*Article* **Chiral Porphyrin Assemblies Investigated by a Modified Reflectance Anisotropy Spectroscopy Spectrometer**

**Ilaria Tomei 1, Beatrice Bonanni 1, Anna Sgarlata 1, Massimo Fanfoni 1, Roberto Martini 2, Ilaria Di Filippo 2, Gabriele Magna 2, Manuela Stefanelli 2, Donato Monti 3, Roberto Paolesse <sup>2</sup> and Claudio Goletti 1,\***


**Abstract:** Reflectance anisotropy spectroscopy (RAS) has been largely used to investigate organic compounds: Langmuir–Blodgett and Langmuir–Schaeffer layers, the organic molecular beam epitaxy growth in situ and in real time, thin and ultrathin organic films exposed to volatiles, in ultra-high vacuum (UHV), in controlled atmosphere and even in liquid. In all these cases, porphyrins and porphyrin-related compounds have often been used, taking advantage of the peculiar characteristics of RAS with respect to other techniques. The technical modification of a RAS spectrometer (CD-RAS: circular dichroism RAS) allows us to investigate the circular dichroism of samples instead of the normally studied linear dichroism: CD-RAS measures (in transmission mode) the anisotropy of the optical properties of a sample under right and left circularly polarized light. Although commercial spectrometers exist to measure the circular dichroism of substances, the "open structure" of this new spectrometer and its higher flexibility in design makes it possible to couple it with UHV systems or other experimental configurations. The importance of chirality in the development of organic materials (from solutions to the solid state, as thin layers deposited—in liquid or in vacuum—on transparent substrates) could open interesting possibilities to a development in the investigation of the chirality of organic and biological layers. In this manuscript, after the detailed explanation of the CD-RAS technique, some calibration tests with chiral porphyrin assemblies in solution or deposited in solid film are reported to demonstrate the quality of the results, comparing curves obtained with CD-RAS and a commercial spectrometer.

**Keywords:** circular dichroism; porphyrins; chirality; chiral layers; supramolecular chirality; reflectance anisotropy spectroscopy

#### **1. Introduction**

Reflectance anisotropy spectroscopy (RAS) is a surface-sensitive optical technique with significant peculiarities: it strongly reduces or avoids the contamination or even damage of the sample (when electrons or charged particles are used as probes in other experimental techniques), is utilizable in vacuum, in atmosphere or in transparent media (also in liquids) without limitations due to pressure, allows for the investigation of insulating samples without problems of charging, and gives the possibility to study the structure of surfaces/layers and even buried or immersed interfaces [1–3].

RAS has been applied to clean surfaces of metals and semiconductors in ultra-high vacuum (UHV), boosting a significant development of the knowledge in surface science [4–7], then to low-dimensionality solid state systems (films, wires, dots) [8,9], and finally to organic samples (for example, samples grown by organic molecular beam epitaxy (OMBE) [10,11] or porphyrin and sapphyrin layers deposited by Langmuir–Blodgett and Langmuir–Schaeffer technique) [12,13]. Clean surfaces of metals and semiconductors immersed in solutions

**Citation:** Tomei, I.; Bonanni, B.; Sgarlata, A.; Fanfoni, M.; Martini, R.; Di Filippo, I.; Magna, G.; Stefanelli, M.; Monti, D.; Paolesse, R.; et al. Chiral Porphyrin Assemblies Investigated by a Modified Reflectance Anisotropy Spectroscopy Spectrometer. *Molecules* **2023**, *28*, 3471. https://doi.org/10.3390/ molecules28083471

Academic Editors: Carlos J. P. Monteiro, M. Amparo F. Faustino and Carlos Serpa

Received: 8 March 2023 Revised: 6 April 2023 Accepted: 7 April 2023 Published: 14 April 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

and organic and biological layers deposited in liquid have been recently studied, down to the characterization of samples in electrolytes, monitoring, for example, in real time and in situ by RAS and scanning tunneling microscope (STM), the electrochemical reactions at the Cu(110)–liquid interface triggered by voltage cycles applied to the sample [14–16].

In all these experiments, RAS has been limited to the anisotropy of the linear dichroism of matter. As we will show hereinafter, the same technique can be used to investigate the circular dichroism in transmittance by a proper modification in the experimental apparatus, opening intriguing perspectives in the experimental study of chirality. This topic constitutes one of the most active and dynamic research areas related to varied disciplines such as chemistry, physics, biology, and material science [17].

Chirality is a characteristic that pervades the universe, from the small range of the subatomic particles to the immensity of the spiral galaxies: it is defined as the geometric property of a rigid object (or spatial arrangement of points or atoms) of being non-superposable on its mirror image [18]. This term derives from the Ancient Greek word "cheir" (χε*ι*´ρ) for hand, to give a pictorial sketch of the chirality meaning.

In chemistry, molecules that feature chirality can be spatially arranged into two specular, nonsuperimposable structures called enantiomers [19]. Since enantiomeric pairs are the same chemical species, the different spatial arrangement does not induce physical or chemical changes in the properties of these two isomers unless they are placed or interact with an asymmetrical environment. To have an idea of the importance of chirality, it is sufficient to note that all living systems are inherently chiral and represent an asymmetrical environment since, during evolution, only a single enantiomer has been selected for the synthons of essential biological macromolecules, such as L-aminoacids for proteins and Dsugars for nucleic acids. As a consequence, chirality plays a fundamental role in biological processes, driving the selectivity of most interactions essential for life.

Although conventionally considered important at the molecular level, especially in the pharmaceutical field, chirality is even more pursued at the supramolecular level to produce novel artificial chiral materials suitable for emerging applications such as circular polarizers, chiral chromatography, and (bio)sensor devices. Among the countless combinations of assembling methods and building blocks chosen, porphyrins have been intensively investigated since they offer the possibility to easily modulate molecular interactions, leading to stereospecific chiral systems by means of different approaches that use both chiral and achiral porphyrin units [20]. Additionally, the high absorptions in the ultraviolet (UV) and visible range possessed by these chromophores guarantee strong exciton coupling over large distances that result in diagnostic chiroptical signals unveiling the specific molecular orientation of the macrocycles within the aggregated species [21,22].

We are traditionally interested in characterizing the aggregation mechanisms of chiral metalloporphyrins with an appended proline group in hydroalcoholic solutions [23]. More recently, we have broadened this interest in the preparation of chemical sensors for chiral discrimination by using materials based on the same chiral porphyrins [24]. The enantiorecognition with chemical sensors is particularly challenging since these devices can rely only on a single binding event for chiral recognition. Additionally, controlling the deposition of chiral layers onto the surface of transducers constitutes another critical step since additional factors that may affect the film morphology/chiroptical properties, such as solvent, concentration, and evaporation time should be considered during the formation of sensing films [25].

Here, we investigated the possibility of utilizing a RAS spectrometer in transmission to measure features of porphyrins in solution and in solid films related to circular dichroism (CD), comparing the spectra obtained with the ones produced by commercial extended wavelength CD (ECD) apparatuses [26]. The possibility of using a RAS spectrometer with this new configuration for investigating CD (hereafter indicated as circular dichroism RAS, CD-RAS) could in principle represent just another way of performing experiments that are possible with the existing commercial instruments and then of questionable utility. However, unlike commercial CD spectrometers, the "open structure" of our CD-RAS spec-

trometer allows for higher flexibility in designing the optical path, making it possible to couple optics with UHV systems, with electrochemical cells, or more generally with other experimental configurations that otherwise should be impossible to configure. The increasing importance of experiments investigating the time domain of the behavior of chiral films (from evolution of the chiral signal after deposition to the investigation of kinetics when exposed to chiral molecules) suggests to go beyond the quite rigid methodology of application of the optical methods normally available in laboratory.

#### **2. Experimental Methods and Techniques**

The RAS signal is defined as the ratio ΔR/R between the difference ΔR of the light intensity R<sup>α</sup> and R<sup>β</sup> reflected by the sample for a beam polarized into two different and independent polarization states α and β, and their average R = (R<sup>α</sup> + Rβ)/2, as a function of photon energy [2,3]:

$$\frac{\Delta \mathcal{R}}{\mathcal{R}} = 2 \frac{\mathcal{R}\_{\alpha} - \mathcal{R}\_{\beta}}{\mathcal{R}\_{\alpha} + \mathcal{R}\_{\beta}} \tag{1}$$

In the case of linearly polarized light, the electric field is directed along two orthogonal directions α and β, usually aligned with the main anisotropy axes of the sample surface plane. In this version, a RAS spectrometer is essentially an ellipsometer at near normal incidence, whose experimental set-up strongly simplifies the interpretation of data with respect to traditional ellipsometry. A (sometimes complex) deconvolution from rough data is still necessary to correctly understand and explain the obtained results by modeling [2,3]: but the existence of a non-vanishing anisotropy often represents a meaningful result per se, suitable to characterize a certain physical system, or to follow its dependence upon the variation of definite experimental conditions as temperature, contamination, strain, coverage, pressure, electric field, etc., finally defined as the signature of the existence of a certain, well-defined phase at the surface. Linearly polarized RAS has been widely applied to investigate surfaces and interfaces, more generally 2D systems, exploiting the different symmetries of the bulk/substrate with respect to the surface/top layer [4–7,15]: the signal anisotropy coming from centrosymmetric crystals or amorphous substrates is ideally null, and then, any anisotropy signal measured by RAS is recognized and isolated as due to the surface/top layer [2,3]. In the experiments reported in Figure 1, α and β were parallel to the [211] and [110] directions on the (111) cleavage plane of a diamond sample 2 × 1 reconstructed. The evident peak dominating the spectrum is due to optical transitions between electronic surface states of the π-bonded reconstructed surface [27].

In a typical RAS system, light from a source (usually Xe or tungsten lamp, emitting photons in the near UV–visible–near infrared (IR) range, that is 280–1000 nm) is shined and focused into a polarizer, then onto a photoelastic modulator (PEM) and finally onto the sample (properly oriented), eventually passing through a special low-birefringence window if the sample is in ultra-high vacuum (UHV) or in liquid. The PEM (properly driven by an oscillating circuit at the resonance frequency ν<sup>0</sup> of the piezoelectric crystal) introduces a phase shift equal to ±π between light beams propagating along ordinary and extraordinary axes, thus modulating the linear polarization of light between two orthogonal, independent states *x* and *y*. The light intensity reflected by the sample for two polarizations (R*<sup>x</sup>* and R*y*) is then collected and focused onto a second polarizer (analyzer) and finally into a monochromator. At the exit slit, there is a detector (photomultiplier, photodiode, etc.) chained to a preamplifier and then to a lock-in amplifier to filter the signal, tuned at the correct modulation frequency (exactly 2ν0, that is the second harmonic of the modulated signal). Another version of a RAS spectrometer exists, where the analyzer is not present.

If the PEM introduces a phase shift equal to ±π/2, the outgoing light is alternatively polarized between right and left circularly polarized light, with a resulting signal carrying information about the circular dichroism of the shined sample. The high modulation frequency (about 50 kHz) favors the signal stability, eliminating the mechanical noise and the low frequency modes. In Figure 2, we draw a time-sketch of the successive polarization states of circularly polarized light after the passage through the PEM: the oscillation frequency between circularly left and circularly right polarization states is exactly coincident with the oscillation frequency of the PEM. In the signal, a linear polarization contribution also appears (just for one of the two orthogonal independent states, we will call it "α"), at a frequency that is twice the PEM frequency. Just tuning the lock-in at a frequency equal to ν0, the only contribution due to the circularly polarized light is selected.

**Figure 1.** ΔR/R as a function of photon energy for a single domain C(111)-2 × 1 surface, in the energy range from 0.4 to 2.8 eV. The sharp peak (whose sign means light electric field along the 1D chains of the reconstructed surface) is related to optical transitions between the surface electronic bands. The inset shows a low-energy electron diffraction (LEED) picture taken at 70 eV, demonstrating the existence on the reconstructed surface of a largely dominating single domain. Doubling of the spots is due to the regular array of surface steps. More extended experimental details are reported in Ref. [27]. Reproduced from G. Bussetti et al., Europhys. Lett. 2007, 79, 57002, with permission.

**Figure 2.** Sketch of the polarization of the light beam after the PEM, for an applied voltage value such that the phase shift is Δ*ϕ* = π/2. The red curve represents (vs. time) the voltage driving the PEM at the frequency ν0. At V = 0 volt, the polarization is linear, here directed as α (yellow arrows). When the voltage reaches its maximum, the resulting polarization is circular and right handed (green circles). When the voltage reaches its minimum, the resulting polarization is circular and left handed (blue circles). The oscillation frequency of the signal between the two states of circular polarization is evidently ν0. The frequency of the linearly polarized signal is 2 ν0.

The technical equipment for the CD-RAS experiments is sketched in Figure 3. After being transmitted through the sample, the beam is reflected on an oxidized Si(100), perfectly isotropic for linearly polarized light, not introducing any artifacts in the signal. Strictly speaking, the term "reflectance" in the acronym CD-RAS is inappropriate here, as the experiments are conducted in transmission: reflectance anisotropy for circularly polarized light is rigorously null. This can be demonstrated using the formalism of the Jones matrices [28]; however, an immediate and identical conclusion is reached by evaluating the effect on the

beam of the surface/layer (that is crossed twice). We have experimentally verified this null result for chiral films whose thicknesses were down to 10 nm.

**Figure 3.** Sketch of the CD-RAS experimental spectrometer. The version here reported is the Safarov/Berkovits-like (see Refs. [1–3,29]), with one polarizer. After passing the sample, the light beam is reflected back by an isotropic Si(100) sample. Although the system was mounted horizontally in this case, it can also work vertically.

The apparatus is highly efficient (as we will show in the next section) in extracting the CD-related features in transmission from solutions and thin solid-state layers deposited onto transparent substrates. For CD-RAS, more properly and more generally, we will then define the signal as:

$$\frac{\Delta \mathbf{I}}{\mathbf{I}} = 2 \frac{\mathbf{I}\_{\alpha} - \mathbf{I}\_{\beta}}{\mathbf{I}\_{\alpha} + \mathbf{I}\_{\beta}} \tag{2}$$

where I<sup>α</sup> and I<sup>β</sup> are the light intensities (in this case, transmitted) measured after passing through the sample for a beam alternatively polarized into two different and independent circular polarization states α and β. The "atout" of our CD-RAS (with respect to more traditional apparatuses commonly available, such as the JASCO CD spectrometer [26]) is a more flexible design of the optical path that can be tuned (in length), with a careful setting of the whole experimental setup, according to the peculiar necessity of the experiments, to be coupled with UHV systems, electrochemical cells, or transparent recipients larger that the cuvette used in commercial CD spectrometers.

The data obtained by CD-RAS were compared with the ones obtained by a commercial spectrometer [26]. While CD-RAS data represent the intensity modulation ΔI/I of the signal when passing from the right circularly polarized state to the left circularly polarized state (see Formula (2)), a commercial spectrometer measures the ellipticity due to the sample, which is a composition (generally associated with the elliptic polarization of light) of the two circularly polarized states, with the same intensity when radiation impinges on the sample, but is then differently absorbed. CD spectra are reported as ellipticity, θ, and are measured in units of mdeg. The ellipticity [θ] is defined as:

$$\mathbf{f}[\boldsymbol{\Theta}] = \mathbf{A} \left(\boldsymbol{\varepsilon}\_{\mathrm{L}} - \boldsymbol{\varepsilon}\_{\mathrm{R}}\right) \mathbf{c} \, l \tag{3}$$

where *ε*<sup>L</sup> and *ε*<sup>R</sup> are the molar extinction coefficients for light left- and right-handed polarized ones, respectively; *l* is the optical pathway of the sample, and c is its concentration. A is a constant with appropriate dimensions.

It is possible to demonstrate that the ellipticity [θ] and the ΔI/I result of a CD-RAS experiment are proportional [26]:

$$\left[\Theta\right] \sim \Delta \text{I/I} \tag{4}$$

#### **3. Results and Discussion**

In this section, the CD-RAS spectra measured in transmission mode for different samples are presented and compared with the spectra recorded by a commercial spectrometer for circular dichroism:


In cases (i) and (ii), the spectrometer used in CD-RAS experiments was in the Aspnes version, with two polarizers [30]. In case (iii), we used the Safarov–Berkovits version of the CD-RAS spectrometer (without the analyzer), sketched in Figure 3 [29]. In all the experiments, the beam after transmission through the sample was reflected on an oxidized Si(100), perfectly isotropic for linearly polarized light, and then in principle not introducing artifacts in the signal. The spectra were measured in the range 380 nm < λ < 500 nm (where the main molecular contribution was expected in all cases), although the covered possible range is more extended (300 nm < λ < 700 nm).

The CD-RAS spectrometer was first tested on chiral porphyrin aggregates characterized by intense CD-related signals in solution (samples A1 and A2). In Figure 4, the CD-RAS spectrum (red curve) and the JASCO spectrum (black curve) of Sample A1 are reported and compared. The different axes (with different units) are also drawn. The two curves are in excellent agreement. A similar result was obtained with sample A2 (Figure 5). In this case, with a different molecule and different CD activity, the curves exhibit an excellent overall similitude, with the same lineshape at a very good accuracy. The CD-RAS spectrum was here corrected by subtracting the background measured (in the same experimental conditions) with an achiral solution.

**Figure 4.** CD-RAS experimental spectrum for sample A1 (red curve, right vertical axis), compared with the ellipticity spectrum measured for the same sample by a commercial spectrometer for circular dichroism (black curve, left vertical axis). The CD-RAS spectrum is reported as measured, without correction. The double-arrow reported in the figure is the unit for the CD-RAS data. Here, sample A1 is a solution containing II-type aggregates of ZnP(L)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 5 μM).

**Figure 5.** CD-RAS experimental spectrum for sample A2 (red curve, right vertical axis), compared with the ellipticity spectrum measured for the same sample by a commercial spectrometer for circular dichroism (black curve, left vertical axis). The double-arrow reported in the figure is the unit for the CD-RAS data. The CD-RAS spectrum was corrected by subtracting the background measured (in the same experimental conditions) with an achiral solution. Here, sample A2 is a solution containing aggregates of H2P(L)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 10 μM).

The system was then tested on sample B to check the performance with a sample exhibiting three orders of magnitude lower CD-related signals in the millideg range. Given this low value of the CD, the CD-RAS spectrum reported in Figure 6 was obtained by a longer integration time (2 s/spectral point) to gain a higher signal-to-noise ratio, and from the resulting curve, the optical background of the glass cuvette containing the solution (being a quite larger signal due to the birefringence of the vessel) was subtracted. The good agreement of the two curves in this case allows us to demonstrate the sensitivity of CD-RAS at a higher sensitivity scale.

**Figure 6.** CD-RAS experimental spectrum for sample B (red curve), compared with the ellipticity spectrum measured for the same sample by a commercial spectrometer for circular dichroism (black curve, left vertical axis). The double-arrow reported in the figure is the unit for the CD-RAS data. The CD-RAS spectrum was corrected by subtracting the optical background of the glass cuvette containing an achiral solution (measured in the same experimental conditions). Here, sample B is a solution containing I-type aggregates of ZnP(D)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 5 μM).

Finally, chiral porphyrin films (samples C1 and C2) deposited onto a glass substrate were measured in transmission mode. The films resulting from letting one 50 μL droplet dry out were approximately 10 nm thick (estimated from atomic force microscope images). In Figure 7, the CD-RAS spectra (red and black curves in panel A) and the JASCO spectra (red and black curves in panel B) are shown for both enantiomers, deposited separately following the same protocol. The background spectrum measured in the same configuration for a clean glass substrate identical to the one used for deposition was always subtracted from the CD-RAS data, to eliminate the spurious anisotropy signal due to the experimental configuration. The final agreement is fully satisfactory, with an excellent signal-to-noise ratio in the CD-RAS curve (integration time: 4 s/spectral point). The broader structures in panel A are due to the larger bandpass of the monochromator used in the CD-RAS spectrometer.

**Figure 7.** Panel (**A**): CD-RAS experimental spectra for sample C1 (red curve: enantiomer 1) and sample C2 (black curve: enantiomer 2) (see text). The two enantiomers were obtained from a 50 μL droplet of a ZnOEP solution in dichloromethane deposited onto a glass substrate and measured in transmission mode. The resulting film has an average thickness estimated in the range of 10–15 nm from atomic force microscope images (not shown). Panel (**B**): Ellipticity spectra measured on the same samples by a commercial spectrometer for circular dichroism for sample C1 (red curve: enantiomer 1) and sample C2 (black curve: enantiomer 2).

#### **4. Materials and Methods**

Details about preparation, kinetics and theoretical explanation of samples A1, A2, and B1 are reported in a recent review about the stereospecific self-assembly processes of porphyrin–proline conjugates [23]. In detail, sample A1 contains II-type aggregates of ZnP(L)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 5 μM); sample A2 contains aggregates of H2P(L)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 10 μM); sample

B contains I-type aggregates of ZnP(D)Pro(–) in hydroalcoholic solution (EtOH:water 25:75, 5 μM). Solid samples were prepared as reported in Ref. [25]. In detail, 50 μL of a 0.1 mM solution in toluene of ZnP(D)Pro(–) or ZnP(L)Pro(–) was cast on a ultraflat glass slide to obtain samples C1 and C2, respectively.

The solutions and casted films were analyzed by a JASCO J-1500 Circular Dichroism Spectrophotometer, equipped with a thermostated cell holder set at 298 K and purged with ultra-pure nitrogen gas. The slit width was set to 2 nm and the scanning speed to 20 mdeg/min in continuous scanning mode. Linear dichroism contribution (LD) was found to be <0.0004 DOD units with respect to the baseline in all the cases examined.

AFM measurements were performed in air using a NanoSurf NaioAFM instrument. Experiments were carried out in tapping mode by using silicon tips with a spring constant of about 48 N/m and a curvature radius of less than 10 nm. The film thickness was estimated as the height differences between the molecule layer and the glass slide, made by excavating across the spot with a needle previously dipped in dichloromethane. The average height of the structures emerging from the glass is approximately 12 ± 3 nm, in a region of 30 × <sup>30</sup> <sup>μ</sup>m2.

#### **5. Conclusions**

In this article, we have shown that destructuring a spectrometer for circularly polarized light (CPL) by adapting a RAS system—normally used for linearly polarized light—to CPL, a flexible apparatus is now available to measure, with high sensitivity, the circular dichroism of solutions and very thin solid state films. As a consequence, a new class of experiments would then become accessible as an investigation in real time of the chiral thin film growth from solution drops deposited onto a transparent substrate (that on a vertical glass would undoubtedly glide toward the bottom, while the CD-RAS system can be mounted in a vertical plane, with the sample horizontal). The development of different approaches for the characterization of chiral layers is an important opportunity: the application of a modified RAS spectrometer for CD measurements could represent a significant boost for research in this field.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/molecules28083471/s1, (A): Optical spectra; (B) Rationalization of the relation between the commercial CD unit and the RAS-CD.

**Author Contributions:** Conceptualization and design of the study, C.G., R.P. and D.M.; methodology, A.S., M.F. and C.G.; synthesis of the title porphyrins, I.D.F.; preparation of the porphyrin samples and drop-casting on glass, I.D.F. and R.M.; Circular dichroism measurements, R.M. and I.D.F.; CD-RAS experiments, I.T., B.B. and C.G.; AFM experiments, R.M.; writing—original draft preparation, C.G. and R.P.; writing—review and editing, C.G., M.S., G.M. and R.P.; responsible for the financial support to the work, R.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by MUR PRIN, grant number 2017EKCS35\_002, by University of Rome Tor Vergata, Italy, project ASPIRE E84I20000220005.

**Data Availability Statement:** The data presented in this study and Supplementary Materials are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Sample Availability:** Samples of the compounds are available on request from the authors.

#### **References**

1. Gilp, J.F.M. Epioptics: Linear and non-linear optical spectroscopy of surfaces and interfaces. *J. Phys. Cond. Matt.* **1990**, *2*, 7985.

2. Weightman, P.; Martin, D.S.; Cole, R.J.; Farrell, T. Reflection anisotropy spectroscopy. *Rep. Progr. Phys.* **2005**, *68*, 1251. [CrossRef]


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

**Mateusz P ˛egier 1,\*, Krzysztof Kilian <sup>1</sup> and Krystyna Pyrzynska <sup>2</sup>**


**Abstract:** Searching for new compounds and synthetic routes for medical applications is a great challenge for modern chemistry. Porphyrins, natural macrocycles able to tightly bind metal ions, can serve as complexing and delivering agents in nuclear medicine diagnostic imaging utilizing radioactive nuclides of copper with particular emphasis on 64Cu. This nuclide can, due to multiple decay modes, serve also as a therapeutic agent. As the complexation reaction of porphyrins suffers from relatively poor kinetics, the aim of this study was to optimize the reaction of copper ions with various water-soluble porphyrins in terms of time and chemical conditions, that would meet pharmaceutical requirements and to develop a method that can be applied for various water-soluble porphyrins. In the first method, reactions were conducted in a presence of a reducing agent (ascorbic acid). Optimal conditions, in which the reaction time was 1 min, comprised borate buffer at pH 9 with a 10-fold excess of ascorbic acid over Cu2+. The second approach involved a microwaveassisted synthesis at 140 ◦C for 1–2 min. The proposed method with ascorbic acid was applied for radiolabeling of porphyrin with 64Cu. The complex was then subjected to a purification procedure and the final product was identified using high-performance liquid chromatography with radiometric detection.

**Keywords:** porphyrins; copper-64; positron emission tomography

#### **1. Introduction**

Porphyrins are natural macrocycles that are able to tightly complex metal ions. The application of porphyrins and their complexes with metal ions in non-invasive diagnostics is gaining increasing importance which allows early detection of disturbances or pathological changes caused by diseases [1,2]. Porphyrins are potent fluorophores, biologically compatible, and are known to preferentially accumulate in tumor tissue [1]. These compounds occurring in nature include the well-known, red-colored heme in hemoglobin, which is responsible for the transport of blood as well as the chlorophylls in some bacteria and plants which are utilized for photosynthesis [3]. Inspired by their role in nature, porphyrins and metalloporphyrins are used not only as agents for tumor diagnostics but also as photosensitizers in photodynamic therapy of cancer [4], contrast agents in magnetic resonance imaging [5], photodynamic antimicrobial chemotherapy [6], or enzyme models in bioinorganic chemistry [7].

Single-photon emission computed tomography (SPECT) and positron emission tomography (PET) are both highly sensitive nuclear imaging techniques, where small amounts of the radioactive tracer as prepared radiopharmaceutical are used to produce images of a given body part. SPECT uses gamma emitters injected into the bloodstream and subsequently taken up by certain tissues [8]. In healthy people, the distribution of the dossed radiopharmaceutical follows the physiological pattern, while in the case of some dysfunction, regions with increased or decreased content can be identified. In PET the short-lived radionuclides with the emission of positron are utilized [9]. Positron, after

**Citation:** P ˛egier, M.; Kilian, K.; Pyrzynska, K. Increasing Reaction Rates of Water-Soluble Porphyrins for 64Cu Radiopharmaceutical Labeling. *Molecules* **2023**, *28*, 2350. https://doi.org/10.3390/ molecules28052350

Academic Editors: M. Amparo F. Faustino, Carlos J. P. Monteiro and Carlos Serpa

Received: 5 January 2023 Revised: 28 February 2023 Accepted: 1 March 2023 Published: 3 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

losing kinetic energy, interacts with an electron in an annihilation reaction, resulting in the emission of two gamma photons at an angle of 180 degrees. The distribution of the radiotracer within different tissues, according to the carrier molecule, can be calculated based on the determination of the point of that annihilation.

Generally, the radiopharmaceutical used in SPECT or PET consists of four parts: radionuclide, bifunctional ligand, linker, and target molecule. The target molecules (small proteins, peptides, fragments of monoclonal antibody) are able to transport complex radionuclides to the diseased tissue containing the appropriate target receptor. The complexes used in nuclear medicine have to exhibit high thermodynamic stability as a strong interaction between the metal and the ligand is necessary to ensure the complete complexation of the radionuclide and to limit transmetalation reaction with competing species [10–12]. Moreover, a radioisotope–ligand complex should exhibit high kinetic inertness to prevent the dissociation of the complex and thus, the release of the radionuclide in the biological medium. The bifunctional ligands are covalently attached to the targeting molecule either directly or through various linkers [3]. The linker can be a simple hydrocarbon chain to increase lipophilicity, a peptide sequence to improve hydrophilicity, or a poly(ethyleneglycol) linker to slow extraction by hepatocytes. The selection of suitable chelators can facilitate the development of an effective PET imaging probe by improving targeting properties. Thus, there is a continuous need for new efficient chelators that can satisfy all the requirements.

The choice of radionuclide for diagnostic radiopharmaceuticals in PET and SPECT depends on its nuclear properties (type of radiation, half-life, energy) as well as radionuclide production, the conditions for radiolabeling, and specific activity. 64Cu, with its half-time of 78.4 h, low positron energy (653 keV), and a short average tissue penetration range (0.7 mm), has received attention for radiopharmaceutical development due to its favorable nuclear decay properties that make it useful in the labeling of antibodies for immuno-PET applications [13]. It undergoes multiple decay paths, allowing not only PET imaging through positron emission but also offering the possibility of treatment due to the emission of ΆLJ radiation (theranostics) [13,14]. DOTA (1,4,7,10-tetra- azacyclododecane-1,4,7,10-tetra-acetic acid) derivatives are mostly used to synthesize bifunctional ligands able to tightly bind the copper ions [15,16]. Porphyrins offer potential for application in nuclear medicine techniques, as 64Cu is a potent radionuclide that can serve as a diagnostic and therapeutic agent. With their natural affinity for tumor tissues, porphyrins can also act as targeting molecules for 64Cu, which can efficiently deliver it to pathological tissues. As potent complexing ligands able to tightly bind 64Cu ions, porphyrins can serve in radiopharmaceutical synthesis [2].

Porphyrins are known for forming very stable complexes with metal ions, which is advantageous in view of possible radiopharmaceutical applications [2,10]. Chelating properties are not significantly affected by the type and number of substituents in the porphyrin ring, which allows the choice of a ligand with appropriate characteristics, such as hydrophilicity/hydrophobicity or partition coefficient octanol/water, for the intended purpose of imaging. The main drawback of their coordination reaction is relatively poor kinetics at room temperature, thus heating under reflux conditions was required over an extended period of time [17]. The rate-limiting steps are determined by the kinetic barrier of the metal insertion step into the planar porphyrin structure and its relative rigidity.

To increase the rate of the formation of metalloporphyrins with medium-sized ions (e.g., Cu(II)) different approaches have been adopted. The rate-limiting step in the metalation of these ions is the deformation of the porphyrin ring. One of the interesting methods is the use of large ions (e.g., Hg(II), Pb(II), Cd(II)) that relatively easily form coordination complexes [18–20]. As these cations are too big to be incorporated completely into a porphyrin ring, they form a complex in which the ion "sits" on the top of the porphyrin macrocycle. It causes much faster distortion of the originally planar ring of the porphyrin ligand and allows relatively easy access for smaller ions (e.g., Cu(II)) from the other side. Yet, as far as the radiopharmaceutical application is concerned, the use of these highly toxic metal

ions during synthesis is unfavorable. Another way to implement this approach is to use a reducing agent for Cu(II) such as ascorbic acid, glutathione, or hydroxylamine [18,21].

Using a microwave synthesis unit, higher radiochemical yields can be achieved in shorter reaction time and small quantities of solvents employed [22–24]. Mechanochemistry and ultrasounds have been also recently proposed for the development of a new method for metalloporphyrin synthesis [25–27]. The synthesis of metal complexes under mechanochemical action requires the use of excess metal salt and the addition of NaOH solution to achieve moderate yields. Under ultrasound irradiation using water as a solvent, only an alkaline medium is necessary. The solventless mechanochemical synthesis of copper(II) complex of 5,10,15,20-tetrakis(N-methylpyridinium-4-yl)porphine tetraiodide (Cu-TMPyP) was obtained in 45% yield after 60 min of milling [26]. The synthesized metalloporphyrin required purification to remove the excess metal salt through exclusion chromatography using water as an eluent, followed by water evaporation. The Cu-TMPyP complex under 30 min ultrasound was obtained in 79% yield (determined by HPLC) and 53% isolated yield.

The aim of this work was the optimization of the synthesis of copper complexes with water-soluble porphyrins with cationic and anionic functional groups for potential application in radiopharmaceutical synthesis. Efforts were focused on developing the fastest possible method that would meet the requirements of labeling with short-lived copper isotopes used in PET. An optimized method was applied for the synthesis of the 64Cu–porphyrin complex.

#### **2. Results and Discussion**

Studied porphyrins included 5,10,15,20-tetrakis(N-methylpyridinium-4-yl)porphine tetratosylate (TMPyP), 5,10,15,20-tetrakis(4-sulfonato phenyl)porphyrin (TSPP) and for microwave synthesis also 5,10,15,20-tetrakis(4-carboxy phenyl)porphyrin (TCPP) (Figure 1).

**Figure 1.** Structures of studied porphyrins.

Electronic spectra of studied porphyrins are typical for this group of compounds [28,29]. The spectrum consists of the most intense band named the Soret band coming from the S0-S2 transition and four Q bands corresponding to split S0-S1 transitions. The formation of porphyrin complexes (structure presented in Figure 2) causes characteristic changes in electronic spectra. The number of Q bands decreases as symmetry break caused by core nitrogen protons no longer exists. The presence of metal ions also causes shifts of Soret and remaining Q bands. In the case of copper complexes, the shift in the Soret band is relatively small. Shifted Q(0,1) band is characteristic of the Cu complex of all studied porphyrins.

**Figure 2.** Structure of the copper(II) porphyrin complex on the example of Cu-TMPyP.

Spectra presented in Figure 3 are typical for unprotonated porphyrins (pH 9). They are almost identical to those at pH 7 (with a minimal decrease in absorbance). The change in spectrum in the acidic region refers to the protonation of porphyrin with average pKa~5, depending on functional groups attached to the porphyrin ring. In the case of TMPyP, protonation is not yet visible in the spectrum at pH 4. However, both TSPP and TCPP become protonated at pH 4 as the Soret band is shifted to a higher wavelength and the intensity order of Q bands is inverted. The intensity of all bands is generally higher than in neutral and alkali media. These spectral changes can be observed as a change in color to deep green and it makes spectrophotometry a convenient method for monitoring of porphyrin reaction course. Spectra of highly protonated porphyrins are presented in Figure S1. TCPP tends to precipitate from the solution.

Cu–porphyrin complexes were synthesized in a water environment at three different pH values: acidic pH 4 (acetate buffer), neutral pH 7 (phosphate buffer), and basic pH 9 (borate buffer) to find optimal conditions. Only in the case of TCPP at pH 4 porphyrin tended to precipitate from the solution, especially after heating, so these conditions were not included in studies. Application of a water environment is highly desirable for possible radiopharmaceutical synthesis, as organic solvents generally must be thoroughly separated from the product, which unnecessarily lengthens and complicates the purification step. Different pH values were tested as it has been proven that acid–base equilibria can strongly influence the complexation of porphyrins. The [Cu]:[Porphyrin] ratio for all syntheses was 1:1.

As porphyrins tend to form aggregates in solutions, a study had to be performed to define the appropriate concentration of reagents for spectrophotometric monitoring of complexation reactions. It was completed by recording spectra of porphyrin solutions at different concentrations and analysis of the dependence of absorbance vs. concentration. The deviations from linearity may suggest the formation of aggregates that could influence the identity of the final product, which is important in radiopharmaceutical synthesis and quality control. The high absorbance (well above 1 absorbance unit) of the Soret band made it unavailable to use the most intense porphyrin band, as linearity was quickly violated due to high absorption. For this reason, aggregation was studied by plotting the absorbance of the most intense band in the Q region of the porphyrin spectrum, namely Q1 for all except from TSPP at pH 4, where the Q1 band is virtually invisible and the Q4 band exhibits the highest absorption. Examples of fitted data are presented in Figure S2. Results show that in the studied concentration range, porphyrins show no signs of aggregation, as R2 for all the plots reaches at least 0.999. The only exception is TSPP at pH 4, where the upper limit is 1.5 × <sup>10</sup>−<sup>5</sup> mol L<sup>−</sup>1. Thus, 10−<sup>5</sup> mol L−<sup>1</sup> has been chosen as the optimal concentration for all further studies, as it is far enough from the upper limit and made it possible to study even small changes in the spectrum.

**Figure 3.** Spectra of studied porphyrins and corresponding Cu complexes: (**A**) TSPP, (**B**) TMPyP, (**C**) TCPP. [Cu-porphyrin] = [Porphyrin] = 10−<sup>5</sup> mol L<sup>−</sup>1; pH 9.

Acceleration of the synthesis of Cu(II)–porphyrin complex with the use of large ions can also be realized with the use of a reducing agent [19]. As far as the radiopharmaceutical application is concerned the choice of an acceptable reductant is important. In our study, ascorbic acid (AA) was employed (Figure 4). The reaction was conducted at a sufficiently high pH to maintain a deprotonated porphyrin ring. In the first rapid step, copper(II) is reduced to copper(I). As the ionic radius of Cu+ (96 pm) is larger than that of Cu2+ (72 pm), a complex with Cu<sup>+</sup> ions remaining out of the ligand plane is formed. This leads to the deformation of the porphyrin ring. Subsequent incorporation of Cu2+ ion involves more favorable kinetics leading to almost immediate reaction in optimal conditions.

**Figure 4.** Overview scheme of synthesis of Cu(II)–poprhyrin complex involving formation of complex with large ions generated by reduction of Cu2+ ion to Cu+ ion with ascorbic acid (AA).

Synthesis of Cu-TSPP complex in the presence of ascorbic acid (AA) as a reducing agent was running at the highest rate at pH 9 (Figure 5). Since the [AA]:[Cu] ratio equals to 5:1. maximum of complex absorbance is reached in 1 min. In the case of pH 7, a 20-fold excess of AA is needed to reach a maximum in 10 min or 50-fold (maximum is reached in 8 min). Reaction at pH 4 is the slowest and 20 min is not enough for complete synthesis of the Cu-TSPP complex. The reaction rate for TMPyP is generally higher than for TSPP with maximum absorbance reached at all pH values. Similarly to TSPP, at pH 9 the reaction is the fastest with an almost instant process, and maximum absorbance is reached between mixing reagents and recording first spectrum.

**Figure 5.** Complexation of Cu with porphyrins in a presence of AA at various [AA]:[Cu] ratios: (**A**) TSPP (pH 4); (**B**) TSPP (pH 7); (**C**) TSPP (pH 9); (**D**) TMPyP (pH 4); (**E**) TMPyP (pH 7); (**F**) TMPyP (pH 9). [Cu2+] = [TSPP] = [TMPyP] = 10−<sup>5</sup> mol L<sup>−</sup>1.

Contrary to TSPP, complexation at pH 4 is only slightly inferior to pH 9. From the 10-fold excess of AA upwards the reaction time is about 1 min. For both porphyrins, basic pH led to the deprotonation of pyrrolic nitrogen atoms, which besides the SAT mechanism additionally promoted a high reaction rate of complexation. Strong differences in the behavior of both porphyrins at pH 4 come from the stronger protonation of the porphyrin ring in the case of TSPP. TMPyP remains in a more reactive form with weakly bound protons that can easily be substituted by the incorporation of copper into the π system of porphyrin molecules. As for the other porphyrins, in the case of TCPP, which was the object of the previous study [19], the complexation at pH 9 is the fastest. An almost immediate reaction was achieved with a 10-fold excess of AA. For all studied porphyrins, a significant increase was observed, as the reaction time at room temperature without AA was more than 100 min (Figure S3).

The porphyrin ring is notably resistant to harsh conditions, such as high temperatures. For this reason, to increase the reaction rate microwave-assisted synthesis was applied. Microwave heating provides uniform energy transfer that prevents samples from overheating near reaction vessel walls and the possibility of precise control of reaction parameters. In order to maximize microwave energy, transfer continuous air cooling of the reaction vessel was applied. This allowed for a reduction of the reaction time to a minimum, which is vital for possible radiopharmaceutical synthesis. The obtained results are presented in Figure 6.

Microwave heating drastically reduced the time required for the synthesis of the Cu–porphyrin complex without the addition of any other chemicals, which is important in pharmaceutical preparations. Generally, 1 min was enough to reach the maximum absorbance of the corresponding complex. Only in the case of TMPyP after 2 min small increase was observed. Further elongation of synthesis did not cause any positive change. In the case of Cu-TCPP, a decrease in complex absorbance was noticeable, which shows the limited stability of the product. Compared to conventional heating, where in many cases 60 min was necessary to complete labeling [2], microwave-assisted synthesis showed a significant increase in reaction rate.

To assess the usefulness of the proposed complexation method, the reaction of TCPP with radioactive 64Cu (t1/2 = 12.4 h) was performed. As mentioned earlier, the complexation of macrocyclic ligands often requires harsh conditions that can be destructive for the targeting molecules due to irreversible modification of their structure. Thus, the rapid and stable complexation in mild conditions is a scientific challenge for the production of radiopharmaceuticals.

The 64Cu isotope was produced by proton irradiation of 64Ni in a medical cyclotron, then separated and finally dried in the form of 64CuCl2. Before further experiments, it was dissolved in deionized water. This form, without any purification, was used for labeling according to the procedure developed, achieving satisfactory yield with good radiochemical purity. Direct complexation resulted in a 78.7% yield at room temperature immediately. Further purification using solid phase extraction with an anion exchange column eliminated free 64Cu and increased the radiochemical purity up to 99.0%. It was confirmed by radio-HPLC where 64Cu was eluted in 2.3 min, while 64CuTCPP was eluted in 4.0 min referring to the radiometric detector signal (Figure S4).

Peaks are identified according to retention time (Figure 7) and specific spectra (Figure S5). This confirmed that the complexation of Cu(II) with the porphyrin ligand is fast and occurs within a single minute at pH = 9.

An interesting behavior was observed during purification on the anion exchange column. The complex could be eluted only in a specific sequence of eluents. After loading the column, water was used to remove water-soluble and cationic impurities (64Cu2+, buffers, and other metal ions). After that, gentle acidification (2 mL 0.1 mol L−<sup>1</sup> HCl) of the column was required to remove the complex quantitatively (>98.5%) with 0.5 mL of ethanol. Any other sequences of eluents, including concentrated acids (1 mol L−<sup>1</sup> HCl), acidification of the solution before the solid phase extraction step, and elution only with organic solvents, resulted in lower recovery or complex decomposition due to high concentration of acids. The explanation could be the mixed sorption mechanism; anion exchange for carboxylic groups combined with π–π interactions between the porphyrin ring and the solid phase

of the column. Acidification breaks the ionic interactions and aprotic solvent releases the complex.

**Figure 6.** Microwave-assisted synthesis of Cu complex with: (**A**) TMPyP, (**B**) TSPP, (**C**) TCPP. [Cu2+] = [TSPP] = [TMPyP] = [TCPP] = 10−<sup>5</sup> mol L<sup>−</sup>1.

**Figure 7.** Separation of (**A**) TCPP (retention time 2.2 min) and (**B**) Cu-TCPP complex (retention time 3.3 min). Conditions: Phenomenex Gemini C18 column (150 mm × 4.0 mm i.d., 10 μm), mobile phase: 40:60 acetonitrile:0.05 mol L−<sup>1</sup> CH3COOH/CH3COONa, pH 4.8, 1 mL min−<sup>1</sup> flow rate.

Another favorable phenomenon can be observed in the chromatographic separation of the post-reaction mixture. Other divalent metal ions may compete in the complexation process [30,31]. According to the supplier's declaration for 64Cu2+, the solution may contain Ni(II) in a two-fold excess and Zn(II) ions in an approx. 40-fold excess vs. copper-64. As porphyrin ligands are used in excess they are not competitive. However, due to formal requirements in the development of radiopharmaceuticals, it may be necessary to remove these metal ions. The proposed HPLC method also allows for the separation of impurities in the form of Ni-TCPP and ZnTCPP complexes [28].

#### **3. Reagents and Methods**

#### *3.1. Reagents*

5,10,15,20-tetrakis(4-carboxyphenyl)porphyrin (TCPP) and 5,10,15,20-tetrakis(N-methylpyridinium-4-yl)porphine tetratosylate salt (TMPyP) were obtained from Sigma-Aldrich. 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin tetraammonium salt was obtained from PorphyChem. Stock solutions (10−<sup>3</sup> mol L−1) were prepared by dissolution of the appropriate amount of porphyrin in 5 mL of specific solvent: 3.59 mg of TCPP in 0.02 mol L−<sup>1</sup> NaOH, 6.82 mg of TMPyP in 0.05 mol L−<sup>1</sup> HCl and 5.02 mg of TSPP in deionized water.

Amounts of 0.1 mol L−<sup>1</sup> borate buffer (pH 9) and phosphate buffer (pH 7) were prepared by dissolution of boric acid or sodium dihydrogen phosphate in water. Acetate buffer (pH 4) was prepared by dilution of 99.5% acetic acid with water. Further pH adjustment was performed using concentrated NaOH solution under control of Hanna HI 2210 pH meter (Hanna Instruments Inc., Smithfield, VA, USA) with combined glass electrode.

An amount of 1000 mg L−<sup>1</sup> standard of Cu(II) ions as nitrates in 2–5% nitric acid (Merck) was further diluted with deionized water.

#### *3.2. Study of Porphyrin Aggregation*

Series of 2.5 × <sup>10</sup>−6–2 × <sup>10</sup>−<sup>5</sup> mol L−<sup>1</sup> porphyrin solutions at different pH were prepared and spectra were recorded. Then, linear fitting of the data using least squares method was performed and R<sup>2</sup> factor was calculated to check the linearity of obtained results.

#### *3.3. Direct Synthesis of Cu–porphyrin Complexes*

Reactions were monitored with the Perkin Elmer Lambda 20 UV–VIS spectrophotometer (Perkin Elmer Inc., Waltham, MA, USA) with UV WinLab V2.85 software. Spectra were recorded in a range of 350–700 nm in polystyrene cuvettes of 1 cm length. To perform general synthesis at room temperature, 40 μL of stock porphyrin solution (10−<sup>3</sup> mol L−1) was mixed with pH buffer and then 25 μL of stock Cu solution (100 mg L−<sup>1</sup> in nitric acid) was added ([Cu] = [Porphyrin] = 10−<sup>5</sup> mol L−1). Final volume of the sample was 4 mL. Then, spectra were recorded for 4 h.

#### *3.4. Increasing Reaction Rate with the Use of Ascorbic Acid at Room Temperature*

The study was conducted at three pH values—4, 7, and 9 at room temperature. Final concentrations of copper and porphyrins were 10−<sup>5</sup> mol L−1. To study the influence of ascorbic acid (AA) on the complexation the concentration of AA was changed to achieve desired excess in relation to Cu. To carry out the study 40 μL of porphyrin was added to the proper buffer. Then, an appropriate volume of ascorbic acid and 25 μL of copper was added consecutively to reach desired excess of ascorbic acid in relation to Cu. Following excesses of AA were studied: 1, 2, 5, 10, 20, 50. For this reason, two stock solutions of ascorbic acid (AA) in acetate buffer were prepared by dissolution of solid ascorbic acid (Merck KGaA, Darmstadt, Germany): 0.05 mol L−<sup>1</sup> for 50 and 20-fold excess of AA and 0.001 < for 1, 2, 5 and 10-fold excess of AA. Final concentrations of Cu(II) and porphyrin were 10−<sup>5</sup> mol L−1. Reaction was monitored spectrophotometrically for 20 min. Due to low stability of AA, stock solutions wereprepared every day.

#### *3.5. Microwave-Assisted Synthesis of Cu–porphyrin Complexes (without Ascorbic Acid)*

Microwave-assisted synthesis was conducted without the addition of the AA due to its lack of thermal stability using CEM Discover SP microwave synthesizer (CEM Corporation, Matthews, NC, USA), which heated the sample with automatic real-time control of reaction parameters to prevent overheating and rapid increase in the pressure. Synthesis was performed in sealed 10 mL glass vessels with magnetic stirring. Temperature was set to 140 ◦C. The amounts of reagents were the same as for direct syntheses. In order to increase microwave power transfer continuous air cooling of the reaction vessel was applied. Reaction time was 1, 2, or 5 min. After cooling step sample was released and a spectrum was recorded.

#### *3.6. Radiosynthesis of 64CuTCPP Complexes*

64Cu was purchased from PET radioisotopes manufacturer (Voxel, Cracow, Poland) and was produced in 64Ni(p,n)64Cu reaction conducted on ALCEO solid target system (Comecer, Italy) coupled with GE PETrace 840 cyclotron (GE Healthcare Technologies Inc., Chicago, IL, USA). The requested activity was about 200 MBq in the form of dried 64CuCl2 reconstituted on-site with 1 mL of water. Stock porphyrin solution, buffers, and 0.05 mol L−<sup>1</sup> ascorbic acid (Merck KGaA, Darmstadt, Germany) were prepared as described above. Water and phosphate-buffered saline (PBS) were of pharmaceutical grade. 64CuTCPP was synthesized at room temperature by mixing 200 μL of 64CuCl2 solution, 400 μL of TCPP, and 200 μL AA stock solutions, filled up with a borate buffer up to 5 mL, and vortexed. For final purification and formulation solution was loaded on anion exchange Waters Sep-Pak Accell Plus QMA column (Waters Corporation, Milford, CT, USA) with an ISM833 Ismatec peristaltic pump (VWR, Radnor, PA, USA) at 1 mL min<sup>−</sup>1, flushed with 5 mL of water at 2 mL min−1, acidified with 1 mL of 0.1 mol L−<sup>1</sup> HCl and finally eluted with 0.5 mL of ethanol. After evaporation near dryness, small aliquot of PBS was added to reach expected activity concentration. An Atomlab 500 dose calibrator and wipe tester (Biodex, Shirley, NY, USA) was used for activity determination.

#### *3.7. Determination of Radiochemical Purity of 64CuTCPP Complexes*

Radiochemical purity was analyzed with the chromatography system Shimadzu AD20 with SPD-M20A UV–Vis photodiode array and radiometric (GabiStar, Raytest, Germany) detectors. The separation was performed on Phenomenex Gemini C18 column (150 mm × 4.0 mm i.d., 10 μm), with 40:60 acetonitrile: 0.05 M CH3COOH/CH3COONa adjusted to pH 4.8 as a mobile phase and 1 mL min−<sup>1</sup> flow rate.

#### **4. Conclusions**

Studied porphyrins form stable complexes with copper(II) ions. The application of both methods, one with the use of ascorbic acid and the second with microwave heating drastically reduced the time required for the synthesis of Cu(II)–porphyrin complexes.

In the case of the method with AA for all studied porphyrins, optimum conditions for fast and efficient complexation at room temperature comprise at least 10-fold excess of AA over Cu. Reactions with TSPP and TCPP should be conducted at pH 9, while for TMPyP pH 4 also offers a satisfying reaction rate. A method with microwave heating would be applicable only for thermally resistant molecules. If the porphyrin would act as a complexing agent attached to other, thermally sensitive targeting moiety, the method with the use of AA would be more convenient. Both proposed methods are convenient for application in radiopharmaceutical synthesis as they are one step, fast, and no toxic or harsh chemicals are used that would have to be separated. Additionally, they could be potentially adopted for the synthesis of radiopharmaceuticals labeled with short-lived copper isotopes ( 60Cu (t1/2 = 23.7 min), 61Cu (t1/2 = 3.3 h), and even 62Cu(t1/2 = 9.67 min)).

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/molecules28052350/s1, Figure S1. Spectra of protonated porphyrins. [TMPyP] = [TSPP] = [TCPP] = 10−<sup>5</sup> mol L<sup>−</sup>1; pH 1.5.; Figure S2: Study of aggregation of porphyrins at different pH and concentrations. Wavelength of peak maximum is given in parentheses.; Figure S3. Complexation of Cu with porphyrins with and without the addition of AA (pH and [AA]:[Cu] ratio in brackets): (A) TSPP (pH 4, 50:1); (B) TSPP (pH 7, 50:1); (C) TSPP (pH 9, 10:1); (D) TMPyP (pH 4, 10:1); (E) TMPyP (pH 7, 20:1); (F) TMPyP (pH 9, 10:1). [Cu2+] = [TSPP] = [TMPyP] = 10−<sup>5</sup> mol L<sup>−</sup>1.; Figure S4. Radio-chromatogram of raw product (A) and purified final product (B). Separation: Phenomenex Gemini C18 column (150 mm × 4.0 mm i.d., 10 μm), mobile phase: 40:60 acetonitrile: 0.05 mol L−<sup>1</sup> CH3COOH/CH3COONa pH 4.8, 1 mL min−<sup>1</sup> flow rate.; Figure S5. Spectra of (A) TCPP and (B) Cu-TCPP recorded at retention times: 2.2 and 3.3 min, respectively. Separation: Phenomenex Gemini C18 column (150 mm <sup>×</sup> 4.0 mm i.d., 10 <sup>μ</sup>m), mobile phase: 40:60 acetonitrile: 0.05 mol L−<sup>1</sup> CH3COOH/CH3COONa pH 4.8, 1 mL min−<sup>1</sup> flow rate.

**Author Contributions:** Conceptualization, K.P., K.K. and M.P; supervision: K.P.; data curation, K.K. and M.P.; investigation, K.K. and M.P.; visualization, M.P.; writing—original draft, K.P., K.K. and M.P; writing—review and editing, K.P., K.K. and M.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data generated during the study are available on request.

**Acknowledgments:** The authors thank the company Voxel for providing copper-64 used for the experiments.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Sample Availability:** Samples of the compounds TCPP, TSPP, TMPyP, and CuTCPP are available from the authors.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
