Nanomaterials Based on Fe₃O₄ and Phthalocyanines Derived from Cashew Nut Shell Liquid

Abstract

In this work we report the synthesis of new hybrid nanomaterials in the core/shell/shell morphology, consisting of a magnetite core (Fe₃O₄) and two consecutive layers of oleic acid (OA) and phthalocyanine molecules, the latter derived from cashew nut shell liquid (CNSL). The synthesis of Fe₃O₄ nanoparticle was performed via co-precipitation procedure, followed by the nanoparticle coating with OA by hydrothermal method. The phthalocyanines anchorage on the Fe₃O₄/OA core/shell nanomaterial was performed by facile and effective sonication method. The as obtained Fe₃O₄/OA/phthalocyanine hybrids were investigated by Fourier transform infrared spectroscopy, X-ray diffraction, UV-visible spectroscopy, transmission electron microscopy (TEM), thermogravimetric analysis and magnetic measurements. TEM showed round-shaped nanomaterials with sizes in the range of 12–15 nm. Nanomaterials presented saturation magnetization (Ms) in the 1–16 emu/g and superparamagnetic behavior. Furthermore, it was observed that the thermal stability of the samples was directly affected by the insertion of different transition metals in the ring cavity of the phthalocyanine molecule.

1. Introduction

Nanomaterials containing Fe₃O₄ nanoparticles have received great attention due to their promising and multifunctional properties, besides potential medical applications in magnetic resonance imaging (MRI) and for cancer therapy [1], water treatment membranes [2], catalysis [3], magnetic separation [4] and detection DNA [5]. Previous studies [6] have shown the possibility of forming bifunctional therapeutic agents for use in photodynamic therapy (PDT) and magnetic hyperthermia. Specifically, the PDT aims at the treatment of several types of cancer, which combines a photosensitizer drug and light to generate active species of oxygen that fight the disease [7]. For this application, phthalocyanines commonly appear as excellent photosensitizers.

Phthalocyanines and their metal complexes have been considered versatile molecules, with remarkable optical and lasing properties, thermal and chemical stabilities [8], with maximum light absorption occurring in the visible region as well as being capable of producing singlet oxygen with high quantum yields [9]. Moreover, they have shown several extents of applicability in the fields of solar cells [10], liquid crystals [11] and as systems for fabrication of organic light-emitting diodes (OLED) [12]. Phthalocyanines consist of an aromatic macrocycle with an extensive conjugated π-electron system that can be readily modified by central coordination of metal atoms and peripheral substitution on the phthalocyanine ring, either by electron-donating or withdrawing groups [13,14].

The preparation of new eco-friendly fine chemicals derived from renewable and natural sources is of great interest. In this sense, the synthesis of cardanol-based phthalocyanines has been described in the literature [15,16,17]. Cardanol is the main component of technical cashew nut shell liquid (t-CNSL), obtained as a by-product during the industrial processing of cashew nut (Anacardium occidentale L.). t-CNSL is considered one of the richest natural sources of phenolic compounds; it is a renewable oil with low commercial value [18,19]. It consists of a mixture of long alkyl chain phenols (cardanol, cardol, anacardic acid and 2-methylcardol), with interesting properties and the possibility of chemical transformation for various applications [20,21,22,23].

Many studies related to hybrids have described the chemical interaction between phthalocyanine molecules and solid-state nanoparticles occurring through phthalocyanine’s peripheral functional groups that can anchor the particles’ surfaces [24,25,26,27]. Other reports have addressed a physical interaction, including alkyl chains in the phthalocyanine peripheral structure [28,29,30]. In the present study we report the successful synthesis of novel magnetic nanomaterials made by the combination of magnetite Fe₃O₄ nanoparticles, with the surface functionalized in sequence with oleic acid and cardanol-derived phthalocyanines; the latter passivation is favored by Van der Waals interactions between alkyl chains. We show also the influence of different metals coordinated to the center of phthalocyanines on the thermal and magnetic properties of these hybrid inorganic–organic nanomaterials.

2. Results

2.1. FT-IR Spectra

FT-IR spectroscopy is usually employed for the fast and efficient identification of chemical molecules and, in this view, of molecular ligands possibly functionalizing magnetite nanoparticles [31]. FTIR carried out on Fe₃O₄ sample shows characteristic bands of Fe–O at 580 cm⁻¹, as seen in Figure 1a. The presence of hydroxyl groups that reside on the Fe₃O₄ nanoparticles’ surface is attributed to the synthetic procedure conducted in aqueous media. The vibrational modes assigned to O–H stretching were observed at 3411cm⁻¹ and the band at 1637cm⁻¹ was attributed to angular vibration of O–H. As expected, the characteristic bands of Fe₃O₄ nanoparticles decreased after coating (Figure 1b–f). The spectrum of the Fe₃O₄@OA/Pc (Figure 1b) presents a band at 3292 cm⁻¹ attributed to the N–H stretching of phthalocyanine core [32]. The spectra of other hybrids, based on different metal phthalocyanines (Fe₃O₄@OA/CoPc, Fe₃O₄@OA/CuPc, Fe₃O₄@OA/NiPc and Fe₃O₄@OA/ZnPc) are virtually identical; bands between 2950 and 2850 cm⁻¹ were assigned to the C–H bond stretching from aliphatic groups, and the bands at 1610 cm⁻¹ to the aromatic C=C bonds stretching. The absence of the characteristic absorption bands at 3292 and 1014 cm⁻¹, assigned to N–H bond vibration, confirms that no free-base phthalocyanine is present.

2.2. X-Ray Powder Diffraction

XRD analysis was performed to study the crystalline structure of the Fe₃O₄ nanoparticles. As shown in Figure 2 nine diffraction peaks were viewed at 2θ = 21.5°, 35.2°, 41.5°, 50.6°, 63.1°, 67.4°, 74.5°, 88.9° and 110.5°, corresponding to (hkl) Miller indices of (111), (220), (311), (400), (422), (511), (440), (533) and (553), respectively. All peaks were consistent with the reference pattern of the Fe₃O₄ cubic inverse spinel structure (JCPDS card # N° 08-4611) with spatial group Fd3M. The size of the Fe₃O₄ was calculated using the Debye–Scherrer equation (1.0) based on the (311) peak.

$$D=\frac{K.\lambda }{\beta .cos \theta }$$

(1)

where D is crystal size (nm), K is a constant with value 0.9, λ is the wavelength of the X-ray radiation source (λ = 0.17889 nm, CoKα1), β is the measure of the width of maximum diffraction peak at the point where the intensity falls in half or “full width at half maximum (FWHM)”, and θ is the peak position [33]. The crystal size calculated for magnetite was of 11 nm, a value compatible with those already reported in the literature [34,35].

2.3. Thermal Analysis

Thermogravimetric results of the Fe₃O₄@OA/Pc, Fe₃O₄@OA/CoPc, Fe₃O₄@OA/CuPc, Fe₃O₄@OA/NiPc and Fe₃O₄@OA/ZnPc samples under a heating rate of 10 °C min⁻¹, and N₂ atmosphere, are shown in

Table 1. Degradation temperatures, mass losses and mass residues of pyrolysis decomposition at heating rates of 10 °C min^−1.

Nanomaterials	First Stage		Second Stage		Residue/%
	Range/°C	Mass Loss/%	Range/°C	Mass Loss/%
Fe3O4@OA/Pc	386–603	65.00	615–800	22.22	12.32
Fe3O4@OA/CoPc	292–673	35.39	675–641	12.88	48.71
Fe3O4@OA/CuPc	285–690	42.10	692–725	22.18	34.32
Fe3O4@OA/NiPc	255–673	34.71	675–707	12.53	47.95
Fe3O4@OA/ZnPc	167–740	47.74	741–784	12.07	38.83

and Figure 3. The decomposition of the materials occurred in two different stages and presented distinct residual masses in the range of 12–48%, which correspond to their metal oxides. All samples showed gradual weight loss of 34–65% in the temperature range of 167–740 °C. This event corresponds to the degradation of oleic acid and of peripheral chains attached to the macrocyclic moiety of the macromolecule. Decomposition and smaller weight losses of 13–22%, observed in the temperature range of 600–800 °C, correspond to the oxidative degradation of macrocyclic moiety of the phthalocyanines. Moreover, it was observed that there was a larger mass loss in Fe₃O₄@OA/Pc around 400 °C than in the other nanomaterials. This may be related to differences in the stacking arrangement of phthalocyanine molecules on the nanoparticle surface, which in the case of the free base forms aggregates, by π–π interaction, more efficiently than metallophthalocyanines when interacting with Fe₃O₄@OA, influencing the final mass of nanomaterial. In general, the complex aggregated with magnetite showed a initial degradation temperature lower than that of the non-metaled macromolecule. According to Barreto et. al. [34], this behavior is associated with the capacity of the nanoparticles to act as promoters of the degradation process. The order of observed thermal stabilities of metal complexes in inert atmosphere was: Fe₃O₄@OA/Pc > Fe₃O₄@OA/CuPc > Fe₃O₄@OA/CoPc > Fe₃O₄@OA/ZnPc > Fe₃O₄@OA/NiPc.

2.4. Transmission Electron Microscopy

The morphology and particle size of the samples were studied in more detail by TEM analysis, as shown in Figure 4. The TEM images showed that most of nanoparticles exhibited a roughly spherical shape and the spontaneous tendency to aggregate when deposited out from a solution onto a substrate. The Van der Waals and π-conjugated macrocyclic interactions of the phthalocyanines in Fe₃O₄ surface seem to favor this agglomeration. Moreover, it was possible to observe nanoparticle sizes with diameters of 12 ± 2 nm, 16 ± 3 nm, 13 ± 2 nm, 15 ± 4 nm and 13 ± 2 nm for Fe₃O₄@OA/Pc, Fe₃O₄@OA/CuPc, Fe₃O₄@OA/NiPc, Fe₃O₄@OA/ZnPc and Fe₃O₄@OA/CoPc, respectively. Fe₃O₄ isolated with a diameter of 10 ± 2 nm tended to agglomerate due to strong interparticle magneto-dipole interactions [35], but the coating with organic molecules noticeably decreased this behavior, as observed for the magnetic nanomaterials (Figure 4).

2.5. Magnetic Measurements

For verifying the influence of the different metals of the phthalocyanines on the magnetization of the nanomaterials, measures of the magnetization were performed as a function of magnetic field at room temperature using a vibrating sample magnetometer. The magnetization curves are presented in Figure 5. The magnetization curves of all nanomaterials exhibited a magnetic response to magnetic field variation and a characteristic superparamagnetic behavior, as evidenced in other previous works [17,36]. The values of saturation magnetization (Ms) found for Fe₃O₄@OA/Pc, Fe₃O₄@OA/CoPc, Fe₃O₄@OA/ZnPc, Fe₃O₄@OA/CuPc and Fe₃O₄@OA/NiPc were about 16, 11, 5, 3 and 1 emu/g, respectively. In general, the magnetization values obtained from the nanomaterials were significantly smaller than Fe₃O₄ nanoparticles (81 emu/g), in total agreement with similar examples reported in the literature [37]. At first such differences in magnetization values can be explained by the presence of non-magnetic components, namely oleic acid and phthalocyanines, which adhere to the nanoparticle surfaces, and by different relative Fe₃O₄ contents [36,37]. Thus, the reduction in saturation magnetization is basically due to the fact that these aggregated magnetite materials have a low magnetic response. Therefore, when phthalocyanine derivatives adhere to the nanoparticle surface, they contribute to the total sample mass, but do not contribute significantly with magnetic moments. Thus, like magnetization, the ratio of magnetic moments to mass (emu/g) showed a natural decreasing trend.

2.6. Optical Properties

The absorption and emission spectra of phthalocyanines in the ultraviolet-visible region are shown in Figure 6. Previous studies have shown that the presence of magnetic nanoparticles do not affect the absorbance and emission properties of the phthalocyanine derivatives [6]. Phthalocyanine exhibits typical electronic spectra with two absorption regions (Figure 6a), where the Q band is around 600–700 nm and the B band is near the UV region, around 300–450 nm. In metallo-phthalocyanines, Q bands consist of one sharp and intense band and two shoulders of low intensity, while in free base phthalocyanine these are of equal intensity. Figure 6b shows the maximum emission spectra of the phthalocyanines that occurs after excitation at 595 nm. The excitation spectra were similar to absorption spectra for Pc and ZnPc with similar wavelengths at 710 and 695 nm, respectively. This may suggest that the nuclear configurations of the ground and excited states are similar and not affected by excitation. This behavior is typical of phthalocyanine complexes [38]. The difference between the fluorescence intensities of the Pc and ZnPc was due to influence of axial metal coordinated to the center of macrocycle. It is important to note that emission corresponding in this spectral region was not observed for other compounds (NiPc, CoPc and CuPc), due to paramagnetic behavior of central metals in the phthalocyanine cavity [39].

3. Materials and Methods

3.1. Synthesis and Metallation of Phthalocyanines from CNSL

Figure 7 shows the synthetic route of phthalocyanines obtained from hydrogenated cardanol; that is 3-n-pentadecylphenol (1). The compound (1) was preferred as starting material due to its homogeneous chemical composition [40]. The compounds 4-(3-pentadecylphenoxy)phthalonitrile (2) and tetrakis-(3-pentadecylphenoxy)phthalocyanine (3) were prepared according to procedures previously described in the literature [15,16,41] with some modifications. Metal-free phthalocyanine (3) was prepared by mixing 0.20 g of compound 2 and 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) (0.70 g, 70.0 µL) that was refluxed in butan-1-ol (10 mL) for 16 h. To obtain metallo-phthalocyanines 0.04 g of an acetate-based metal salt X(CH₃COO)₂2H₂O (X = Zn, Cu, Co and Ni) was added to the reaction batch. After cooling to room temperature, ethanol was added and the solid filtered. The product was purified by column chromatography (silica gel) using methylene chloride (CH₂Cl₂) as eluent. The solvent was evaporated and the product obtained as a blue solid (3). Metal-free phthalocyanine (Pc): Yield 46%. ¹H NMR (CDCl₃, ppm) broad signals: −3.65 (2H), 0.84 (t, J = 8.4 Hz, 12H), 1.16–1.38 (m, 96H), 1.76 (m, 8H), 2.76 (t, J = 7.8 Hz, 8H), 7.13–8.62 (m, 28H). FT-IR (KBr, cm⁻¹): 3292, 2922, 2852, 1606, 1583, 1523, 1467, 1249, 1140, 1093, 1012, 742. UV-vis (λmax/nm, CHCl₃): 607, 639, 670, 704. LC-MS (m/z): 1724 [M + H]^–. Cobalt phthalocyanine (CoPc): Yield 68%. ¹H NMR (CDCl₃, ppm): 0.84 (t, J = 6.9 Hz, 12H), 1.25 (s, 96H), 1.59 (m, 8H), 2.62 (t, J = 7.7 Hz, 8H), 7.70–6.83 (28H). FT-IR (KBr, cm⁻¹): 2922, 2852, 1608, 1583, 1468, 1248, 1142, 960, 750. UV-vis (λmax/nm, CHCl₃): 612, 675. Zinc phthalocyanine (ZnPc): Yield 55%. ¹H NMR (CDCl₃, ppm) broad signals: 0.94–0.86 (m, 12H), 1.65–1.08 (m, 104H), 2.62 (t, J = 7.8 Hz, 8H), 6.81–7.73 (28H). FT-IR (KBr, cm⁻¹): 2922, 2852, 1606, 1581, 1485, 1246, 1142, 948, 744. UV-vis (λmax/nm, CHCl₃): 616, 684. Nickel phthalocyanine (NiPc): Yield 63%. ¹H NMR (CDCl₃, ppm) broad signals: 0.84 (t, J = 6.5 Hz, 12H), 1.17–1.27 (m, 96H), 1.74 (m, 8H), 2.72 (d, J = 7.5 Hz, 8H), 7.08–7.76 (m, 28H). FT-IR (KBr, cm^-1): 2923, 2852, 1608, 1583, 1473, 1250, 1143, 961, 751. UV-vis (λmax/nm, CHCl₃): 612, 676. Copper phthalocyanine (CuPc): Yield 72%. ¹H NMR (CDCl₃, ppm) broad signals: 0.87 (d, J = 6.7 Hz, 12H), 1.26 (s, 96H), 1.61 (m, 8H), 2.63 (t, J = 8.6 Hz, 8H), 6.88–7.72 (28H). FT-IR (KBr, cm⁻¹): 2921, 2852, 1607, 1582, 1467, 1250, 1143, 957, 746. UV-vis (λmax/nm, CHCl₃): 617, 684.

¹³C NMR (300 MHz, CDCl₃), ¹H NMR (300 MHz, CDCl₃) and Mass Spectra (MALDI-TOF-MS) of the compounds (a) Pc, (b) CoPc, (c) CuPc, (d) NiPc and (e) ZnPc have been reported in the Supplementary Materials section respectively in Figure S1, Figure S2 and Figure S3.

3.2. Synthesis of Magnetite

To form the spinel phase Fe₃O₄ the co-precipitation processing route was followed [34,35]. Thus, a solution of metal salts containing Fe²⁺/Fe³⁺ in the molar ratio of 1:2 was dissolved in Milli-Q water and heated to 80 °C. Then, under to vigorous stirring, a 30 wt% NH₄OH solution was added until the mixture reached pH = 10. The as-obtained black precipitate was washed several times with Milli-Q water until the residual solution became neutral. Lastly, the magnetic nanoparticles were dried. The chemical reaction of Fe₃O₄ formation is written in Equation (2).

Fe²⁺ + 2Fe³⁺ + 8O⁻ → Fe₃O₄ + 4H₂O

(2)

3.3. Preparation of Magnetic Nanomaterials

The hybrid nanomaterial consisted of magnetic nanoparticles stabilized by an inner layer of oleic acid, and an outer layer of phthalocyanine. Initially, magnetite (125.0 mg) was subjected to mechanical stirring in 50 mL of Milli-Q water, then few drops of NH₄OH (30%) were added until pH = 10 was reached, afterwards, 50 mL of oleic acid (OA) were further incorporated in the batch. Posteriorly, the ferrofluid obtained was separated from the organic phase by magnetic attraction. The formation of the second passivation layer was performed through the addition of a solution of phthalocyanine (12.0 mg), dichloromethane (40 mL) and ethanol (60 mL) under sonication at 40 ºC for 30 min, with subsequent addition of an excess of ferrofluid (Figure 8a). The magnetic hybrid (Fe₃O₄@OA/Pc) was separated by using a magnet (Figure 8b). This methodology was reproduced for the preparation of all nanomaterials described herein (Fe₃O₄@OA/CoPc, Fe₃O₄@OA/NiPc, Fe₃O₄@OA/CuPc and Fe₃O₄@OA/ZnPc).

3.4. Characterization

FTIR measurements of the samples were performed using a Perkin Elmer 2000 spectrophotometer in a 400–4000 cm⁻¹ range (KBr pellet). The phthalocyanines were analyzed by ¹H NMR spectra recorded on a Bruker Avance DRX-500 (500 MHz ¹H) using CDCl₃ as solvent and by liquid chromatography–mass spectrometry (LC-MS) on a Bruker Microflex LT (Maldi Tof) using alphacyano and dichloromethane as a matrix, and 5% TFA as mobile phase. Furthermore, UV–Vis absorption was recorded by a Varian Cary 5000 spectrophotometer (in CHCl₃). Photoluminescence excitation (PLE) measurements were obtained using a Horiba Nanolog fluorometer equipped with a liquid N₂-cooled InGaAs detector with the excitation range from 410 to 550 nm (coming from a Xe lamp) and emission range from 600 to 800 nm. The X-ray powder diffraction (XRPD) analysis was performed in an X-ray powder diffractometer Xpert Pro MPD (Panalytical) using Bragg–Brentano geometry in the range of 20–120° with a rate of 1° min⁻¹. CoKα radiation (λ = 1.7889 Å) was used and the tube operated at 40 kV and 30 mA, with the experimental patterns using the Rietveld algorithm to better identify and quantify crystallographic phases. Thermogravimetric analysis (TGA) experiments of the nanomaterials were all operated on a Mettler Toledo TGA/SDTA 851e equipment, under nitrogen atmospheres (constant flow of 50 cm³ min⁻¹), with a heating rate of 10 °C min^-1, sample mass of 10 mg, and temperature range from 25 to 800 °C. The magnetization measurements were performed at room temperature with a home-made vibrating sample magnetometer (VSM), calibrated using a pure Ni wire and, after measuring the mass of each sample, the magnetization was given in emu/g. The morphologies and microstructures of prepared samples were analyzed by a Jeol JEM-1011 transmission electron microscope (TEM) operating at 100 kV, equipped with a CCD camera ORIUS 831 from Gatan.

4. Conclusions

In summary, we demonstrated a facile strategy for the development of magnetic nanomaterials based on Fe₃O₄ and phthalocyanines derived from biomass. The results of TEM confirmed the nano-sized particles with spherical morphology. The decomposition temperatures of Fe₃O₄@OA/Pcs can reach 380 °C, indicating their excellent thermal stability. We also observed that, despite showing superparamagnetic behavior, the transition metals of metallo-phthalocyanines appear to negatively influence the magnitude of the nanomaterial magnetization. Thus, we showed that by Van der Waals interactions occurring between the phthalocyanines and oleic acid alkyl chains emerging from the surface of Fe₃O₄, it was possible to obtain stable and magnetic hybrid nanomaterials.