Merging Porphyrins with Gold Nanorods: Self Assembly Construct to High Fluorescent Polyelectrolyte Microcapsules

Dual probe porphyrin-gold nanorod polyelectrolyte microcapsules were developed to explore the enhancing effects of a plasmonic interface of self-assembled gold nanoparticles in the fluorescence emission from porphyrins loaded into the capsules’ core. An analysis of fluorescence lifetime imaging microscopy (FLIM) data reports a notable 105–106-fold increase in the maximum detected photon rates from diffraction-limited spots and an overall six-fold increase in fluorescence as averaged over the whole microcapsule area. Large emission enhancements were correlated with decreases in fluorescence lifetimes. The microcapsule’s design proved effective in achieving high fluorescent hybrids and may shed light on new possibilities for advanced materials imaging applications.


Introduction
In recent years, polyelectrolyte microcapsules (PECs) have received a great deal of attention due to their possible use as drug delivery systems, biosensors, and microreactors [1,2]. Using the adequate polymers and number of assembled layers, fine control over a PEC's structure can be achieved, allowing for devices with high biocompatibility and mechanical/chemical stability [1,[3][4][5], which can load hydrophobic and hydrophilic molecules and promote their controlled and targeted release [6][7][8].
The integration of dyes, such as porphyrins, within these platforms enables the production of optically addressable devices and expands PECs' use towards fields involving light. Porphyrins are important light-absorbing molecules that are largely recognized as photosensitizers for cancer photodynamic therapy and diagnosis due to their intrinsic ability to generate cytotoxic oxygen species and to emit in the red or near infrared when exposed to visible or UV light [9]. Nevertheless, most porphyrins are weak emitters, have solubility issues, and tend to aggregate in water [10,11], thus hampering most cases for their application in biomedicine. Porphyrinoids' functionalization with hydrophilic units [12][13][14][15] is an elegant synthetic strategy used to overcome porphyrins' low solubility. Nanotechnological development, is also encouraging the use of different nanoparticles such as gold nanoparticles (AuNP), as porphyrinoid carriers to improve imaging and photodynamic capabilities [16][17][18][19]. Bédard et al. were the first to incorporate porphyrins in the PEC's shell and to study the use of 1 O 2 formed upon light irradiation to promote dye release [20]. Other examples include the studies of the doping procedure impact and shell nature on the optical and electronic properties of porphyrins [21,22] and as templates for the construction of supramolecular porphyrin assemblies [11,23]. Laser-responsive AuNP photothermal

Porphyrin-Nanogold PEC's Hybrid Preparation
Porphyrin-nanogold PEC hybrids were prepared according to the following steps: (i) Porphyrin embedded CaCO3 template preparation and polyelectrolytes wrapping (Scheme 1a-c

Porphyrin-Nanogold PEC's Hybrid Preparation
Porphyrin-nanogold PEC hybrids were prepared according to the following steps: (i) Porphyrin embedded CaCO 3 template preparation and polyelectrolytes wrapping (Scheme 1a-c). Equal volumes of 100 µL of saturated solutions of Na 2 CO 3 (0.33 M with PSS 18 wt%) and CaCl 2 (0.33 M) were added under intense stirring to 50 µL of an aqueous solution of TMPyP (260 µM). After 30 s, the mixture was allowed to rest for 15 min. Porphyrin doped CaCO 3 microparticles were obtained after supernatant removal, followed by three washing/centrifugation cycles (3000 rpm, 2 min) to remove any non-encapsulated dye. Similar procedures in the absence of porphyrin were used to prepare nonfunctionalized CaCO 3 PECs. CaCO 3 microparticles (with or without TMPyP) were then dispersed in an aqueous solution of PSS (1 mL, 3 mg/mL, 0.5 M NaCl). After stirring for 15 min, the resulting particles were centrifuged (6000 rpm, 10 min), and the supernatant was removed. After three washing/centrifugation cycles to remove residual PSS, the obtained microparticles were dispersed in an aqueous solution of PAH (1 mL, 12 mg/mL, 0.5 M NaCl). After the usual washing/centrifugation procedures, PECs were obtained free from residual PSS and stored in bi-distilled water. (ii) Gold nanorods adsorption onto porphyrin embedded PECs (Scheme 1d-e): PSS wrapped gold nanorods: AuNR solution (1 mL, OD = 1) were centrifuged (6000 rpm, 15 min). After supernatant removal, PSS polyelectrolyte (1 mL, 1.7 mg/mL, 1 mM NaCl) was added. The mixture was vigorously stirred and after 1 h it was centrifuged and washed with bi-distilled water (3 cycles, 6000 rpm, 15 min) to remove the excess of PSS. These PSS-wrapped gold nanorods were then resuspended in 500 µL of bidistilled water and added to previously prepared porphyrin embedded polyelectrolyte microcapsules. After the usual washing/centrifugation steps, porphyrin-nanogold PECs hybrids were stored in bi-distilled water.

Characterization
UV-Vis absorption measurements were performed in a PerkinElmer Lambda 35 spectrophotometer (PerkinElmer Inc., Ltd., Waltham, MA, USA). Corrected fluorescence measurements were recorded with a Fluorolog-3 spectrophotometer (Horiba Jobin Yvon, Tokyo, Japan). The cuvettes used for spectrophotometry were quartz cuvettes from Hellma (optical path length 10 mm, 700 µL volume). The effective zeta potential values were measured in a Zetasizer Nano ZS Doppler electrophoretic light scattering analyzer from Malvern Instruments Inc., Ltd. (Malvern, UK). The pH was measured at 24.0 • C with a Crison microph 2002 (Crison, Barcelona, Spain). The samples were centrifuged using a Spectrafuge 24D Digital Microcentrifuge (Labnet, Orange, CA, USA), with a centrifuging radius of 8.23 cm. Samples for electron microscopy were prepared by dispensing one or two drops of PECs dispersed in water on carbon-coated copper grids and were allowed to dry at room temperature overnight. Scanning transmission electron microscopy (STEM) measurements were performed on a TEM/STEM Hitachi microscope model HD2700 (Schaumburg, IL, USA) operating at 200 kV, equipped with EDS and HAADF detector. Scanning electron microscopy measurements were performed with Analytical FEG-SEM equipment JEOL 7001F (JEOL Inc., Ltd., Tokyo, Japan) with an Oxford light elements EDS detector and an EBSD detector. FLIM measurements were performed with a time-resolved confocal microscope (MicroTime 200, PicoQuant GmbH, Berlin, Germany). Cover-slips were carefully cleaned prior to use. Cover glasses were first immersed in an RBS solution (5%) for 2 h followed by rinsing with ultrapure water. After drying, cover slips were exposed to UV/Ozone (equipment PSD-UV3 from Novascan) for 2 h. Briefly, excitation at 639 nm was carried out by pulsed diode laser heads at a repetition rate of 20 MHz. Band-pass filter 695AF55 was used for the red diode laser. The fluorescence lifetimes were detected with a single-photon-counting avalanche diode (SPAD) (PerkinElmer Inc., Ltd., Waltham, MA, USA) whose signal is processed by a TimeHarp 200 TC-SPC PCboard (PicoQuant GmbH, Berlin, Germany) working in time-tagged time-resolved (TTTR) operation mode. The emission intensity time traces were collected from random points, chosen within PECs during a time interval of 180 s, using an excitation power of ca 0.44 kW/cm 2 to avoid porphyrin's photobleaching.

Discrete Dipole Approximation Simulations
The method of discrete dipole approximation (DDA) was used [40,41] to model plasmonic hot spots between closely packed AuNRs and the enhancement effect on porphyrin's emission. A full detailed description of the calculation procedures can be found elsewhere [35,42]. Firstly, the near field maps were calculated for a set of particle configurations comprising two, four, and seven nanorods, in order to identify the possible plasmon hot spots. The shortest gap distance between nanorods was set at 3 nm to account for the polymer coating in the experiment. The model nanorods are spherically capped cylinders with a diameter of 25 nm and length of 61 nm that were discretized as an array of cubic elements with a side of 0.5 nm. The dielectric function of gold was used as reported by Johnson and Christy [43]. Then, the porphyrin was modeled as a point-like dipole emitting at specific wavelengths, each representing a spectral component within the dye's emission spectrum [37]. The plasmon-enhanced radiative and non-radiative decay rates for the emitter in the plasmon hot spots were calculated using the theoretical formalism described by D'Agostino et al. [42].

Preparation and Characterization of Porphyrin-Nanogold Polyelectrolyte Microcapsules Hybrids
TMPyP is a tetracationic porphyrin known for its high singlet oxygen quantum yield (Φ ∆ = 0.7 in phosphate-buffered solution), a noteworthy affinity for single and doublestrand DNA polyanion [44]. It is highly selective, targeting cancer cells over non-tumorous cells in a wide variety of established cancer models [45,46]. TMPyP has a molar absorption coefficient (ε) value of around 10 3 M −1 cm −1 at the excitation wavelength of 639 nm and a low fluorescence quantum yield of 5% due to the mixing of the S1 state with a charge transfer state (CT) involving an electron transfer pathway between the N atoms of the pyrrolic units and the meso pyridinium substituents which has been well covered in other studies [47] Due to four N-methylpyridinium meso-substituents, this porphyrin is water soluble and able to self-assemble through electrostatic interactions with negatively charged polyelectrolytes, the primary constituents of the polyelectrolyte shell. Figure 1 shows the optical spectra of individual TMPyP and AuNRs that were obtained in water before encapsulation in PECs.

Preparation and Characterization of Porphyrin-Nanogold Polyelectrolyte Microcapsules Hybrids
TMPyP is a tetracationic porphyrin known for its high singlet oxygen quantum yield (Φ∆ = 0.7 in phosphate-buffered solution), a noteworthy affinity for single and doublestrand DNA polyanion [44]. It is highly selective, targeting cancer cells over non-tumorous cells in a wide variety of established cancer models [45,46]. TMPyP has a molar absorption coefficient (ε) value of around 10 3 M −1 cm −1 at the excitation wavelength of 639 nm and a low fluorescence quantum yield of 5% due to the mixing of the S1 state with a charge transfer state (CT) involving an electron transfer pathway between the N atoms of the pyrrolic units and the meso pyridinium substituents which has been well covered in other studies [47] Due to four N-methylpyridinium meso-substituents, this porphyrin is water soluble and able to self-assemble through electrostatic interactions with negatively charged polyelectrolytes, the primary constituents of the polyelectrolyte shell. Figure 1 shows the optical spectra of individual TMPyP and AuNRs that were obtained in water before encapsulation in PECs. The absorption spectrum of TMPyP displays a Soret maximum at 422 nm (from S0 → S2 transition) and four satellite bands in the 500-700 nm region (from S0 → S1 transition). In general, free base water soluble monomeric porphyrins present a characteristic emission spectrum with two vibrational bands, Q(0,0) and Q(0,1), around 650 nm and 720 nm, respectively. The unresolved emission spectra of the porphyrin TMPyP is explained by the above-mentioned CT state [47]. AuNRs have two plasmon modes; the transversal surface plasmon resonance (SPR) mode at 518 nm, excited by the light polarized along the rod's transverse direction, and the longitudinal SPR mode at 658 nm, which is associated with electron oscillations across the rod's long axis [48] that gives rise to large nearfield enhancements. The observed spectral overlap between AuNRs' longitudinal mode and TMPyPs' absorption (Q bands) and emission spectra, enables the direct excitation of both dye and plasmon oscillation. Therefore, we would expect a resonant plasmonic enhancement of porphyrin's molecular emission. Still, TMPyP and AuNR may exhibit small differences in absorption and emission due to encapsulation.
The preparation and characterization of the porphyrin-nanogold PEC hybrids will be discussed in detail in the next sections, following the summarized procedure in Scheme 1 (Section 2.2).
The typical design of PECs doped with low molecular dyes usually starts with the dyes' adsorption on the polyelectrolyte shell by taking advantage of their ionic character. In this work, an alternative co-precipitation approach was used to promote porphyrin encapsulation within the CaCO3 core matrix during microparticle growth (spectroscopic The absorption spectrum of TMPyP displays a Soret maximum at 422 nm (from S 0 → S 2 transition) and four satellite bands in the 500-700 nm region (from S 0 → S 1 transition). In general, free base water soluble monomeric porphyrins present a characteristic emission spectrum with two vibrational bands, Q(0,0) and Q(0,1), around 650 nm and 720 nm, respectively. The unresolved emission spectra of the porphyrin TMPyP is explained by the above-mentioned CT state [47]. AuNRs have two plasmon modes; the transversal surface plasmon resonance (SPR) mode at 518 nm, excited by the light polarized along the rod's transverse direction, and the longitudinal SPR mode at 658 nm, which is associated with electron oscillations across the rod's long axis [48] that gives rise to large nearfield enhancements. The observed spectral overlap between AuNRs' longitudinal mode and TMPyPs' absorption (Q bands) and emission spectra, enables the direct excitation of both dye and plasmon oscillation. Therefore, we would expect a resonant plasmonic enhancement of porphyrin's molecular emission. Still, TMPyP and AuNR may exhibit small differences in absorption and emission due to encapsulation.
The preparation and characterization of the porphyrin-nanogold PEC hybrids will be discussed in detail in the next sections, following the summarized procedure in Scheme 1 (Section 2.2).
The typical design of PECs doped with low molecular dyes usually starts with the dyes' adsorption on the polyelectrolyte shell by taking advantage of their ionic character. In this work, an alternative co-precipitation approach was used to promote porphyrin encapsulation within the CaCO 3 core matrix during microparticle growth (spectroscopic characterization details can be found in Table S1). Afterward, the full coverage of TMPyP-CaCO 3 microparticles with a bilayer of PSS-PAH polyelectrolytes ( Figure S2) took place, using well-known layer-by-layer methods [11,22] to prevent TMPyP release and provide a fair nanometric control over the distance between porphyrin and AuNRs. Commercially available AuNRs are positively charged due to the presence of a bilayer of CTAB (ζ = +30 mV) [49,50], and therefore are not suitable for PAH adsorption without further functionalization.
In our case, the modification of the entire nanorod surface is preferable to its tip functionalization (where plasmonic hot spots are located) [51][52][53] since a higher number of attaching points will increase the contact area, strengthening AuNR adsorption onto a (CaCO 3 -TMPyP)-PSS-PAH surface. Due to the polymeric nature of PECs, AuNR overcoating with PSS [49,50], which introduces sulfonate groups at their surface which allows for their adsorption to positively charged -NH 3 + groups in the PECs last layer, seems to us a natural choice. Furthermore, polyelectrolyte coverage significantly reduces CTAB cytotoxicity and increases AuNRs' stability in water [52,54,55].
Our UV-Vis data documented a 3 nm red shift (from 658 to 661 nm, Figure S3) in the longitudinal plasmon mode after PSS adsorption, which is consistent with the UV-Vis changes reported in reference [49]. Zeta potential measurements also support this, as a charge reversal is observed from +29.5 ± 5 mV to −29.9 ± 2 mV.
The as-prepared AuNR-PSS was added to a known amount of (CaCO 3 -TMPyP)-PSS-PAH (Scheme 1, Section 2.2), and a change in the PEC's color from light red to dark blue was observed. The UV-Vis spectrum of the porphyrin-nanogold PEC hybrids (Figure 2A) in water presents the typical signatures of TMPyP and AuNR entities, confirming their successful encapsulation: a TMPyP Soret band around 432 nm, a broad band between 450-650 nm attributed to a superposition of TMPyP Q bands and the transverse plasmon mode, and a NIR band at 717 nm that originates from the longitudinal plasmon mode. characterization details can be found in Table S1). Afterward, the full coverage of TMPyP-CaCO3 microparticles with a bilayer of PSS-PAH polyelectrolytes ( Figure S2) took place, using well-known layer-by-layer methods [11,22] to prevent TMPyP release and provide a fair nanometric control over the distance between porphyrin and AuNRs. Commercially available AuNRs are positively charged due to the presence of a bilayer of CTAB (= + 30 mV) [49,50], and therefore are not suitable for PAH adsorption without further functionalization. In our case, the modification of the entire nanorod surface is preferable to its tip functionalization (where plasmonic hot spots are located) [51][52][53] since a higher number of attaching points will increase the contact area, strengthening AuNR adsorption onto a (CaCO3-TMPyP)-PSS-PAH surface. Due to the polymeric nature of PECs, AuNR overcoating with PSS [49,50], which introduces sulfonate groups at their surface which allows for their adsorption to positively charged -NH3 + groups in the PECs last layer, seems to us a natural choice. Furthermore, polyelectrolyte coverage significantly reduces CTAB cytotoxicity and increases AuNRs' stability in water [52,54,55].
Our UV-Vis data documented a 3 nm red shift (from 658 to 661 nm, Figure S3) in the longitudinal plasmon mode after PSS adsorption, which is consistent with the UV-Vis changes reported in reference [49]. Zeta potential measurements also support this, as a charge reversal is observed from + 29.5 ± 5 mV to −29.9 ± 2 mV.
The as-prepared AuNR-PSS was added to a known amount of (CaCO3-TMPyP)-PSS-PAH (Scheme 1, Section 2.2), and a change in the PEC's color from light red to dark blue was observed. The UV-Vis spectrum of the porphyrin-nanogold PEC hybrids (Figure 2A) in water presents the typical signatures of TMPyP and AuNR entities, confirming their successful encapsulation: a TMPyP Soret band around 432 nm, a broad band between 450-650 nm attributed to a superposition of TMPyP Q bands and the transverse plasmon mode, and a NIR band at 717 nm that originates from the longitudinal plasmon mode. It is important to note that neither the width nor the wavelength maxima of TMPyP in PEC hybrids are changed by the adsorption of polyelectrolytes or by the proximity of the AuNRs, which means that the porphyrin remains intact during the experiment. However, significant changes are observed in the AuNR's longitudinal mode peak wavelength with a red shift from 661 to 717 nm (56 nm), due to the combined effects of changes in the dielectric constants of AuNRs adsorbed and coupling of plasmon resonances from neighboring AuNRs which are closely adsorbed in the same microcapsule [56].
Calculation of the amount of AuNR-PSS adsorbed based on UV-Vis spectra of the supernatant cannot be readily applied in this instance, due to aggregation of unbound It is important to note that neither the width nor the wavelength maxima of TMPyP in PEC hybrids are changed by the adsorption of polyelectrolytes or by the proximity of the AuNRs, which means that the porphyrin remains intact during the experiment. However, significant changes are observed in the AuNR's longitudinal mode peak wavelength with a red shift from 661 to 717 nm (56 nm), due to the combined effects of changes in the dielectric constants of AuNRs adsorbed and coupling of plasmon resonances from neighboring AuNRs which are closely adsorbed in the same microcapsule [56].

Plasmonic Fluorescence Enhancement in Polyelectrolyte Microcapsules Hybrid
The antenna effect of the plasmonic interface of randomly assembled AuNRs on the porphyrins' emission was evaluated by FLIM. Images of more than 30 individual (CaCO3-TMPyP)-PSS-PAH-(PSS-AuNR) particles (5 × 5 µ m) were collected using laser excitation of λexc = 639 nm (0.44 kW/cm 2 ), which closely matches the longitudinal plasmon resonance of AuNRs, and the absorption from TMPyP Q bands ( Figures 3C,F and S5). The same type of image analysis was performed on two control samples prepared accordingly but containing only TMPyP or only AuNR, organized as (CaCO3-TMPyP)-PSS-PAH (Figure 3 A,D) and CaCO3-PSS-PAH-(PSS-AuNR) ( Figures 3B,E).   The average fluorescence enhancement for porphyrin-nanogold PEC's hybrid (160.9 counts) was experimentally determined after subtraction of the background signal of unenhanced porphyrin emission (average number of counts of control (CaCO 3 -TMPyP)-PSS-PAH 23.1 counts)- Figure 3G.
According to these data, we estimate an average enhancement factor of six times more fluorescence from TMPyP in porphyrin-nanogold PEC's hybrids due to their plasmonic outer layer of AuNRs. This is a remarkable result since it translates the overall picture of compact or sparse regions of AuNRs within a single capsule, and it also accounts for local heterogeneity between different PECs resulting from small differences in the surface distribution of AuNRs. The photoluminescence spectra obtained for porphyrin embedded PECs with and without plasmonic layer (green and red curves, respectively, Figure 3I), also reflect a significant increase in the number of photon counts/ms measured in the hybrid.
We have also used FLIM to probe maximum local fluorescence enhancement factors and compare our results with other works from the literature. Several fluorescence time traces were recorded in randomly chosen diffraction-limited spots of individual porphyrinnanogold PECs and controls (CaCO 3 -TMPyP)-PSS-PAH and CaCO 3 -PSS-PAH-(PSS-AuNR). Figure 4 shows various sections of different emission behavior extracted from typical fluorescence time traces acquired in our hybrids, that display strong and discrete fluorescence emission events of much higher intensities than the trace baseline. Sometimes these events are several consecutive bursts (B to D) while in other cases they are prolonged emission events (E to G). Note that strong emission is absent in the fluorescence time traces measured for controls. Therefore, the events observed in time traces B to G (Figure 4) are attributed to fluorescence enhancement of porphyrin emission. The average fluorescence enhancement for porphyrin-nanogold PEC's hybrid (160.9 counts) was experimentally determined after subtraction of the background signal of unenhanced porphyrin emission (average number of counts of control (CaCO3-TMPyP)-PSS-PAH 23.1 counts)- Figure 3G.
According to these data, we estimate an average enhancement factor of six times more fluorescence from TMPyP in porphyrin-nanogold PEC's hybrids due to their plasmonic outer layer of AuNRs. This is a remarkable result since it translates the overall picture of compact or sparse regions of AuNRs within a single capsule, and it also accounts for local heterogeneity between different PECs resulting from small differences in the surface distribution of AuNRs. The photoluminescence spectra obtained for porphyrin embedded PECs with and without plasmonic layer (green and red curves, respectively, Figure 3I), also reflect a significant increase in the number of photon counts/ms measured in the hybrid.
We have also used FLIM to probe maximum local fluorescence enhancement factors and compare our results with other works from the literature. Several fluorescence time traces were recorded in randomly chosen diffraction-limited spots of individual porphyrin-nanogold PECs and controls (CaCO3-TMPyP)-PSS-PAH and CaCO3-PSS-PAH-(PSS-AuNR). Figure 4 shows various sections of different emission behavior extracted from typical fluorescence time traces acquired in our hybrids, that display strong and discrete fluorescence emission events of much higher intensities than the trace baseline. Sometimes these events are several consecutive bursts (B to D) while in other cases they are prolonged emission events (E to G). Note that strong emission is absent in the fluorescence time traces measured for controls. Therefore, the events observed in time traces B to G (Figure 4) are attributed to fluorescence enhancement of porphyrin emission. While our measurements are not strictly limited to single-molecule fluorescence, because many porphyrins are present within the confocal volume, the discrete characteristic of these intense emission events suggests that these may arise from a single porphyrin molecule passing through interparticle gap hot-spots and, consequently, experiencing a strong enhancement effect on its emission. The stochastic nature of these events is reflected in the number, intensity, and duration of fluorescence bursts that vary from time-trace to time-trace. This result was expected because gold nanorods are randomly dispersed on the PEC's surface. Below we further discuss this issue based on DDA simulations. Additionally, different orientations and gap separations between neighbor gold nanorods may give rise to plasmonic hot spots with significant differences in the plasmon near-field enhancements that affect the strength of the antenna effect. Furthermore, the polyelectrolyte bilayer that separates the porphyrin (core) from the gold nanorods offers an ionic barrier that is likely to slow down porphyrin diffusion in the vicinity of gold nanorods, allowing TMPyP to probe local electromagnetic fields that translate into different fluorescence enhancements.
The top enhancement factor was calculated from the intensity of the strongest emission event (I ev ) corrected from the background signal I bg (intrinsic photoluminescence of individual porphyrins and AuNRs in the same capsule) and normalized to the intensity of one emitter in the absence of the antenna (I 0 ), according to Equation (1): For porphyrin-nanogold PEC's hybrid, the top fluorescence bursts obtained from 30 independent fluorescence time traces were found to range from 345,000 to 954,000 counts/s with an average I bg of 50,000 counts/s. Considering that, in the same experimental conditions, the non-enhanced emission intensity (I 0 ) of a single porphyrin is 0.8 counts/s [35], one can estimate that the top events measured range from 10 5 to 10 6 more photons detected during the events of enhanced emission than those obtained from TMPyP alone. Such large enhancement factors are in the same range as those reported earlier for the same porphyrin in gold nanodimer antennas made of gold nano-spheres linked through a DNA spacer [35]. Moreover, the maximum enhancement found here compares favorably with those reported in studies on 10 3 -fold emission increase for crystal violet in the vicinity of a single AuNR [37] or a 10 4 -fold increase in the emission of the same molecule in end-to-end assemblies of AuNRs [59].
To confirm the singularity of the (CaCO 3 -TMPyP)-PSS-PAH-(AuNR-PSS) design, microcapsules containing both porphyrin and AuNR now co-precipitated in the same CaCO 3 core and covered by a bi-layer of polyelectrolytes (CaCO 3 -TMPyP-AuNR)-PSS-PAH were also prepared. No evidence of plasmonic fluorescence enhancement was found, probably due to a low amount of AuNRs adsorbed ( Figures S6 and S7).

DDA Simulations
We have performed DDA simulations of a set of particle configurations idealized as models of coupled AuNRs in the plasmonic outer layer of PECs ( Figure 5A-H) to confirm that emission enhancements of several orders of magnitude are attainable for a porphyrin dye interacting with nanogap hot spots.
For each particle configuration, these maps were used to identify hot spot regions, and, therein, a position was defined for a point-like dipole emitter that simulates the emission of TMPyP (Figures S8-S10). Another set of simulations were performed, now in the spectral range of the dye's emission, which estimates the relative emission quantum yield, Φ/Φ 0 , i.e., the ratio between the modified quantum yields due to the particle's antenna effect and the dye's intrinsic one. In most configurations, the selected hot spot positions allowed for the enhancement of Φ/Φ 0 below 10-fold, which may seem modest, however, this factor is intrinsically limited to 20-fold ( Figure 5I). The excitation rate enhancement, E exc , with estimated values that can reach up to 1000-fold is then the prevailing factor for the overall emission enhancement, E exc × Φ/Φ 0 . The overall enhancement factors obtained from these model simulations reach a 10 4 -fold emission increase ( Figure 5J) for comparable conditions to those used in the experiments. The latter has shown that larger emission enhancements are experimentally observed. A similar difference between simulation and experiment has been observed before, for single porphyrin's emission enhanced by nanogap antennas [35], which deserves a more detailed investigation in future works. however, this factor is intrinsically limited to 20-fold ( Figure 5I). The excitation rate enhancement, Eexc, with estimated values that can reach up to 1000-fold is then the prevailing factor for the overall emission enhancement, Eexc × / 0 . The overall enhancement factors obtained from these model simulations reach a 10 4 -fold emission increase ( Figure 5J) for comparable conditions to those used in the experiments. The latter has shown that larger emission enhancements are experimentally observed. A similar difference between simulation and experiment has been observed before, for single porphyrin's emission enhanced by nanogap antennas [35], which deserves a more detailed investigation in future works.

Conclusions
In summary, herein we show that a compartmentalized structure of polyelectrolyte microcapsules can be used as efficient platforms to create spectrally engineered microsystems for plasmonic fluorescence enhancement of weak emitters. Particularly, the assembly of AuNRs strongly enhances porphyrin emission, resulting in an overall PECs fluorescence enhancement of six times and on fluorescence enhancements from discrete emission events, which are 10 5 to 10 6 -fold. To the best of our knowledge, this is the first example of PECs being used as a depot system to maintain and supply a defined level of weak emitters in the vicinity of gold nanoantennas. This result is an important step towards designing and improving devices targeting a wide range of fluorescence imaging applications.

Conclusions
In summary, herein we show that a compartmentalized structure of polyelectrolyte microcapsules can be used as efficient platforms to create spectrally engineered microsystems for plasmonic fluorescence enhancement of weak emitters. Particularly, the assembly of AuNRs strongly enhances porphyrin emission, resulting in an overall PECs fluorescence enhancement of six times and on fluorescence enhancements from discrete emission events, which are 10 5 to 10 6 -fold. To the best of our knowledge, this is the first example of PECs being used as a depot system to maintain and supply a defined level of weak emitters in the vicinity of gold nanoantennas. This result is an important step towards designing and improving devices targeting a wide range of fluorescence imaging applications.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nano12050872/s1, Table S1: Absorption (λ max A ) and emission (λ max F ) maxima wavelengths of TMPyP in water or PSS solution and TMPyP doped CaCO 3 microparticles (CaCO 3 -TMPyP), Figure S1: Molecular structures of PSS and PAH, Figure S2: SEM images of non-functionalized PECs, one probe polyelectrolyte microcapsules and Porphyrin-nanogold PEC's hybrid, Figure S3: Normalized extinction spectra of AuNR and respective PSS wrapped AuNR, Figure S4: Representative TEM images of TMPyP-nanogold PEC's hybrid, Figure S5: Examples of typical FLIM images measured for porphyrin-nanogold PEC's hybrid, Figure S6: Representative TEM and SEM images of alternative PEC's hybrid obtained by co-precipitation of both TMPyP and gold nanorods, Figure S7: Representative FLIM images obtained for TMPyP doped PECs, Figure S8: Model simulations of plasmonic hot-spots between AuNRs, Figure S9: Near-field enhancement maps calculated from model simulations, Figure S10: Calculated spectra of particle configurations shown in Figure S8 or Figure S9.