by the excitation source, the sample cuvette, and detection. “H” refers to the linear polarization horizontal to this plane. Detailed data analysis procedure including the sample inner-filter-effect correction, solvent- and cuvette-background-removal, light scattering and fluorescence depolarization determination, and cross-section quantification were presented in recent publications.28,29 The fluorescence and PRS2 spectra are acquired by normalization of the signal intensity of the sample detector by that of the reference detector. This is important for eliminating spectral variation because of the time- and/or wavelength-dependent excitation light intensity fluctuation from measurements to measurements. Filtration Separation of Isolated and Assembled Porphyrin. TSPP (5 μM) solution was prepared with a final solution of pH 0.6. The samples were aged for 3 days before the membrane filtration separation of the isolated and aggregated porphyrin. Concentration-Dependent Porphyrin Assembly. Porphyrin aggregation was conducted at strongly acidic conditions. TSPP (10 μM) stock solution was prepared using 0.01 M NaOH solution as a solvent. A series of pH 0.6 solutions were then prepared by diluting the stock solution with 2 M HCl. The final TSPP nominal concentrations in the sample solutions are 0.05, 0.1, 0.15, 0.2, 0.3, 0.4, 0.5, 0.7, 0.8, 0.9, 1, 2, and 3 μM, respectively. The samples were incubated at room temperature in dark for 3 days before measurements. Time-Dependence of TSPP Assembly and Disassembly. Both TSPP assembly and disassembly were studied as a

function of sample incubation time. The TSPP assembly was performed by diluting the 10 μM TSPP solution prepared in 0.01 M NaOH into 1 μM with 2 M HCl. The TSPP disassembly was conducted by diluting a 10 μM TSPP in 2 M HCl solution into 1 μM with 2 M HCl. pHs of both samples were approximately 0.6.

■

RESULTS AND DISCUSSION UV−vis, Fluorescence, and PRS2 Spectra of Isolated and Aggregated TSPPs. Experimental identification of the spectral features for the isolated and aggregated TSPP was performed with membrane filtration method described in the Experimental Section. The 423, 490, and 710 nm peak in both the UV−vis (Figure 1A) and PRS2 (Figure 1J) spectra in the as-prepared TSPP solutions are all absent in the spectra obtained with the TSPP filtrate (Figure 1B,K), confirming that these spectral features are distinctive markers for the aggregated TSPP (Figure 1C,L). The unique spectral feature for the isolated TSPP monomers includes the 435 and 655 nm peak in the UV−vis spectrum (Figure 1B) and the 655 nm peak (Figure 1K) in the PRS2 spectrum. The PRS2 peak position of 490 nm was deduced from the samples with lower TSPP concentrations shown later in this work. Pinpointing the position for this PRS2 peak with the 5 μM sample was not possible because its peak intensity exceeded the instrument signal saturation (Figure 1J,L). Exciting the as-prepared and filtrated TSPP sample with 430 nm light produced a broad SSF emission with peak wavelength at 665 nm (Figure 1D,E), but exciting the as-prepared TSPP with 490 nm light generated SSF emission with the peak wavelength at 712 nm with no identifiable spectral feature at the 665 nm region (Figure 1G). Furthermore, exciting the TSPP filtrate with 490 nm produced no detectable fluorescence features (Figure 1H). This is consistent with reports that the 665 nm peak and 712 nm SSF features are marker peaks for isolated and aggregated TSPPs, respectively. It also confirms that the 435 and 655 nm UV−vis peaks are specific for the monomeric TSPP, whereas UV−vis peaks of 423, 490, and 710 nm are for aggregated TSPPs.44 The reason we used the relatively concentrated TSPP sample (5 μM) in this filtration study is to ensure that the aggregated TSPP has a relatively large size; therefore, it can be separated with the filtration membrane. It is important to note that the TSPP filtrate most likely has no significant aggregated TSPPs, which is evident from the fact that the spectra obtained with the filtrate are totally free of any marker peaks associated with aggregated TSPPs, but the sample retained in the filtrate membrane likely contains isolated TSPPs. The amount of the TSPP in the filtrate monotonically decreased in three consecutive filtrations (data not shown). Therefore, the difference spectra can contain both spectral features from aggregated and monomeric TSPP. For the sake of convenience, however, we refer to the difference UV−vis, SSF, and PRS2 spectra as the TSPP aggregate spectra. Conversely, the spectra obtained with the filtrate are referred to as the isolated TSPP UV−vis and SSF spectra. Material UV−vis spectral features can be because of the combination of its photon absorption and scattering, whereas the sample PRS2 features can be because of its photon scattering and fluorescence emission under the resonance excitation and detection conditions. The 423 and 490 nm peaks observed in the UV−vis spectrum of the TSPP aggregates are due to the combination of their photon 8431


Article

The Journal of Physical Chemistry B Table 1. Quantification of Optical Activities of the Isolated and Assembled TSPPsg peak (nm)

[TSPP]a (μM)

σExt (×10−16 cm2)b

σF (×10−19 cm2)c

% SARd

423 435 490 655 710

10.0 0.2 1.0 0.7 10.0

3.1 18.7 2.0 1.9 2.9

NA NA NA 4.6 NA

0.3 0.0 19.5 0.0 1.4

e

PPRS2

f

0.15 NA 0.15 0.98 0.15

NA NA NA 0.97 0.50

PSSF

a

Nominal TPPS concentration of the samples used for specified peak analysis. bExtinction cross section. cFluorescence cross section calculated using the specified nominal TPPS concentration. dPercentage scattering-to-absorption ratio (SAR). eCalculated PRS2 depolarization. fCalculated SSF depolarization. gNA represents not available.

655 nm is 0.24% (σF/σabs ≈ σF/σabs because the light scattering is negligible for isolated TSPP at this wavelength). If we assume that the fluorescence quantum yield of the TSPP aggregate at 710 nm is the same as that for the isolated TSPP at its resonance wavelength of 650 nm, this fluorescence quantum yield of the TSPP aggregate should only be 4.7 times smaller than its RLS quantum yield, defined as the SAR value in Table 1. The fact that there is no detectable fluorescence signal in the PRS2 spectrum of the TSPP aggregates at 710 nm strongly suggests that the fluorescence quantum yield for the TSPP aggregate at this wavelength is even smaller than that for the isolated TSPP at 650 nm. In other words, TSPP aggregation not only induces red-shift in the TSPP emission wavelength but also reduces its fluorescence activity. The light-scattering and fluorescence depolarization data support the UV−vis and PRS2 peak assignment made in this work. Mathematically, the light scattering and fluorescence depolarization were defined as the G-factor-corrected ratio of IVH versus IVV.29 It varies from 0, indicating that the scattered or emitted photons have the same polarization of the excitation photons, to 1, in which the polarization of the scattered or emitted photon were totally randomized. The SSF depolarization of the isolated TSPP is close to unity (Figure 2A,C), which is consistent with the fluorescence depolarization observed for other dispersed molecular fluorophores.28,29 However, the fluorescence depolarization of the TSPP aggregates is close to 0.5 (Figure 2B,C), which is exceptionally small in comparison with the values reported for other molecular assembly such as aggregated fluorescein isothiocya-

absorption and scattering at these wavelengths, whereas the PRS2 peaks at these wavelengths are due predominantly to the TSPP light scattering, but not to fluorescence emission. This is because fluorescence under the resonance excitation and detection conditions occurs only in the wavelength region where the sample both absorbs and emits.30 No significant fluorescence emission was observed at 423 or 490 nm when the as-prepared TSPP were excited with wavelengths from 280 to 400 nm and from 280 to 470 nm for the respective possible 423 and 490 nm emissions. Pinpointing the optical processes responsible for the 655 and 710 nm peaks in the TSPP UV−vis and PRS2 spectra obtained with the isolated and assembled TSPP, respectively, is much more challenging. This is because these peaks simultaneously appeared in the TSPP UV−vis, SSF, and PRS2 spectra. Consequently, the TSPP can be simultaneously photon absorbers, scatterers, and emitters at these wavelengths. One must quantitatively decouple the photon absorption, scattering, and fluorescence emissions to identify the main optical processes contributing to these spectral signals. Using the recent ratiometric BVPRS2 method, we evaluated the TSPP UV−vis photon scattering and fluorescence contribution to its PRS2 signal (Figure S2, Supporting Information). The ratiometric BVPRS2 signal from the lightscattering component is independent of the monochromator wavelength bandwidth, whereas that of fluorescence emission is linearly dependent of the wavelength bandwidth.28 The ratiometric BVPRS2 signals of TSPP at 423, 490, and 710 nm are totally independent of the wavelength bandwidth of the excitation and detection monochromators, indicating that light scattering is the predominant contributor (Figure S2). In contrast, the ratiometric BVPRS2 intensity centered at 655 nm peak is linearly dependent on the wavelength bandwidth, indicating that the signal at this wavelength region is attributed mainly to porphyrin fluorescence emission. Using the ratiometric BVPRS2 method,28 the light-scattering cross sections of 423, 490, and 710 nm peak for the aggregated porphyrin were computed, and so was the fluorescence cross section for the 655 nm peak of the isolated porphyrin (Table 1). Because the sample total photon extinction can be readily obtained with the conventional UV−vis measurement, one can readily decompose the TSPP extinction spectrum into its absorption and scattering component spectrum. Apparently, the aggregated porphyrin is a simultaneous photon absorber and scatterer in the UV−vis peaks centered at wavelengths of 423, 490, and 710 nm, respectively. The UV−vis peak at 435 and 650 nm is due predominantly to the photon absorption by the dispersed porphyrin with no significant scattering (scattering-to-absorption ratio, SAR % = 0). The absolute fluorescence quantum yield of the isolated TSPP at the resonance excitation and detection conditions at

Figure 2. SSF spectra of (A) isolated TSPPs (emission at 675 nm when excited at 430 nm) and (B) assembled TSPPs (emission at 710 nm when excited at 490 nm). (C) Comparison of the depolarization spectra of SSF for the isolated and assembled TSPPs. (D) Transmission electron microscope image of assembled TSPPs. 8432


Article


Figure 3. (A) UV−vis extinction spectra of TSPP solutions at different concentrations varied from 0.05 to 3 μM with a pH of 0.6, (B) UV−vis extinction spectral intensity change at 435 and 490 nm as a function of TSPP concentration. The red lines are linear fitting of the starting four data points and last three data points, respectively. (C) UV−vis extinction spectral intensity changes at 655 and 710 nm as a function of TSPP concentration. (D) Polarized SSF spectra excited with the wavelength of 430 nm for the as-prepared TSPP with an excitation and detection polarization combination of VV. (E) Polarized SSF spectra of the as-prepared TSPP at the excitation wavelength of 490 nm and with excitation and detection polarization combination of VV. (F) Polarized SSF intensity at 655 nm as a function of TSPP concentrations at 655 and 710 nm. (G) PRS2 spectra of TSPP at different concentrations with excitation and detection polarization combination of VV. (H,I) PRS2 VV spectral intensity as a function of TSPP concentration at wavelengths of 423 and 490, and 655 and 710 nm, respectively. Arrows indicate the increasing TSPP concentration.

nate (FITC).28 The fluorescence depolarization of the aggregated FITC is statistically identical to that of isolated FITC (0.96 ± 0.045). Indeed, the fluorescence depolarization of the aggregated TSPP is, to our knowledge, the smallest among all of the fluorophores we have investigated so far. The extraordinarily small fluorescence depolarization of the assembled porphyrin is due to the combined effects of its short fluorescence lifetime and low diffusion coefficient. Akins et al. showed that the fluorescence lifetime of the assembled TSPP is in the range from 0.082 to 0.30 ns, more than 10 times shorter than that of the dispersed monomer which is in the range from 3.5 to 9.5 ns.45 The small diffusion coefficient of the TSPP aggregates is due to the large sizes and tubular geometries. The wall of the tubular aggregate comprises tens of thousands monomers,14 and there are millions of solvent molecules inside the tube. Therefore, the average volume of individual TSPP aggregates is most likely tens of thousands times larger than the isolated TSPPs. Because the diffusion coefficient is inversely proportional to the radius of the solute, the diffusion coefficient of the aggregated TSPP can be hundreds to thousands times smaller than that of the isolated TSPP. The light-scattering depolarizations at 423, 490, and 710 nm wavelengths are all approximately 0.15, which are drastically higher than the spherical PSNPs and the aggregated FITC whose light-scattering depolarizations are essentially zero.29,31 The nonzero light-scattering depolarization is consistent with the fact the TSPP is a tubular geometry (Figure 2D). Earlier research conducted with small molecular and the plasmonic AuNPs revealed that light-scattering depolarization was very sensitive to the scatterer’s geometry.31,32 The scattering depolarization of spherical molecules such as CCl4 and gold nanosphere is below 0.03 (the limit of quantification), whereas the light-scattering depolarization of the rod-shaped molecules such as CS2 can be as high as 0.5.32

The fact that the depolarization of the PRS2 peak at 655 nm under the resonance excitation and detection conditions is the same as that in the SSF measurement, indicating that this PRS2 peak is due predominantly to the fluorescence emission of the isolated porphyrin. In contrast, the PRS2 depolarization of the 710 nm peak is similar to that for the 423 and 490 nm peaks, further validating that light scattering but not fluorescence emission is the main contributor to the PRS2 signal at 710 nm. Otherwise, PRS2 depolarization at this wavelength should be very similar to SSF depolarization at 710 nm. Earlier reports showed that fluorescence depolarization from the same fluorophore are approximately independent of the excitation and detection wavelength.29 Concentration-Dependent TSPP Assembly. The results shown in Table 1 are strictly sample-specific and timedependent. This is because the TSPP optical constants are strongly concentration-dependent, as shown in Figure 3, and time-dependent, as shown in the following section. Figure 3 shows the TSPP UV−vis (Figure 3A−C), polarized SSF (Figure 3D−F), and PRS2 (3G−I) spectra where TSPP concentration varies from 0.05 to 3 μM. Only the polarized SSF and PRS2 spectra acquired with excitation and detection polarization combination of “VV” was shown in Figure 3; the ones obtained with polarization combination of “VH” are shown in Supporting Information (Figure S3). PRS2 is drastically more sensitive than UV−vis and fluorescence in detecting the TSPP aggregate formation. The threshold concentration for the TSPP to have a detectable PRS2 peak of 490 nm is 0.05 nM (Figure S3, Supporting Information), which is more than 20-folds smaller than the 1 μM threshold concentration of the appearance of the 490 nm peak in the UV−vis spectra of the TSPP samples (Figure 3A). The intensity of 435 nm UV−vis extinction peak of the isolated TSPP is approximately linearly dependent on the 8433


Article


Figure 4. Time-dependent study of TSPP (top) assembly and (bottom) disassembly conducted with the first 3 h of sample incubation. The composition of the TSPP assembly and disassembly differs only in their method of preparation as specified in the method section, and their nominal TSPP concentrations are both 1.0 μM. (A,G) UV−vis extinction spectra of assembling and disassembling TSPP, PRS2 spectra of (B) assembling and (H) disassembling TSPP with excitation and detection polarization combination of VV. PRS2 spectra of (C) assembling and (I) disassembling TSPP with excitation and detection polarization in combination with VH. (D) Extinction cross sections and intensity at 435 nm of assembling TSPP as a function of time. (J) Extinction cross sections of disassembling TSPP at (red) 435 and (black) 490 nm as a function of time. (E,K) Scattering cross-sectional intensity at 490 nm as a function of the sample incubation time of TSPP assembly and disassembly, respectively. (F,L) Light-scattering depolarization at 490 nm of the assembling and disassembling TSPP as the function of time. Arrows indicate increasing sample incubation time.

that of the nominal TSPP concentrations used in the calculations. The light-scattering activities of the aggregated TSPP in the 2 and 3 μM TSPP samples are even higher. Unfortunately, their light-scattering cross sections cannot be determined because its scattering signal exceeds the saturation intensity of the FluoroMax-4 instrument used in this study. The fact that UV−vis extinction of the aggregated porphyrin contains significantly a fraction of light-scattering extinction that raises concerns on a rather common literature practice that directly assigns UV−vis extinction spectrum obtained with aggregated fluorophores or fluorophore/NP assemblies to their UV−vis absorbance or absorption.20,21,23,46 Kinetics of the TSPP Assembly and Disassembly. The most critical new learning is that the TSPP assembly and disassembly can be extraordinarily lengthy processes, which take several months to reach an approximate equilibrium in terms of concentration of the isolated and aggregated TSPPs. Even after extremely long incubation time, the size of the TSPP aggregates remains different depending on the method of sample preparations. This conclusion is drawn from a headto-head comparison of the kinetics of the porphyrin assembly and disassembly under acidic pHs (Figures 4 and 5). The porphyrin assembly and disassembly samples differ essentially only in the way of the sample preparation, as specified in the Experimental Section, but their sample compositions in terms

nominal TSPP concentration only when the TSPP concentration varies from 0.05 to 0.2 μM (Figures 3B, and S4, Supporting Information). Large deviation appears when the TSPP concentration is above 0.2 μM, and it reaches a plateau when the nominal TSPP concentration is 1 μM (Figure 3A,B). Using the linear calibration curve derived with the 435 nm marker peak for the isolated TSPP with the four leastconcentrated samples used in Figure 3 (Figure S4, Supporting Information), one can estimate the concentration of the isolated and aggregated TSPPs (in terms of their constituent monomers) for all of the samples with nominal TSPP concentration above 0.2 μM. The percentage of the TSPP aggregates increases from less than 5% into more than 70%, when the nominal TSPP concentration increases from 0.3 to 3 μM (Figure 3). The average size of the TSPP aggregates increases with increasing nominal TSPP concentration, which is deduced from the TSPP light-scattering cross-section measurements. The nominal TSPP light-scattering cross section at 490 nm increases from 2.0 × 10−19 to 54 × 10−19 cm2, when the nominal TSPP concentration increases from 0.15 to 1 μM (Figure S4, Supporting Information). The actual lightscattering cross section of the TSPP aggregates should be significantly larger than the nominal one. This is because the molarity of the aggregates should be drastically smaller than 8434


Article


Figure 5. Time-dependent study of TSPP (red) assembly and (black) disassembly for up to 127 days. The nominal TSPP concentration of the TSPP assembly and disassembly samples is the same. Plots (A−E) in the first row are UV−vis extinction spectra of TSPP assembly and disassembly samples. Plots (F−J) in the second rows are PRS2 spectra of TSPP assembly and disassembly samples. Data in the third and fourth row are the sample-polarized SSF spectra obtained with excitation wavelengths of (K−O) 430 nm and (P−T) 490 nm, respectively. The peak intensity of extinction, PRS2 VV, and SSF VV spectra as a function of number of incubation days are shown in Supporting Information (Figure S5).

porphyrin can also be readily deduced from the TSPP monomer’s 435 nm UV−vis peak, which increases right after the sample preparation (Figure 4G). Despite their rapid onset, however, the TSPP assembly and disassembly last at least for 4 months (Figure 5). This is because the 490 nm UV−vis and PRS2 associated with the TSPP aggregates monotonically decreases in TSPP disassembly sample but increases in the assembly sample during the 127 days sample incubation period. The time courses of the fluorescence marker peaks for the isolated TSPP (655 nm) and aggregated TSPP (710 nm) are much more complicated (third and fourth rows in Figure 5). Such a complication is due most likely to the fluorescence photobleaching occurring during such a prolonged sample incubation period and/or during the periodic optical spectroscopic measurements. This is in spite of our effort in minimizing the light exposure of the samples by placing them in a box covered with an aluminum foil. Indeed, the decreased TSPP SSF intensities at 655 and 710 nm in the 127 days samples (Figure 5O,T) than their respective 72 days counterpart (Figure 5N,S) are due likely to the fluorescence photobleaching of isolated and aggregated TSPPs during their long-term incubation. The fact that the 490 nm UV−vis and PRS2 associated with the TSPP assembling sample become increasingly similar to their respective counterpart of the TSPP disassembling sample, strongly indicating that the TSPP assembly and disassembly are reversible, even though the kinetics to reach such equilibrium state is extraordinarily slow. With the intensity of the 435 nm UV−vis peak (first row in Figure 5) and the 655 nm fluorescence peak (third row in Figure 5), the two markers

of the HCl, ionic strength, and porphyrin concentration are approximately the same. The kinetics of the porphyrin assembly and disassembly within a 3 h sample preparation were monitored with time-dependent UV−vis and PRS2 spectral measurements with relatively short intervals (5−10 min) between sequential measurements (Figure 4). The data obtained with long-term kinetics study over a time course of 127 days are shown in Figure 5. This prolonged kinetics study offers a series of new insights to porphyrin assembly and disassembly processes that are not available from existing studies. Apparently, the PRS2 data indicated that both porphyrin assembly and disassembly occur immediately after the samples were prepared (Figure 4). The 490 nm PRS2 peak appears upon acidification of the neutrally prepared porphyrin solution (within a measurement dead time of ∼3 min) and then monotonically increases during an entire period of sample incubation. This is in contrast to the earlier report that the porphyrin aggregation undergoes an “induction period”.47 The initial porphyrin assembly totally escaped the UV−vis spectral acquisition. Indeed, there is no spectral variation during the entire first 3 h sample incubation at either the 435 or 490 nm UV−vis peak in the TSPP assembly sample (Figure 4A). This provides further evidence of superior PRS2 sensitivity than UV−vis for detecting the fluorophore assembly. The poor UV−vis sensitivity in detecting the early-stage TSPP aggregation may explain the “induction period” reported in a previous UV−vis study of the TSPP assembly.47 Conversely, the 490 nm PRS2 peak in the disassembly sample rapidly monotonically decreases as a function of the sample incubation time. The disassembly of the aggregated 8435


Article


months of sample incubation. One important implication of this slow kinetics is that the conclusions drawn from all of the existing porphyrin studies essentially completed within a few hours of the sample preparation are likely snapshots of the porphyrin aggregation and disintegration processes that may be far from equilibrium. The insights provided in this study are important for in-vitro monitoring of the molecular selfassembly of dye molecules that involves photon/matter interactions, whereas the provided methodology is directly applicable for experimental quantification of optical properties of dye aggregates that are simultaneous photon absorbers, emitters, and scatters.

of isolated TSPPs become essentially the same after the 127 days sample incubation. This shows that the TSPP assembly and disassembly have reached equilibrium in terms of the isolated and aggregated TSPP concentrations somewhere between the 72 and 127 days sample incubation. However, even after the 127 days sample incubation, the marker peaks (490 nm UV−vis and PRS2 peaks, and the 710 nm fluorescence peak) associated with aggregated TSPPs remain significantly different in the assembly and disassembly samples, indicating that the structures of the TSPP aggregates in these two samples are different even with such a prolonged sample incubation. The fact that the 490 nm PRS2 peak in the disassembly sample is substantially higher than that in the assembly sample, indicating that the average size of the final TSPP aggregates remain significantly larger in the disassembly sample than that in the assembly sample. The change of depolarization in the short-term kinetics study is informative on the aggregation process (Figure 4F). The time course of the light-scattering depolarization at 490 nm of the TSPP assembling sample follows the first-order reaction scheme in which the scattering depolarization rapidly increases from 0 to 0.15 during initial the 50 min sample incubation (Figure 4) but varies very slowly afterward. Earlier research indicates that TSPP forms a nanoscale ring that subsequently assembles into a tubular structure. The fact that the light-scattering depolarization at 490 nm rapidly rises to its maximum in the assembly sample indicates that such lightscattering depolarization is sensitive only to the ring formation or to the initial ring assembly to the tubular structure. Once the length of the tube reaches a threshold value, further elongation of the tubes has no significant effect on the light scattering of the TSPP aggregates. This explanation is consistent with the observation that the light-scattering depolarization of the 490 nm peak remains constant during the entire sample incubation. It is also in agreement with the earlier computational simulation performed with the AuNP as a function of its aspect ratio. The AuNP light-scattering depolarization increases initially when the aspect ratio increases from 1 (perfectly spherical) to ∼2.5 (rod) and then remains as an essential constant when the rod is further elongated.48

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.8b05965. As-acquired SSF and PRS2 spectrum before the IFEand solvent background correction of the as-prepared and filtrate TSPP solutions; ratiometric BVPRS2 study of TSPP; polarized SSF and PRS2 spectra acquired with excitation and detection polarization combination of “VH”; UV−vis extinction intensity at 435 nm and scattering cross sections at 490 nm as a function of TSPP concentration; and peak intensity of extinction, PRS2 VV, and SSF VV spectra for TSPP association and disassociation as a function of number of incubation days (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], dmzhang@mail. xhu.edu.cn. Phone: (662) 325-6752. Fax: 662-325-1618. ORCID

Dongmao Zhang: 0000-0002-2303-7338 Author Contributions §

B.C.N.V. and J.X.X. equally Contributed.

Notes

The authors declare no competing financial interest.

■

■ ■

CONCLUSIONS Using UV−vis and polarized SSF in combination with the recently developed PRS2 method, we have experimentally quantified the photon extinction, absorption, and scattering and fluorescence cross sections, as well as scattering and fluorescence depolarizations of isolated and aggregated TSPP in acidic medium. PRS2 is drastically more sensitive than the popular UV−vis extinction measurement in detecting the TSPP assembly and disassembly. The structural evolution and kinetics of the TSPP assembly and disassembly has also been investigated. The fluorescence depolarization of the aggregated TSPP is 0.5, which is extraordinarily small in comparison with the isolated TSPP and other dispersed or aggregated fluorophores. The BVPRS2 measurement provides conclusive information that the porphyrin PRS2 spectrum contains both light-scattering and fluorescence spectral features and not just the RLS indicated in earlier literature. Another critical new learning is that while the onset of the TSPP assembly and disassembly is instantaneous upon the sample preparations, the process for the aggregated TSPP to reach their equilibrium configurations is extraordinarily lengthy, taking at least 4

ACKNOWLEDGMENTS This work was supported by the NSF funds (CHE 1151057) and (EPS-0903787) provided to D.Z. REFERENCES

(1) Parkash, J.; Robblee, J. H.; Agnew, J.; Gibbs, E.; Collings, P.; Pasternack, R. F.; de Paula, J. C. Depolarized Resonance Light Scattering by Porphyrin and Chlorophyll a Aggregates. Biophys. J. 1998, 74, 2089−2099. (2) Deng, S.; Advincula, R. C. Dual Aggregation Behavior of Porphyrin-Cored Thiophene Dendrimers. Macromol. Rapid Commun. 2011, 32, 1634−1639. (3) Blase, X.; Attaccalite, C.; Olevano, V. First-principles GW calculations for fullerenes, porphyrins, phtalocyanine, and other molecules of interest for organic photovoltaic applications. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 83, 115103. (4) Wang, Z.; Medforth, C. J.; Shelnutt, J. A. Porphyrin nanotubes by ionic self-assembly. J. Am. Chem. Soc. 2004, 126, 15954−15955. (5) Berg, K.; Bommer, J. C.; Winkelman, J. W.; Moan, J. Cellular uptake and relative efficiency in cell inactivation by photo activated sulfonated meso-tetraphenylporphines. J. Photochem. Photobiol., A 1990, 52, 775−781.

8436


Article


for Photodynamic Cancer Therapy. ACS Appl. Mater. Interfaces 2017, 9, 26731−26739. (24) Pasternack, R.; Collings, P. Resonance light scattering: a new technique for studying chromophore aggregation. Science 1995, 269, 935−939. (25) Pasternack, R. F.; Bustamante, C.; Collings, P. J.; Giannetto, A.; Gibbs, E. J. Porphyrin assemblies on DNA as studied by a resonance light-scattering technique. J. Am. Chem. Soc. 1993, 115, 5393−5399. (26) Shortell, M. P.; Hewins, R. A.; Fernando, J. F. S.; Walden, S. L.; Waclawik, E. R.; Jaatinen, E. A. Multi-angle fluorometer technique for the determination of absorption and scattering coefficients of subwavelength nanoparticles. Opt. Express 2016, 24, 17090−17102. (27) Chen, Z.; Lei, Y.; Gao, W.; Liu, J. Detection of Vascular Endothelial Growth Factor Based on Gold Nanoparticles and Immunoreaction Using Resonance Light Scattering. Plasmonics 2013, 8, 605−611. (28) Xu, J. X.; Hu, J.; Zhang, D. Quantification of Material Fluorescence and Light Scattering Cross Sections Using Ratiometric Bandwidth-Varied Polarized Resonance Synchronous Spectroscopy. Anal. Chem. 2018, 90, 7406−7414. (29) Siriwardana, K.; Vithanage, B. C. N.; Zou, S.; Zhang, D. Quantification of the Depolarization and Anisotropy of Fluorophore Stokes-Shifted Fluorescence, On-Resonance Fluorescence, and Rayleigh Scattering. Anal. Chem. 2017, 89, 6686−6694. (30) Siriwardana, K.; Nettles, C. B.; Vithanage, B. C. N.; Zhou, Y.; Zou, S.; Zhang, D. On-Resonance Fluorescence, Resonance Rayleigh Scattering, and Ratiometric Resonance Synchronous Spectroscopy of Molecular- and Quantum Dot-Fluorophores. Anal. Chem. 2016, 88, 9199−9206. (31) Xu, J. X.; Siriwardana, K.; Zhou, Y.; Zou, S.; Zhang, D. Quantification of Gold Nanoparticle Ultraviolet-Visible Extinction, Absorption, and Scattering Cross-Section Spectra and Scattering Depolarization Spectra: The Effects of Nanoparticle Geometry, Solvent Composition, Ligand Functionalization, and Nanoparticle Aggregation. Anal. Chem. 2017, 90, 785−793. (32) Athukorale, S. A.; Zhou, Y.; Zou, S.; Zhang, D. Determining the Liquid Light Scattering Cross Section and Depolarization Spectra Using Polarized Resonance Synchronous Spectroscopy. Anal. Chem. 2017, 89, 12705−12712. (33) Deng, S.; Advincula, R. C. Dual Aggregation Behavior of Porphyrin-Cored Thiophene Dendrimers. Macromol. Rapid Commun. 2011, 32, 1634−1639. (34) Cui, L.; Lin, Q.; Jin, C. S.; Jiang, W.; Huang, H.; Ding, L.; Muhanna, N.; Irish, J. C.; Wang, F.; Chen, J.; Zheng, G. A PEGylation-Free Biomimetic Porphyrin Nanoplatform for Personalized Cancer Theranostics. ACS Nano 2015, 9, 4484−4495. (35) Bahsoun, H.; Chervy, T.; Thomas, A.; Börjesson, K.; Hertzog, M.; George, J.; Devaux, E.; Genet, C.; Hutchison, J. A.; Ebbesen, T. W. Electronic Light-Matter Strong Coupling in Nanofluidic FabryPérot Cavities. ACS Photonics 2018, 5, 225−232. (36) Purrello, R.; Monsu’ Scolaro, L.; Bellacchio, E.; Gurrieri, S.; Romeo, A. Chiral H- and J-Type Aggregates ofmeso-Tetrakis(4sulfonatophenyl)porphine on α-Helical Polyglutamic Acid Induced by Cationic Porphyrins. Inorg. Chem. 1998, 37, 3647−3648. (37) Wan, Y.; Stradomska, A.; Knoester, J.; Huang, L. Direct Imaging of Exciton Transport in Tubular Porphyrin Aggregates by Ultrafast Microscopy. J. Am. Chem. Soc. 2017, 139, 7287−7293. (38) Wan, Y.; Stradomska, A.; Fong, S.; Guo, Z.; Schaller, R. D.; Wiederrecht, G. P.; Knoester, J.; Huang, L. Exciton level structure and dynamics in tubular porphyrin aggregates. J. Phys. Chem. C 2014, 118, 24854−24865. (39) Schwab, A. D.; Smith, D. E.; Rich, C. S.; Young, E. R.; Smith, W. F.; de Paula, J. C. Porphyrin nanorods. J. Phys. Chem. B 2003, 107, 11339−11345. (40) Vlaming, S. M.; Augulis, R.; Stuart, M. C. A.; Knoester, J.; van Loosdrecht, P. H. M. Exciton spectra and the microscopic structure of self-assembled porphyrin nanotubes. J. Phys. Chem. B 2009, 113, 2273−2283.

(6) Gibbs, E. J.; Tinoco, I., Jr.; Maestre, M. F.; Ellinas, P. A.; Pasternack, R. F. Self-assembly of porphyrins on nucleic acid templates. Biochem. Biophys. Res. Commun. 1988, 157, 350−358. (7) Gandini, S. C. M.; Borissevitch, I. E.; Perussi, J. R.; Imasato, H.; Tabak, M. Aggregation of meso-tetrakis(4-N-methyl-pyridiniumyl) porphyrin in its free base, Fe(III) and Mn(III) forms due to the interaction with DNA in aqueous solutions: Optical absorption, fluorescence and light scattering studies. J. Lumin. 1998, 78, 53−61. (8) Hollingsworth, J. Synthesis, characterization, and self-assembly of porphyrins conjugated to superparamagnetic colloidal particles for enhanced photodynamic therapy. Ph.D. Dissertation, Louisiana State University, 2012. (9) Micali, N.; Mallamace, F.; Romeo, A.; Purrello, R.; Monsù Scolaro , L. Meso scopi c Struct ure ofme so-Te trakis (4sulfonatophenyl)porphine J-Aggregates. J. Phys. Chem. B 2000, 104, 5897−5904. (10) Andrade, S. M.; Costa, S. M. B. Aggregation kinetics of mesotetrakis (4-sulfonatophenyl) porphine in the presence of proteins: temperature and ionic strength effects. J. Fluoresc. 2002, 12, 77−82. (11) Pasternack, R.; Collings, P. Resonance light scattering: a new technique for studying chromophore aggregation. Science 1995, 269, 935−939. (12) Maiti, N. C.; Ravikanth, M.; Mazumdar, S.; Periasamy, N. Fluorescence dynamics of noncovalently linked porphyrin dimers, and aggregates. J. Phys. Chem. A 1995, 99, 17192−17197. (13) Udal’tsov, A. V.; Kazarin, L. A.; Sweshnikov, A. A. Self-assembly of large-scale aggregates of porphyrin from its dimers and their absorption and luminescence properties. J. Mol. Struct. 2001, 562, 227−239. (14) Collings, P. J.; Gibbs, E. J.; Starr, T. E.; Vafek, O.; Yee, C.; Pomerance, L. A.; Pasternack, R. F. Resonance Light Scattering and Its Application in Determining the Size, Shape, and Aggregation Number for Supramolecular Assemblies of Chromophores. J. Phys. Chem. B 1999, 103, 8474−8481. (15) Purrello, R.; Monsu’ Scolaro, L.; Bellacchio, E.; Gurrieri, S.; Romeo, A. Chiral H- and J-Type Aggregates ofmeso-Tetrakis(4sulfonatophenyl)porphine on α-Helical Polyglutamic Acid Induced by Cationic Porphyrins. Inorg. Chem. 1998, 37, 3647−3648. (16) Kitahama, Y.; Kimura, Y.; Takazawa, K. Study of Internal Structure ofmeso-Tetrakis (4-Sulfonatophenyl) Porphine J-Aggregates in Solution by Fluorescence Microscope Imaging in a Magnetic Field. Langmuir 2006, 22, 7600−7604. (17) Romeo, A.; Castriciano, M. A.; Occhiuto, I.; Zagami, R.; Pasternack, R. F.; Scolaro, L. M. Kinetic Control of Chirality in Porphyrin J-Aggregates. J. Am. Chem. Soc. 2014, 136, 40−43. (18) Fathalla, M.; Neuberger, A.; Li, S.-C.; Schmehl, R.; Diebold, U.; Jayawickramarajah, J. Straightforward Self-Assembly of Porphyrin Nanowires in Water: Harnessing Adamantane/β-Cyclodextrin Interactions. J. Am. Chem. Soc. 2010, 132, 9966−9967. (19) Kobayashi, Y.; Katayama, T.; Yamane, T.; Setoura, K.; Ito, S.; Miyasaka, H.; Abe, J. Stepwise Two-Photon-Induced Fast Photoswitching via Electron Transfer in Higher Excited States of Photochromic Imidazole Dimer. J. Am. Chem. Soc. 2016, 138, 5930−5938. (20) Cao, W.; Sletten, E. M. Fluorescent Cyanine Dye J-Aggregates in the Fluorous Phase. J. Am. Chem. Soc. 2018, 140, 2727−2730. (21) Yang, G.; Tang, Y.; Li, X.; Ågren, H.; Xie, Y. Efficient Solar Cells Based on Porphyrin Dyes with Flexible Chains Attached to the Auxiliary Benzothiadiazole Acceptor: Suppression of Dye Aggregation and the Effect of Distortion. ACS Appl. Mater. Interfaces 2017, 9, 36875−36885. (22) Shih, K.-Y.; Hsiao, T.-S.; Deng, S.-L.; Hong, J.-L. Water-Soluble Poly(γ-propargyl-l-glutamate) Containing Pendant Sulfonate Ions and Terminal Fluorophore: Aggregation-Enhanced Emission and Secondary Structure. Macromolecules 2014, 47, 4037−4047. (23) Guan, Y.; Lu, H.; Li, W.; Zheng, Y.; Jiang, Z.; Zou, J.; Gao, H. Near-Infrared Triggered Upconversion Polymeric Nanoparticles Based on Aggregation-Induced Emission and Mitochondria Targeting 8437


Article

The Journal of Physical Chemistry B (41) Maraka, H. V. R.; Al-Shammari, R. M.; Al-Attar, N.; Lopez, S. G.; Keyes, T. E.; Rice, J. H. Graphene oxide intercalation into selfassembled porphyrin J-aggregates. Mater. Res. Express 2014, 1, 045038. (42) Hollingsworth, J. V.; Richard, A. J.; Vicente, M. G. H.; Russo, P. S. Characterization of the Self-Assembly of meso-Tetra(4sulfonatophenyl)porphyrin (H2TPPS4-) in Aqueous Solutions. Biomacromolecules 2011, 13, 60−72. (43) Nettles, C. B.; Hu, J.; Zhang, D. Using Water Raman Intensities To Determine the Effective Excitation and Emission Path Lengths of Fluorophotometers for Correcting Fluorescence Inner Filter Effect. Anal. Chem. 2015, 87, 4917−4924. (44) Trapani, M.; De Luca, G.; Romeo, A.; Castriciano, M. A.; Scolaro, L. M. Spectroscopic investigation on porphyrins nanoassemblies onto gold nanorods. Spectrochim. Acta, Part A 2017, 173, 343−349. (45) Akins, D. L.; Ö zçelik, S.; Zhu, H.-R.; Guo, C. Fluorescence Decay Kinetics and Structure of Aggregated Tetrakis(pSulfonatophenyl)Porphyrin. J. Phys. Chem. A 1996, 100, 14390− 14396. (46) Zhang, W.; Jin, W.; Fukushima, T.; Saeki, A.; Seki, S.; Aida, T. Supramolecular Linear Heterojunction Composed of Graphite-Like Semiconducting Nanotubular Segments. Science 2011, 334, 340−343. (47) Pasternack, R. F.; Fleming, C.; Herring, S.; Collings, P. J.; dePaula, J.; DeCastro, G.; Gibbs, E. J. Aggregation Kinetics of Extended Porphyrin and Cyanine Dye Assemblies. Biophys. J. 2000, 79, 550−560. (48) Xu, J. X.; Siriwardana, K.; Zhou, Y.; Zou, S.; Zhang, D. Quantification of Gold Nanoparticle Ultraviolet-Visible Extinction, Absorption, and Scattering Cross-Section Spectra and Scattering Depolarization Spectra: The Effects of Nanoparticle Geometry, Solvent Composition, Ligand Functionalization, and Nanoparticle Aggregation. Anal. Chem. 2018, 90, 785−793.

8438


Optical Properties and Kinetics: New Insights to the Porphyrin

Recommend Documents