Zn(II)-tetracarboxy-phthalocyanine-Sensitized TiO2 Thin Films as Antimicrobial Agents under Visible Irradiation: a Combined DFT and Experimental Study.

In this article, we studied the antimicrobial activity of TiO2 sensitized by the Zn(II)-tetracarboxy-phthalocyanine (TcPcZn) complex using TiO2-Degussa P25 as a semiconductor source. The TiO2 thin films were deposited by the doctor blade method and were sensitized by the chemisorption process. The obtained compounds were characterized using Fourier transform infrared spectroscopy, UV-vis spectrophotometry, Raman spectroscopy, diffuse reflectance spectroscopy, and scanning electron microscopy. Furthermore, we studied the stability of the adsorbed sensitizer on the semiconductor surface by using the density functional theory (DFT). Additionally, we determined the antimicrobial activity of TcPcZn-TiO2 against methicillin-resistant Staphylococcus aureus (MRSA). The Raman and optical results confirmed the sensitizing process. The TcPcZn-TiO2 thin films showed radiation absorption in the visible range of the electromagnetic spectrum (600-750 nm), and the dye anchored on the TiO2 surface had a band gap of 1.58 eV. The DFT study showed that TcPcZn supported on any phase of Degussa P25 is stable, making them suitable to act as catalysts in the proposed reactions. Finally, the TcPcZn-TiO2 thin films reached 76.5% of inhibition activity against MRSA.

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) bacteria are a public health problem.[1,2] Reports estimate that in the first decade of this century, in United States alone, nearly 20,000 people died and more than 500,000 were hospitalized due to infections caused by these bacteria.[3,4] The infections caused by such bacteria through skin-to-skin contact are typical in hospital settings, primarily among healthcare workers and patients; therefore, the risk of infection and spreading among doctors, nurses, or other healthcare professionals and patients is continuous.[5]S. aureus bacteria are resistant to some conventional antibiotics (e.g., penicillin derivatives such as methicillin, cephalosporins, and vancomycin). During the last few decades, research on new developments in the field of antibiotics has been conducted to find alternatives for the control of resistant pathogenic microorganisms.[6−8] Various nanomaterials have antimicrobial properties that are not found in their micro-/macrocounterparts (silver nanoparticles are an example); however, technical challenges (e.g., agglomeration and loss of antibacterial activity over time) limit their commercial application.[9,10] Nowadays, advanced oxidation processes (AOPs) involving metal oxides have emerged as an alternative degradation technology both to eliminate recalcitrant compounds from wastewater and to inactivate bacteria and viruses in different types of environments.[11] Different semiconductors, derivatives, and mixtures thereof have shown their potential as antimicrobial agents, and the list includes SnO2, SnS2, Cu2O, ZnO, TiO2/graphene, and its derivatives. Table compiles information on this topic for different types of semiconductors.

Table 1

Reports on Antimicrobial Activities of Different Semiconductors

semiconductor/reference	microorganism	antimicrobial activity (best result)a
SnO2, SnS2, and SnO2/SnS2[12]	Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 6538)	reduction of 73% of cfu of E. coli and reduction of 70% of cfu of S. aureus
ZnO[13]	Escherichia coli and Staphylococcus aureus	25% of bacterial reduction (E. coli) and 50% of bacterial reduction (S. aureus)
TiO[2] thin films[14]	Escherichia coli	100% of lethality under UV light irradiation after 20 min
TiO2/ZnO[15]	Staphylococcus aureus, Pseudomonas fluorescens, Listeria monocytogenes, and Escherichia coliO157:H7	FIC (mg/mL)b, 0.33 against S. aureus and P. fluorescens, 0.40 against L. monocytogenes, and 0.25 against E. coli.
Cu2O/r-GOc[16]	Escherichia coli and Staphylococcus aureus	70% of bacterial reduction of E. coli and 65% of bacterial reduction of S. aureus
TiO2/Garcinia zeylanica extract[17]	MRSA was investigated under light irradiation	reduction of 99.1% of cfu under sunlight exposure
TiO2/5,10,15,20-tetrakis(2,6-difluorosulfonylophenyl)porphyrin[18]	Staphylococcus aureus was investigated under light irradiation	(>6 log killing) under visible irradiation
TiO2/Procion Red[19]	Staphylococcus aureus (ATCC25923) was investigated under light irradiation	99% under visible irradiation

Authors reported many tests, but we cited the best result from each reference.

FCI: fractional inhibitory concentration.

r-GO: stable reduced graphene oxide.

Among metal-oxide semiconductors, TiO2 has been recognized as the most promising photocatalyst owing to its unique electronic configuration, photostability, cost, and nontoxicity.[20] Different authors have stressed that TiO2 is useful in the degradation of various bacteria.[21,22] Matsunaga first published a study on the photocatalytic inactivation of microorganisms using TiO2 nanoparticles.[23] Azizi-Lalabadi et al. reported the antimicrobial activity of TiO2 nanoparticles supported in zeolite.[15] Ripolles-Avila et al. reported bactericidal activity against Gram-positive and Gram-negative bacteria by TiO2 nanoparticles under UV irradiation.[24] Kubacka et al. presented a complete study of TiO2 films as antimicrobial agents against a pathogenic bacterium.[25] Nevertheless, TiO2 is catalytically active only under UV irradiation, a fact that restricts its application under direct solar irradiation, as only 4% of solar spectrum radiation is located in the UV region. Nowadays, solar photocatalytic applications are part of green strategies (clean and cheap photochemical technologies) to improve wastewater and antimicrobial treatments; furthermore, the solar photocatalytic wastewater treatment shows great potential for practical applications.[26] In the last decade, different methodologies have been attempted to improve TiO2 photoactivity in the visible region (e.g., doping,[27−30] semiconductor coupling,[31−33] transition-metal coupling,[34−36] and sensitization with both synthetic[37,38] and natural dyes[39−42]). Among such strategies, the use of synthetic dyes as sensitizers is an efficient process for TiO2 modification. Sensitization improves the absorption of visible light, improving the photocatalytic activity under visible irradiation. The physicochemical process relies on electron transfer at the sensitizer/semiconductor interface—a similar process occurs in dye-sensitized solar cells.[43,44] Currently, N4 macrocycles (e.g., phthalocyanines and porphyrins) have demonstrated their advantages given their low cost and their special physical–chemical properties to convert visible light into chemical energy.[45,46] Phthalocyanines exhibit some attractive properties, namely: (i) high absorption coefficients in the visible region, (ii) high photostability, and (iii) strongly bonded to the TiO2 surface through carboxylate groups (when this group is present).[47,48] Therefore, phthalocyanines are promising candidates for antimicrobial application (e.g., heterogeneous photocatalysis and photodynamic therapy).[49−52] Phthalocyanines have proved their potential as photosensitizers of TiO2. Altin et al. reported the photocatalytic removal of methyl orange and under visible irradiation on zinc phthalocyanine-sensitized TiO2 nanocomposites.[53] Cobalt phthalocyanine–TiO2 nanocomposites were used in the photodegradation of organic textile dyes.[54] In this context, the surface functionalization of TiO2 has potential to be an effective photocatalytic agent on a wide range of microorganisms and chemical contaminants under visible light radiation. Currently, heterogeneous photocatalysis aims at the development of a solar photocatalytic treatment. In the present study, we carried out theoretical and experimental studies of Zn(II)-tetracarboxy-phthalocyanine (TcPcZn)-sensitized TiO2 thin films regarding their antimicrobial activity against MRSA under visible irradiation.

Results and Discussion

Structural Study

The TiO2 semiconductor has three crystalline structures: rutile, anatase, and brookite. Although the rutile-TiO2 phase is the most thermodynamically stable one, the anatase-TiO2 has higher photocatalytic activity. Both the anatase and rutile crystalline structures can be built as chains of TiO6 octahedral, the difference between both chemical structures is the distortion of the octahedra and the binding pattern of their chains along the crystal lattice; because of that, they present different physicochemical properties (e.g., recombination rates, band gaps, photocatalytic activity, and so forth).[55,56]

Figure shows experimental X-ray diffraction patterns for the thin films studied. The XRD-TiO2 pattern showed typical signals corresponding to anatase (JCPDS #071-1166) and rutile (JCPDS #021-1276) crystalline phases. The presence of both crystalline phases is due to the TiO2 source (Degussa P25).[57,58] The TiO2 powder (Degussa, P-25), a standard material in the field of photocatalytic reactions, contains anatase and rutile phases in a ratio of about 3:1; this composition is typical and has been reported by other authors with regard to its photocatalytic applications.[59] After the sensitizing process, the diffraction pattern did not change significantly. Furthermore, none of the films showed any evidence of the presence of other phases, which is a result aligned with other reports.[60] The thin-film-crystal-domain size was determined using the Debye–Scherrer equation and the full width at half maximum (fwhm) for the signal (101). The average crystal size was 21.5 nm for TiO2 and 20.8 nm for TcPcZn–TiO2 thin films.

Figure 1

X-ray diffraction patterns for the catalysts synthesized in this study, where (*) indicates signals for the anatase phase and (**) for the rutile crystalline phase.

Morphological Study

Figure displays the scanning electron microscopy (SEM) images obtained for TiO2 and TcPcZn–TiO2 thin films. Figure a shows formed microaggregates with a narrow size margin of 50-100 nm for the TiO2 films, which is a common result for the TiO2 Degussa P25 used in thin-film fabrication.[61]Figure b shows that the morphological properties of the semiconductor surface changed after the sensitizing process. Thus, the microaggregates of the TcPcZn–TiO2 film shows a larger grain size than the TiO2 films; some reports point out that the aggregation of TcPcZn on the TiO2 surface after the adsorption process is possible.[62] Moreover, in the first stages of the sensitizing process, the dye tends to occupy the spaces with greater surface energy on TiO2 (free spaces on the surface). Finally, this result is in agreement with other reports.[63,64]

Figure 2

SEM images of (a) TiO2 films and (b) TcPcZn–TiO2 films.

Raman Study

The Raman spectra of a typical dye (e.g., ruthenium complexes for DSSC applications) have been studied for characterization of their chemical structures and their adsorption on the TiO2 surface.[65−68]Figure a shows the Raman spectrum for TiO2 thin films; the signals located at 144 and 639 cm–1 are assigned to the E1g vibrational mode; the signal located at 518 cm–1 is assigned to the Ag vibrational mode; and the signal located at 398 cm–1 is assigned to the B1g vibrational mode—all these signals correspond to TiO2.[69]Figure b shows the free TcPcZn (solid state) Raman spectrum. The results show signals located at 1545, 1337, 751, and 641 cm–1, with these vibration modes corresponding to bonding Cβ–Cβ vibrations of the pyrrole, isoindole, and symmetric and asymmetric distortions of the macrocycle, respectively. Finally, Figure c shows the Raman spectrum of the TcPcZn–TiO2 thin films; all signals observed for TcPcZn are also observed here; besides, after TcPcZn adsorption on the TiO2 surface, Figure shows how the signal assigned to the Eg vibration mode increases in its intensity, and this could be associated with surface-enhanced Raman scattering (SERS). SERS is characterized by the increase in the intensity of the signals in the Raman spectrum by several orders of magnitude due to the effect of adsorbed species on the semiconductor surface.[70] The Raman results allowed observing that TcPcZn was adsorbed on the TiO2 thin films; this will also be supported with the obtained optical results.

Figure 3

Raman spectra for the catalysts fabricated in this study: free TiO2, free TcPcZn, and TcPcZn–TiO2.

Optical Study

Figure a shows the diffuse reflectance spectrum of the thin films. As can be observed, TiO2 did not show optical activity in the visible region, and this behavior is in concordance with the high-energy band gap value of TiO2.[71] In contrast, the TcPcZn–TiO2 thin films showed optical activity in the visible region (from 600 to 750 nm). Furthermore, Figure a shows two new signals (639 and 711 nm), which can be associated with (n → π*) transitions and are commonly reported for metal phthalocyanine compounds.[72] The band gap energy values for all photocatalytic systems (TcPcZn–TiO2) were determined for all samples using the Kubelka–Munk remission function,[73] and the analogue to Tauc plots (F(R∞)hv)1/2 against photon energy can be made according to[74]

Figure 4

(a) UV–vis diffuse reflectance spectra and (b) Kubelka–Munk plots and band gap energy estimation for both the TiO2 and TcPcZn–TiO2 thin films.

Figure b shows the plots of (F(R∞)hv)1/2versus (hv), which allows determining the band gap of films.[75]Figure b shows that TiO2 thin films have a band gap value of 3.06 eV, a value that is in concordance with previous reports (3.10 eV for Degussa P25).[76]

The dye anchored on the TiO2 surface (TcPcZn–TiO2 films) has a band gap value of 1.58 eV. As observed, after the sensitizing process, the optical properties of TcPcZn–TiO2 films in the visible region improved significantly, which also suggests that after the sensitizing process this type of film could work as a catalyst under visible irradiation.[77]

Antimicrobial Activity against MRSA

Figure shows the antimicrobial activity of both catalysts. Bare-TiO2 did not affect the inhibition of MRSA when visible light was used as an irradiation source, this result is in concordance with the previously reported TiO2. In general, no toxic effect was found without irradiation. The photodynamically induced death of bacteria was dependent on the type of material and irradiation light, this is according to the photodynamic process requirement.[18] The TcPcZn–TiO2 thin films reached 76.5% of inhibition activity against MRSA. The positive control (ciprofloxacin) reached 80.2% of inhibition activity.

Figure 5

Results of inhibition activity against MRSA for TiO2 and TcPcZn–TiO2 thin films. IC (irradiation control), −C (negative control), and +C (positive control).

These results are significant compared with other reports in the literature. Table lists reports of antimicrobial activities of TiO2 with different surface modifications. The comparison of these results against the TiO2 modified with a natural product (Garcinia zeylanicaextract), which has inherent antimicrobial activity, indicates that phthalocyanine is a potential sensitizer for this kind of application. In a recent study, Sułek et al. reported antimicrobial activity against Gram-positive S. aureus and Gram-negative Escherichia coli bacteria of the system TiO2/5,10,15,20-tetrakis(2,6-difluorosulfonylphenyl)porphyrin and its zinc derivative. They found that the photoinactivation of both systems can eradicate (>6 log killing) S. aureus, whereas no bactericidal effect was obtained against Gram-negative E. coli.[18] Other types of dyes have also been tested as sensitizers. Krishna et al. reported the use of two azo dyes (Mordant Orange and Procion Red) as sensitizers of TiO2 to inactivate S. aureus (ATCC25923) under visible irradiation.[19] Such antimicrobial activity can be explained by the following: (i) after visible absorption, phthalocyanine undergoes excitation, promoting an electron from the HOMO to the LUMO. This electron is injected into the conduction band of TiO2. Afterward, this electron can react to generate reactive oxygen species (ROS)[78−80]

After the electron transfer, ROS (c.a. O2–) are generated and the conventional photocatalytic process starts, with the consequent degradation of MRSA; (ii) after visible absorption, the excited electron located on the LUMO of the phthalocyanine could decay, generating hydroxyl or superoxide radicals (type I mechanism) or singlet oxygen (type II mechanism). In either case, the generated ROS are highly cytotoxic for MRSA(81−85)

Figure shows the HOMO–LUMO transition level of phthalocyanine and the energetic level of the semiconductor, indicating the possible ROS generation mechanism.

Figure 6

HOMO–LUMO transition level of phthalocyanine and the energetic level of the semiconductor. (a) Conventional ROS generation by electron transfer from the LUMO dye state to TiO2 (see eqs –5). (b) ROS generation of the dye through mechanism I (electron transfer) and through mechanism II (energy-transfer reaction) (see eqs and 7).

DFT Results

The TcPcZn macrocycle adsorbed on the semiconductor surface was modeled in the periodic-DFT framework using plane waves. The anatase and rutile structures were both studied and chosen as the crystalline TiO2-Degussa P25 (employed in the present experiments), constituting both phases in the films. Therefore, slab models of the anatase and rutile phases were constructed with (TiO2)24 units. The molecular structure of the sensitizer was first optimized in the gas phase and isolated. Subsequently, the molecule was mono-deprotonated (the hydrogen atom of one −COOH group was abstracted) and the −COO– group was placed close to the slab, facing the Ti atoms at a distance of 1.5 Å to allow the energy and structure optimizations of an adsorbed system through chemical interactions. The obtained results show the bidentate binuclear anchoring form of TcPcZn with anatase TiO2 nanoparticles, where the bond lengths between the carboxylic oxygens with Ti atoms are 2.411 and 2.015 Å (see Figure a). In the case of rutile, a monodentate mononuclear interaction between the sensitizer and the semiconductor was observed, with a bond length of 2.12 Å, as shown in Figure b. For both systems, the hydrogen that migrated to the semiconductor forms a bond with an oxygen of the exposed layer of the TiO2 model. On the other hand, the adsorption energies (Eads) were calculated using eq (see the Computational Methods section). The Eads for the adsorbed systems with the anatase structure is −62.18 kcal/mol, whereas using the rutile phase, the Eads is −50.01 kcal/mol, thus evidencing that considerable interaction and stable adsorption promote electronic coupling between the TcPcZn and the TiO2-nanostructured films and resulting in efficient sensitization. Furthermore, there was a difference regarding the optimized structures, where, in the case of the anatase phase, a full planarity of the macrocycle and the carboxylic acid that bond to the structure was observed. On the other hand, in the case of the rutile phase, an almost 90° rotation between the macrocycle and the carboxylic acid bond to the structure was observed. In this way, both observed facts (bidentate vs monodentate bonding and the torsion angle between the macrocycle and the carboxylic acid) support the lower adsorption energy observed for rutile in comparison with anatase.

Figure 7

Adsorbed TcPcZn/TiO2 systems on (a) anatase and (b) rutile phases. Color representation of the atoms: Ti atom (light blue), O atom (red), C atom (gray), H atom (white), N atom (dark blue), and Zn atom (purple).

Conclusions

We synthesized and characterized TcPcZn-–TiO2 thin films. The Raman results corroborated the sensitizing process. The TcPcZn-–TiO2 thin films showed radiation absorption at the visible range of the electromagnetic spectrum. The DFT results showed that TcPcZn supported on the phases present in Degussa P25, that is, anatase and rutile, is stable and catalytically active, where covalent interactions with the semiconductor were observed. Also, the results showed differences regarding the adsorption energies and structures in the TcPcZn/TiO2 models. The TcPcZn-–TiO2 thin films exhibited greater antimicrobial activity against MRSA, reaching 76.5% (±0.6) of inhibition after visible light irradiation. This behavior can be attributed to (i) ROS generation on the surface of the semiconductor and (ii) ROS generation by the sensitizer. Finally, our results indicated that TcPcZn is a suitable sensitizer for modified TiO2 catalysts to develop an antimicrobial action under visible irradiation.

Experimental Section

Synthesis and Characterization

All reagents were supplied by Merck, TiO2 (Degussa-Nanoshel). TiO2 thin films were deposited on soda-lime glass substrates using the doctor blade technique, and the suspension was placed on a 2 cm height by 2 cm width glass, according to previous reports.[86,87] The tetracarboxy phthalocyanines were synthesized using the method reported by Achar.[88] In summary, zinc sulfate (5.8 × 10–2 mol), trimellitic anhydride (1.8 × 10–1 mol), ammonium chloride (8.2 × 10–2 mol), and ammonium tetramolybdate (4.0 × 10–3 mol) were mixed in the solid state and, after that, the mixture was added to 10.0 mL of nitrobenzene. The mixture was heated to reflux for 4 h at 185 °C. Finally, the phthalocyanines were purified and recrystallized in an acid medium (HCl, 1.0 M). A dark-green solid was obtained. This powder was purified by means of column chromatography, using silica gel (2.5 × 24 cm) as a stationary phase and petroleum ether–ethyl acetate (15:3) as a mobile phase (rf.: 0.60); yield (35%). UV–vis (H2SO4) 230, 320, 630, 715 nm; FT-IR (cm–1): O–H (3292–3509), C=O (1703), C=C (1525), C=N (1282); MS (ESI-IT). The infrared spectrum was measured using an ECO-ART Alpha Bruker Fourier transform infrared spectrometer. The UV–vis assay was recorded on a Shimadzu UV-2401PC UV–vis spectrophotometer. In this assay, we dissolved 2.0 × 10–4 g of dye in H2SO4. For phthalocyanine sensitization of the TiO2 thin films, the previously prepared coatings were immersed in a solution of dye (1.1 × 10–1 M; pH = 5.0) and the chemical adsorption process was carried out for 24 h under constant agitation; after that, the sensitized films were washed with ethanolic solution and dried at room temperature.[42] The amount of dye adsorbed on the TiO2 surface was determined by the sensitizer desorption for using H2SO4 of 98% w/w (a medium in which the dye was entirely soluble). The concentration of phthalocyanines was calculated using spectrophotometry (calibration curve at 715 nm). We determined dye adsorption (20.2 × 10–6 mol dye/g TiO2) in more detail in the Supporting Information section. The physical–chemical properties of the films were studied by X-ray diffraction, diffuse reflectance spectrophotometry, and Raman spectroscopy. X-ray diffraction patterns were obtained with a Shimadzu 6000 diffractometer using Cu Kα radiation (λ = 0.15406 nm) as an X-ray source with a diffraction angle in the 2θ range (20–60°). Diffuse reflectance spectra were obtained with a Lambda 4 PerkinElmer spectrophotometer equipped with an integration sphere. The compositional properties of the materials were studied by Raman spectroscopy in a DXR device equipped with a 780 nm laser. The morphological properties were studied by SEM, under an excitation energy of 5 and 1 kV.[89]

Antimicrobial Activity against MRSA

We estimated viable bacteria in terms of colony-forming units (cfu)[90] based on our previous report.[91] In summary, (i) we used a nutrient broth (30.0 mL) to inoculate resistant Staphylococcus aureus strains (ATCC 43300). After that, the samples were stirred for 12 h at 37 °C. (ii) Then, we prepared a sample dilution until reaching 10–7 (see Figure ). Subsequently, (iii) the catalyst films were plunged into the crops; and visible irradiation (450 μW cm–2) was performed for 30 min. After crop irradiation, the samples were stirred for 12 h. (iv) Afterward, we determined antimicrobial activity by electronic counting of cfu in Petri dishes. Figure illustrates the procedure. Furthermore, several control tests were carried out to verify the inhibitory effect caused by the photodynamic activation of TcPcZn: (a) one test consisted of a sample of TiO2 thin films, with the strain and the nutrient broth exposed to visible light; (b) another test consisted of a sample of TcPcZn-–TiO2 thin films, with the strain and the nutrient broth exposed to visible light; (c) the irradiation control test consisted of a sample of the TcPcZn-–TiO2 thin film, with the strain and the nutrient broth not exposed to visible light (IC); (d) the negative control consisted of the strains and the nutrient broth (−C); and (e) the positive control (+C) consisted of the strain, the nutrient broth, and ciprofloxacin (40 mg/L). The irradiation tests were carried out in a batch reactor using an LED tape as a source of visible light radiation (cold white light, 17 W), and the incident photon flow per unit volume Io was 5.8 × 10–7 einstein L–1 s–1.

Figure 8

The cfu counting method for antimicrobial tests of thin films against MRSA. After that, tubes were irradiated under visible light for 30 min. Then, the tubes were incubated for 12 h. Adapted with permission from Rev. Acad. Colomb. Cienc. Ex. Fis. Nat.,[91] Copyright 2016, Rev. Acad. Colomb. Cienc. Ex. Fis. Nat.

Computational Methods

To evaluate the stability of the adsorbed sensitizer on the semiconductor, the following procedure was carried out. A TiO2 unit cell was modeled following the experimental parameters of a face-centered cubic (fcc) crystal structure for anatase and a simple tetragonal crystal structure for rutile. Computed lattice parameters were in good agreement with those experimental values for anatase (a = 5.365 Å)[92] and rutile (a = 4.593, c = 2.959).[93] The 3 × 2 and 2 × 2 slab models were built for anatase (101) and rutile (110), respectively, from their optimized crystals. Here, two layers were used to get a slab with (TiO2)24 units, as shown in Figure . As a first step, geometry optimizations of the semiconductors, of the TcPcZn macrocycle adsorbed onto TiO2 (anatase and rutile) and of the free TcPcZn were performed to obtain minima structures by periodic DFT calculations, which were modeled using the Perdew–Burke–Ernzerhof (PBE) functional and GBRV 1.5 pseudopotentials.[94] An energy cutoff of 45 Ryd was employed. Particularly, a 3 × 3 × 3 k-point mesh was used to optimize the TiO2 bulk, and the gamma point was employed to optimize the adsorbed complexes (TcPcZn/TiO2 slab models). All extended calculations were performed with Quantum Espresso v6.4.,[95] and the images were created with VMD v1.9.3.[96] A model consisting of a single molecule per supercell was considered. Therefore, an 8 Å vacuum in the Z-axis was included, avoiding some possible interactions between the adsorbate replicas.

Figure 9

Semiconductor slab models: (a) anatase and (b) rutile. (c) Molecular structure of free TcPcZn. Color representation of the atoms: Ti atom (light blue), O atom (red), C atom (gray), H atom (white), N atom (dark blue), and Zn atom (purple).

The starting complex geometry model consisted of TcPcZn being positioned directly onto the TiO2 slab. The carboxylic group (−COOH) of TcPcZn was deprotonated and the proton was adsorbed in a grain boundary oxygen of the slab. Then, the −COO– group was positioned to promote an interaction between the TcPcZn oxygens and a Ti atom of the slab, considering a starting adsorbate–surface distance of ∼1.5 Å.

Afterward, due to the supercell dimensions and their high computational cost, single point calculations were performed using the same functional employed for the periodic DFT computations (PBE), the Gaussian 6-31G(d,p) basis set for nonmetal atoms (C, H, O, and N), and the pseudopotential LANL2DZ for the metals (Zn and Ti). All discrete calculations were performed using Gaussian 09.[97] For a deeper understanding of the interaction between the molecule and the semiconductor, taking into account both the semiconductor structures, the adsorption energy (Eads) was calculated according to eq

In order to get the energy values for TcPcZn (Figure c), the molecular structure was also optimized under extended and discrete calculations, which were carried out under the aforementioned specifications.