2.1. Synthesis and characterization

This new family of dendrimers was prepared by the synthesis of the required dendrons (D1-PhCHO and D2-PhCHO), followed by their condensation with pyrrole to give the desired free-base porphyrins as intermediates, which were then metalated by Zn(II) to give the final dendrimers ZnTPP-D1 and ZnTPP-D2.

2.1.2. Porphyrin synthesis

Two synthetic methods are most often used to synthesize porphyrins: the Adler–Longo [33,34] or the Lindsey reaction [35]. Both are efficient for synthesizing porphyrins substituted at their meso positions. Given that the first-generation dendrimer TPP-D1 was previously isolated using the Adler–Longo approach [28], these reaction conditions were used again to synthesize TPP-D2. However, this compound could not be isolated in pure form even after several purification attempts by chromatography and subsequent recrystallizations (CHCl₃ and MeOH), the yield of crude TPP-D2 in the isolated solid being below 4%. Fortunately, all these porphyrin dendrimers have good solubilities in common organic solvents, allowing their easy metalation, and this approach provided a way to access the pure zinc complex from the mixture in the case of ZnTPP-D2 (Scheme 2). Thus, the corresponding zinc complex ZnTPP-D1 was formed at 40 °C overnight from TPP-D2 using Zn(OAc)₂ in CH˙2Cl˙2/MeOH and isolated pure in 60% yield (Scheme 2). Using similar conditions for metalation, ZnTPP-D1 was isolated in 80% yield from TPP-D1.

From a purely synthetic standpoint, the isolated yields in alkene-bridged dendrimers (D-type series; Figure 1c) were always lower than those of their alkyne-bridged analogs (T-type series; Figure 1b), although rigorously similar reaction conditions have been used (the isolated yields of TPP-T1 and TPP-T2 were 18% and 13%, respectively [16,17]), making the D-type series dendrimers more challenging to obtain via the Adler–Longo approach [16,17,36].

2.1.3. ¹H NMR analysis

The aldehyde dendrons D1-PhCHO and D2-PhCHO, and the corresponding dendrimers, i.e., the free-base and metalated porphyrins MTPP-D1 and MTPP-D2, were characterized by ¹H NMR analysis (Figures 2–4). Figure 2 shows the full ¹H NMR spectra of the dendrons compared to those of their analogs with triple bonds (T1-PhCHO and T2-PhCHO). They all show three diagnostic signatures: (i) the aldehyde proton as a singlet, around 10 ppm; (ii) the aromatic protons located as multiplets, in region 7–8 ppm, belonging to protons of phenyl and fluorenyl, partially identified; (iii) four groups of alkyl protons H_a,b,c,d located at 0.5–2.1 ppm that are assigned to the n-butyl chains of fluorenyl. We can particularly notice that for the double-bonded dendrons D1-PhCHO and D2-PhCHO, we observe four additional alkene protons comming out as a broad peak around 7.18–7.40 ppm.

The full ¹H NMR spectra in CDCl₃ of the corresponding free-base porphyrin TPP-D1 (Figure 3) shows four diagnostic signatures: (i) the β-pyrrolic protons of the porphyrin core (H_𝛽) around 9 ppm, (ii) the aromatic protons around 7.3–8.4 ppm, (iii) the alkyl protons of the various butyl chains around 2.2–0.5 ppm, and (iv) the NH protons of porphyrin cavity around −2.6 ppm. For TPP-D1, as for reported TPP-T1 and TPP-T2 [16,17], we observe eight protons H_𝛽 of porphyrin ring located around 9 ppm. For aromatic and vinylic protons (around 7.2–8.4 ppm), only some of them can be easily assigned like H_A and H_B. Again, the vinyl protons of TPP-D1 give rise to a broad peak around 7.2–7.3 ppm. For all these dendrimers, the n-butyl protons (H_a,b,c,d) are simply assigned to four groups of signals located at 0.5–2.1 ppm as for the corresponding dendrons. In contrast, for the larger dendrimer TPP-D2, a ¹H NMR spectrum with broad signals was obtained (see ESI) and only for the corresponding zinc(II) complex ZnTPP-D2 the spectrum was well resolved (Figure 4). Some characteristic signals are also readily identified; (i) eight protons H_𝛽 of porphyrin ring located around 9 ppm, (ii) H_A and H_B in the phenyl arms. However, most of the aromatic and vinylic protons overlap (7.2–8.4 ppm) because of the larger molecular structure.

2.2. Photophysical properties

UV-visible absorption and emission spectra, after excitation in the Soret band [37,38], were recorded for the isolated dendrimers at room temperature (Table 1 and Figures 5 and 6). The free-base tetraphenylporphyrin (TPP; Figure 1a) and the corresponding zinc complex (ZnTPP) were chosen as references compounds. Their two-photon oxygen-photosensitizing yields were subsequently evaluated and compared to those of their alkyne analogs (TPP-T1 and TPP-T2) to analyze the impact of this structural modification (Tables 2–3).

For the free-base porphyrin series, all absorption spectra are typical of porphyrin derivatives with an intense Soret band around 430 nm and four Q-bands in the 520–650 nm range [37]. Compared to TPP, the dendrimers TPP-D1, TPP-T1, and TPP-T2 present an additional absorption around 300–400 nm, corresponding to the conjugated fluorenyl dendrons. This band is almost of similar intensity for TPP-D1 and TPP-T1, most likely in reason of their identical terminal fluorenyl number; however, a red shift is observed for TPP-D1 compared to TPP-T1, likely due to the improved conjugation between core and arms in the former compound. Concerning the Soret band, the vinyl-bridged dendrimer TPP-D1 presents also a larger red shift (15 nm) than its alkyne analog TPP-T1 (9 nm) versus the Soret band of the reference TPP, which reflect the extension of the porphyrin π manifold [40] through conjugation with the peripheral arms at the meso positions.

The corresponding zinc complexes ZnTPP-D1 and ZnTPP-D2 exhibit characteristic changes in their electronic spectra compared to those of the corresponding free-base porphyrins TPP-D1 and TPP-D2 [37,41]. Only two Q-band absorption are now observed, around 552 nm and 635 nm, due to the symmetry change from D_2h to D_4h symmetry upon metalation and an intense Soret band around 435 nm (Table 1 and Figure 6). An additional broad band is also observed in UV range (346 nm) which corresponds to π–π* absorption of the fluorenyl chromophores. This UV absorption is weaker for the first-generation zinc(II) complex (ZnTPP-D1) than for the higher generation dendrimer (ZnTPP-D2) due to the smaller number (eight versus sixteen) of fluorenyl groups present in the peripheral arms.

Upon excitation in their Soret band, all these compounds exhibit the characteristic porphyrin emission peaks Q(0,0) and Q(0,1) [12,40]. After normalizing their emission spectra on their Q(0,0) peaks, these compounds exhibit two emission peaks at similar wavelengths, but with different intensities (Figure 5). Compared to TPP, the Q(0,1) band of all dendrimers does not change in intensity for TPP-D1 but increases for TPP-T1 and TPP-T2. The emission spectra of zinc(II) porphyrin complexes usually consist of three sub-bands assigned to a vibronic progression from a Q state: Q(0,0), Q(0,1), and weak Q(0,2), the last one, near 720 nm being usually too weak to be observed [11]. Presently, for ZnTPP-D1 and ZnTPP-D2, the emission spectra exhibit the two expected Q-bands around 603 nm and 653 nm (Figure 6), the blue shift of these bands compared to the corresponding free bases being ascribable to the metal coordination. The quantum yields (𝛷_F) were then measured for all these compounds (Table 1). While the free-base porphyrin dendrimers have similar quantum yields (𝛷_F = 12–13%) than TPP (𝛷_F = 11%), these values drop drastically (3–6%) when a metal like zinc(II) is introduced in the porphyrin cavity, as also observed for the reference ZnTPP.

The existence of an energy-transfer (ET) process from the peripheral 2-fluorenyl donors toward the central porphyrin acceptor core was subsequently studied. The emission spectra were measured from 450 to 800 nm, using two excitation wavelengths: the dendron absorption (325–351 nm) and the Soret-band absorption (419–434 nm). As expected, all dendrimers showed exclusive Q-band emission around 650–720 nm, in both cases, with no residual dendron emission (usually observed around 400 nm). This suggests that the peripheral fluorenyl groups transfer their energy very efficiently to the porphyrin core, given that any dendron emission is totally quenched (SI Figure S14). This very efficient energy transfer most likely corresponds to a so-called “through-bond” energy-transfer process (TBET) [42]. Thanks to this very efficient “antenna effect”, the dendron absorption band, when intense (as in ZnTPP-D2), might be efficiently used for exciting these compounds.

2.3. Two-photon absorption

As these dendrimers exhibit good fluorescence properties, their intrinsic two-photon absorption cross-sections were determined by two-photon excited fluorescence (TPEF) in CH˙2Cl˙2. Measurements were performed with 10⁻⁴ M solutions, using a mode-locked Ti:sapphire laser delivering femtosecond pulses, following the experimental protocol described by Xu and Webb [43]. A fully quadratic dependence of the fluorescence intensity on the excitation power was observed for each sample at all the wavelengths of the spectra (790–920 nm), indicating that the cross-sections determined are only due to TPA. A significant increase of their TPA cross-sections compared to that of TPP (12 GM at 790 nm) was observed for all porphyrins possessing fluorenyl dendrons (Table 2 and Figure 7). Comparison between the free-base dendrimers TPP-D1 and TPP-T1 reveals that replacing triple bonds with double bonds in the peripheral dendrons leads to a significant increase of the TPA cross-sections.

The zinc complexes of the D-type series (ZnTPP-D1 and ZnTPP-D2) also exhibit high TPA cross-sections (𝜎₂ = 260 and 450 GM, respectively) at 790 nm, ZnTPP-D2 being the best two-photon absorber of the series of compounds presently investigated. From the comparison between TPP-D1 and ZnTPP-D1, metalation by Zn(II) does not result in a significant change in cross-section at this wavelength and induces a slight decrease of 𝜎₂ at higher wavelengths compared to the corresponding free-base porphyrin. This statement suggests that the cross-section of the missing TPP-D2 free-base porphyrin would be similar or slightly above that of ZnTPP-D2 in the 790–920 nm range. In this respect, the clear enhancement of 𝜎₂ observed for ZnTPP-D2 relative to ZnTPP-D1 is particularly remarkable. First, it reveals a more pronounced structural difference in two-photon cross-sections for the D-type series than for the T-type series, in favor of the largest dendrimers. Then it suggests that the free-base dendrimer TPP-D2 would be a better two-photon absorber than its TPP-T2 analog in the investigated wavelength range.

The two-photon brightness (𝜎₂ ⋅𝛷_F) is a figure of merit allowing the evaluation of the potential of two-photon absorbers for fluorescence imaging. For the zinc complexes, any change in 𝜎₂ is combined with a strong decrease in 𝛷_F leading to a decrease of this figure of merit for ZnTPP-D1 relative to that of its free-base analog TPP-D1, which is the highest of the compound presently discussed. The ZnTPP-D2 dendrimer still exhibits an interesting two-photon brightness which is enhanced more than 20 times compared to TPP used as reference. Among free-base porphyrins, the two-photon brightness of TPP-D1 is also significantly higher than that of its analog TPP-T1, revealing the positive impact of replacing triple bond by double bonds in the peripheral dendrons for imaging purposes.

The oxygen-photosensitizing properties of these dendrimers were also studied. Their quantum yields of singlet oxygen generation (𝛷_𝛥) were determined and compared to those of analogous T dendrimers and TPP used as reference (Table 3). All these dendrimers exhibit values comprised between 0.55 and 0.64, comparable to that of reference TPP (0.60). Interestingly, the free-base dendrimer TPP-D1 shows the highest value 𝛷_𝛥 = 64%, whereas the two zinc complexes ZnTPP-D1 and ZnTPP-D2 show the lowest ones (59% and 55%, respectively). As previously noticed for T-type dendrimers [16,17,36], the increase in fluorescence quantum yield of the new free-base dendrimer TPP-D1 relative to TPP is not obtained at the expense of the singlet oxygen production.

In combination with the notable increase of the TPA cross-sections of the dendrimer TPP-D1 compared to its TPP-T1 analog, significant enhancements of the figure of merit for the two-photon excited oxygen sensitization (𝛷𝛥⋅𝜎2max=177 GM) can be achieved. For the zinc complexes, this value goes up to 153 GM for ZnTPP-D1 and even to 248 GM for larger ZnTPP-D2. The free-base TPP-D1 dendrimer exhibits a clearly higher enhancement factor than its T-type analog, in relation with its higher 𝜎₂ and its slightly increased 𝛷_𝛥. This compound, easy to synthesize, appears therefore particularly promising for two-photon photodynamic therapy, and, considering its TPEF properties, also for theranostic applications provided it can be made water-soluble by proper functionalization. It should be emphasized that other porphyrin-based systems with more efficient conjugation between the sub-chromophoric units have often been shown to exhibit higher TPA cross-sections, but these are generally accompanied by strong modifications of their other photophysical properties such as the red shift of their linear absorption range [24,44,45,46,47,48,49,50,51], which somewhat limitates their interest for theranostics. Indeed, most often these highly efficient two-photon absorbers exhibit a modest to negligible fluorescence or some interfering residual one-photon absorption above 800 nm, which leads to the loss of the 3D resolution. In contrast, the dendrimers presently reported, with a more restricted π-conjugation between the dendrons and the porphyrin core,¹ exhibit an improved trade-off [16,17,36,53] between intrinsic TPA, fluorescence, and photosensitizing properties.

3.1. General

Unless otherwise stated, all solvents used in reactions were distilled using common purification protocols [54], except DMF and ⁱPr₂NH, which were dried on molecular sieves (3 Å). All chromatographic separations were effected on silica gel (40–60 μ, 60 Å). ¹H and ¹³C NMR spectra were recorded on BRUKER Ascend 400 and 500 at 298 K. The chemical shifts are referenced to internal tetramethylsilane. High-resolution mass spectra were recorded on Bruker MicrOTOF-Q II in ESI positive mode in dried solvent at CRMPO in Rennes. Reagents were purchased from commercial suppliers and used as received. Element analyses were collected on a Microanalyser Flash EA1112. UV-visible absorption and photoluminescence spectroscopy measurements for all porphyrin dendrimers in solution were performed on Edinburg FLS920 Fluorimeter (Xe900) and BIO-TEK instrument UVIKON XL spectrometer at room temperature. Toluene and dichloromethane for spectral analysis were HPLC grade.

3.2. Dendron synthesis and characterization

The two dendrons D1-PhCHO and D2-PhCHO were obtained after a multistep synthesis from the brominated 1-bromo-3,5-xylene precursors and the corresponding 2-fluorenylaldehyde via Horner–Wadsworth–Emmons reactions followed by carbonylation using butyllithium and DMF.

1-bromo-3,5-bis(bromomethyl)benzene. Commercial 1-bromo-3,5-xylene (5.0 g, 3.67 mL, 27.02 mmol, 1 eq) was added into CH˙2Cl˙2(100 mL, distilled), together with NBS (9.6 g, 54.04 mmol, 2 eq) and AIBN (220 mg, 1.35 mmol, 0.05 eq). The mixture was stirred for 30 min at room temperature, and then refluxed for 30 h. Then cooled in ice-water bath and filtered, washing residue with heptane. The solvents were evaporated and the residue was further purified by chromatography (heptane), collecting the target product (4.56 g, 49% yield) admixed with 1-bromo-3-bromomethyl-5-methylbenzene (20%) as white powder, as well as pure 1-bromo-3-bromomethyl-5-methylbenzene (3.55 g) and 1-bromo-3-(bromomethyl)-5-(dibromomethyl)benzene (1.4 g). ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.47 (s, 2H), 7.34 (s, 1H), 4.41 (s, 4H).

Tetraethyl (5-bromo-1,3-phenylene)bis(methylene)diphosphonate (3). In a flask, the previously isolated [4:1] mixture of 1-bromo-3,5-bromomethyl-benzene and 1-bromo-3-bromomethyl-5-methylbenzene (2.4 g, 7 mmol, 1 eq) and P(OEt)₃ (2.4 mL, 14 mmol, 2 eq) were added, respectively. The mixture was refluxed for 4 h at 140 °C. The excess of P(OEt)₃ was removed under reduced pressure. Then the title product was purified by chromatography using CH˙2Cl˙2 to remove other byproducts and then collected by ethyl acetate, giving a colorless oil (2.66 g, 83% yield). ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.34 (s, 2H), 7.16 (s, 1H), 4.07–4.00 (m, 8H), 3.08 (d, J = 22.0 Hz, 4H), 1.26 (t, J = 7.0 Hz, 12 H).

Intermediate D1-PhBr. In a Schlenk tube, fluorenylaldehyde 1 (1.69 g, 5.51 mmol, 2.2 eq) and previously obtained 3 (1.15 g, 2.51 mmol, 1 eq) were added, then THF (100 mL, dried) was injected. After cooling the Schlenk with an ice-water bath (0 °C), t-BuOK (1.20 g, 10.69 mmol, 4.4 eq) was added under Argon and the reaction was kept stirring for 1 h at 0 °C. The bath was removed, a saturated NH˙4Cl solution (aq) added and the resulting solution extracted with ethyl acetate. After evaporating the solvents, it was further purified by chromatography (CH˙2Cl˙2/heptane [1:10]), giving D1-PhBr as a white powder (1.7 g, 89% yield). ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.71 (d, J = 7.6 Hz, 4H), 7.61 (s, 1H), 7.60 (s, 2H), 7.52 (d, J = 8.4 Hz, 2H), 7.50 (s, 2H), 7.37–7.31 (m, 6H), 7.27 (d, J = 16.0 Hz, 2H), 7.12 (d, J = 16.4 Hz, 2H), 2.01 (t, J = 8.0 Hz, 8H), 1.15–1.06 (m, 8 H), 0.71–0.58 (m, 20H).

Dendron D1-PhCHO. In a Schlenk tube, D1-PhBr (720 mg, 0.94 mmol, 1 eq) was dissolved in THF (60 mL) and n-BuLi (0.59 mL, 0.94 mmol, 1.6 M, 1 eq) was added dropwise at −78 °C during 15 min. The reaction was stirred for additional 40 min at low temperature. Then DMF (1 mL, dried) was added and stirring was continued for 1 h at −78 °C. The bath was removed, a saturated NH˙4Cl solution (aq) added and the resulting solution extracted with ethyl acetate. After evaporating the solvents, it was further purified by chromatography (CH˙2Cl˙2/heptane [1:5]), giving D1-PhCHO as a light-yellow powder (580 mg, 86% yield). ¹H NMR (400 MHz, CDCl₃, ppm): δ 10.11 (s, 1H), 7.96 (s, 3H), 7.73–7.71 (m, 4H), 7.56 (d, J = 8.0 Hz, 2H), 7.53 (s, 2H), 7.40–7.23 (m, 10H), 2.02 (t, J = 8.0 Hz, 8H), 1.15–1.06 (m, 8 H), 0.71–0.56 (m, 20H). HRMS-ESI: m/z calcd for C₅₃H₅₈O: 710.44822 [M]^+.; found 710.4481.

Intermediate D2-PhBr. This synthesis is a classical procedure similar to that previously used for D1-PhBr. The purification was completed by chromatography (heptane/CH˙2Cl˙2 [10:1]), providing D2-PhBr as a white powder (93% yield). ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.72–7.71 (m, 10H), 7.64 (s, 7H), 7.57 (d, J = 8.0 Hz, 4H), 7.55 (s, 4H), 7.37–7.31 (m, 16H), 7.30–7.18 (m, 8H), 2.03 (t, J = 8.0 Hz, 16H), 1.16–1.07 (m, 16 H), 0.72–0.57 (m, 40H).

Dendron D2-PhCHO. This synthesis is a classical procedure similar to that previously used for D1-PhCHO. The purification was completed by chromatography (heptane/CH˙2Cl˙2 [5:1]), providing D2-PhCHO as a yellow powder (65% yield). ¹H NMR (400 MHz, CD₂Cl₂, ppm): δ 10.15 (s, 1H), 8.05 (s, 3H), 7.76–7.73 (m, 13H), 7.62–7.59 (m, 8H), 7.42–7.29 (m, 25H), 2.06 (t, J = 8.0 Hz, 16H), 1.17–1.08 (m, 16 H), 0.72–0.54 (m, 40H). HRMS-ESI: m/z calcd for C₁₁₅H₁₂₂O: 1518.94902 [M]^+.; found 1518.9487.

3.3. Porphyrin synthesis and characterization

Reference porphyrins TPP, TPP-T1, and TPP-T2 were synthesized as described earlier by our group [32,40]. The generation G1 dendrimer TPP-D1 was obtained under Adler–Longo conditions as described earlier (8% yield) [28].

ZnTPP-D1. The free-base porphyrin TPP-D1 reacts with excess of Zn(OAc)₂ in [3:1] mixture of CH˙2Cl˙2 and MeOH at 40 °C overnight. After evaporating the solvents, ZnTPP-D1 was purified by chromatography (petroleum ether/CH˙2Cl˙2 [5:1]) and after evaporation of the volatiles was obtained as a pink powder (80% yield). ¹H NMR (400 MHz, CD₂Cl₂, ppm): δ 9.20 (s, 8H), 8.42 (s, 8H), 8.21 (s, 4H), 7.71–7.68 (m, 32H), 7.60–7.54 (m, 60H), 7.48 (d, J = 8.4 Hz, 8H), 7.42–7.27 (m, 80H), 7.15 (dd, J₁ = 8.4 Hz, J₂ = 2.4 Hz, 4H), 2.01 (t, J = 7.2 Hz, 64H), 1.12–1.01 (m, 64H), 0.65–0.51 (m, 160H) [28]. 13C NMR (125 MHz, CDCl₃, ppm): δ 151.3, 151.0, 143.0, 141.2, 140.8, 136.3, 136.1, 132.0, 130.7, 127.6, 127.1, 126.8, 125.8, 124.0, 122.8, 120.8, 119.9, 119.7, 54.9, 40.3, 25.9, 23.1, 13.8. HRMS MALDI: m/z calcd for C₂₂₈H₂₃₆N₄Zn: 3093.7876 [M]^+.; found 3093.782.

Dendrimer TPP-D2. The mixture of D2-PhCHO (250 mg, 0.16 mmol, 1 eq) and propionic acid (4 mL) was heated to 120 °C. After pyrrole (0.01 mL, 0.16 mmol, 1 eq) in propionic acid (1 mL) was added into the mixture dropwise, the reaction was kept refluxing for 5.5 h. After cooling to room temperature, MeOH was then added to the reaction mixture and the precipitate was filtered. The residue could be purified by chromatography (petroleum ether/CH˙2Cl˙2 [5:1]) as a red powder (10 mg, 4% yield). ¹H NMR (400 MHz, CD₂Cl₂, ppm): δ 9.10 (broad s, 8H), 8.50 (s, 8H), 8.20–7.10 (large signals, 188H), 2.01 (large s, 64H), 1.00 (m, 64H), 0.60–0.50 (m, 160H).

ZnTPP-D2. Previous crude mixture TPP-D2 (10 mg, 1.6 × 10⁻⁶ mol, 1 eq) reacts with excess of Zn(OAc)₂ (3 mg, 1.6 × 10⁻⁵ mol, 10 eq) in a [3:1] mixture of CH˙2Cl˙2 and MeOH (1 mL) at 40 °C overnight. After evaporating the solvents, the Zn complex ZnTPP-D2 could be isolated by chromatography (petroleum ether/CH˙2Cl˙2 [5:1]), as a pink powder (60% yield). ¹H NMR (400 MHz, CD₂Cl₂, ppm): δ 9.20 (s, 8H), 8.42 (s, 8H), 8.21 (s, 4H), 7.71–7.68 (m, 32H), 7.60–7.54 (m, 60H), 7.48 (d, J = 8.4 Hz, 8H), 7.42–7.27 (m, 80H), 7.15 (dd, J₁ = 8.4 Hz, J₂ = 2.4 Hz, 4H), 2.01 (t, J = 7.2 Hz, 64H), 1.12–1.01 (m, 64H), 0.65–0.51 (m, 160H).

3.4. Spectroscopic measurements

All measurements have been performed with freshly prepared air-equilibrated solutions at room temperature (298 K). Fluorescence measurements were performed on dilute solutions (ca. 10⁻⁶ M, optical density <0.1) contained in standard 1 cm quartz cuvettes. Fully corrected emission spectra were obtained, for each compound, under excitation at the wavelength of the absorption maximum, with A_𝜆ex < 0.1 to minimize internal absorption.

Measurements of singlet oxygen quantum yield (𝛷_𝛥). Measurements were performed on a Fluorolog-3 (Horiba Jobin Yvon), using a 450 W Xenon lamp. The emission at 1272 nm was detected using a liquid nitrogen-cooled Ge-detector model (EO-817L, North Coast Scientific Co). Singlet oxygen quantum yields 𝛷_𝛥 were determined in dichloromethane solutions, using tetraphenylporphyrin (TPP) in dichloromethane as reference solution (𝛷_𝛥 [TPP] = 0.60) and were estimated from ¹O₂ luminescence at 1272 nm.

Two-Photon Absorption Experiments. To span the 790–920 nm range, a Nd:YLF-pumped Ti:sapphire oscillator (Chameleon Ultra, Coherent) was used generating 140 fs pulses at a 80 MHz rate. The excitation power is controlled using neutral density filters of varying optical density mounted in a computer-controlled filter wheel. After five-fold expansion through two achromatic doublets, the laser beam is focused by a microscope objective (10×, NA 0.25, Olympus, Japan) into a standard 1 cm absorption cuvette containing the sample. The applied average laser power arriving at the sample is typically between 0.5 and 40 mW, leading to a time-averaged light flux in the focal volume on the order of 0.1–10 mW/mm². The fluorescence from the sample is collected in epifluorescence mode, through the microscope objective, and reflected by a dichroic mirror (Chroma Technology Corporation, USA; “blue” filter set: 675dcxru; “red” filter set: 780dxcrr). This makes it possible to avoid the inner filter effects related to the high dye concentrations used (10⁻⁴ M) by focusing the laser near the cuvette window. Residual excitation light is removed using a barrier filter (Chroma Technology; “blue”: e650–2p, “red”: e750sp–2p). The fluorescence is coupled into a 600 μm multimode fiber by an achromatic doublet. The fiber is connected to a compact CCD-based spectrometer (BTC112-E, B&WTek, USA), which measures the two-photon excited emission spectrum. The emission spectra are corrected for the wavelength dependence of the detection efficiency using correction factors established through the measurement of reference compounds having known fluorescence emission spectra. Briefly, the setup allows for the recording of corrected fluorescence emission spectra under multiphoton excitation at variable excitation power and wavelength. TPA cross-sections (𝜎₂) were determined from the two-photon excited fluorescence (TPEF) cross-sections (𝜎₂ ⋅𝛷_F) and the fluorescence emission quantum yield (𝛷_F). TPEF cross-sections of 10⁻⁴ M CH˙2Cl˙2 solutions were measured relative to fluorescein in 0.01 M aqueous NaOH using the well-established method described by Xu and Webb [39,43] and the appropriate solvent-related refractive index corrections [55]. The quadratic dependence of the fluorescence intensity on the excitation power was checked for each sample and all wavelengths.