Layered Double Hydroxide-Based Composites for Concerted Decontamination of Water

Abstract

A series of composites was prepared starting from five types of LDHs, which were then exchanged with three types of metallo-phthalocyanines, and, in the end, magnetic nanoparticles were attached. In the case of LDHs containing Fe, characterization data showed that there was a partial oxidation from Fe²⁺ to Fe³⁺. Samples containing evident LDH structures performed better in general than the ones containing iron oxide mixtures, the composites being more active towards amoxicillin removal compared with ampicillin removal. The nature of the phthalocyanine did not have such a great influence, although some differences in the activity were observed. The removal was a combination between adsorption and photocatalytic degradation. The best composites for this application were those based on Mg_0.325Fe_0.325Al_0.25-LDH prepared by co-precipitation in the presence of NaOH and Na₂CO₃. They presented high adsorption capacity in 10 min and, at the same time, high photocatalytic activity for both amoxicillin and ampicillin.

1. Introduction

Active pharmaceutical compounds are important for public health and quality of life. However, like many things, we consume these compounds, but they are not completely absorbed or metabolized in organisms. Moreover, these compounds can reach the environment without modifications or being non-metabolized. Among the pharmaceutical compounds which can cause major damage to the environment, antibiotics have an important contribution due to their high consumption rate in both veterinary and human medicine [1,2]. Antibiotics act as important drugs to inhibit the growth of bacteria and eliminate or kill microbes. As such, antibiotics have been classified upon on their chemical structure, spectrum, mechanism of action, method of administration, and effect. However, classification based on mechanism of action is the most common [3]. These are used in present in large quantities and because of that they have become a focal point in the research regarding the resistance of pathogens to antibiotics and, in recent years, in environmental research [4].

Although they are used as a treatment for many human or animal diseases, antibiotics from the pharmaceutical industry, animals, poultry, households, hospitals, and aquaculture or livestock farms represent a challenge [5]. Hence, efforts towards removing antibiotics from water were made, although several limitations (incomplete removal, lack of sustainability, cost, and production of toxic side products) remain. Compared to the general water treatment techniques, photocatalytic degradation of antibiotics is still an efficient method used to remove them from water systems [6]. However, the use of “classical” photocatalysts such as TiO₂ has limitations, mostly due to their large band-gap (3.2 eV), which makes them efficient only under UV irradiation.

The catalytic systems which involve supramolecular chemistry concepts use chromophores as photosensitizers. For example, metallo-phthalocyanines (MPcs) are electron donors that play the role of a substrate that can absorb the radiation in the visible range. Phthalocyanines, with planar structure, can incorporate either two hydrogen atoms to form metal-free phthalocyanine (H₂Pc) or a transition metal atom in the center of their circular structure to form metal-organic complexes [7]. Complexes or phthalocyanines containing metal cations with complete electron shells or d-orbitals such as Zn²⁺, Mg²⁺, and A1³⁺ [8] present catalytic activity. Layered double hydroxides (LDHs), which belong to the anionic clays [9], can act with good results in the degradation of these drugs, showing the general formula [M²⁺_1−xM³⁺ _x(OH)₂]^x+Aⁿ⁻_x/n·mH₂O, where M²⁺ and M³⁺ are, respectively, divalent and trivalent metallic cations (e.g., Mg²⁺ and Al³⁺), A^n- is an anion, x is the mole ratio (varies between 0.2 and 0.33), and m is the number of molecules of water of crystallization within the inter lamella. LDHs can be prepared in the laboratory in an easy and timely way [10,11]. Positively charged octahedral sheets are obtained when M²⁺ cations from brucite, Mg(OH)₂, are replaced by M³⁺ cations [12]. Further, charge-compensating anions (Aⁿ⁻) are incorporated into the interlayer in order to maintain charge balance [13]. LDHs can precipitate heavy metal cations in aqueous solutions (acid buffer role) [14] as a result of their alkaline nature. There are several categories of LDH: Mg/Al-LDH, Ca/Al-LDH, Mg/Fe-LDH, or Cu/Al-LDH [15,16]. The representative material of this class of compounds is hydrotalcite, in which Mg²⁺ and Al³⁺ are present as cations. Traditionally, these materials are obtained with the tandem co-precipitation method and inorganic alkalis [17]. A new approach which is a viable alternative to the traditional one is the use of the mechanochemical synthesis method but also the use of organic alkalis [11,18]. However, using organic alkalis presents several advantages, such as the elimination of the possibility of contamination of the final material with alkaline cations, fewer steps in the synthesis process, the low amount of energy involved and also of water used, etc.

Organic-inorganic hybrid supramolecular systems are developed by encapsulating the metallo-phthalocyanines into a layered double hydroxide (LDH) structure [19,20]. To avoid conventional catalyst-separation methods (filtration, centrifugation), magnetic nanoparticles (MNPs) were considered as alternative catalytic supports [21]. MNPs represent a widely used class of nanoparticles due to their medical and pharmaceutical applications. The main representatives are magnetite (Fe₃O₄) and maghemite (γ-Fe₂O₃). They are ferrite colloids that present a spinel crystal structure with oxygen ions forming a close-packed cubic lattice with iron ions located at the interstices [22,23]. The antiferromagnetic coupling (super exchange through oxygens) that takes place between the Fe³⁺ ions in the octahedral and tetrahedral interstices leads to the magnetization of Fe₃O₄ [22].

This research aimed to synthesize different magnetic-Fe-containing LDH systems incorporating different metallo-phthalocyanines (Pcs) and their application into the removal of β-lactam antibiotics from water through a simple and environmentally friendly method. The preparation method for the LDHs and the nature of the precipitating agent were studied in order to achieve efficient photocatalytic systems and to determine the most efficient method from an environmental point of view.

2. Results and Discussion

2.1. Characterization of the Materials

2.1.1. X-ray Diffraction (XRD)

X-ray patterns obtained for magnetic particles (Fe₃O₄) and hydrotalcite (Mg, Al) prepared by co-precipitation and hydrotalcite (Mg, Al) deposited on magnetic particles are typical of these materials (Figure 1). The X-ray diffractograms show narrow and intense diffraction lines located at low angles of 2 theta and broad and less intense lines at high diffraction angles, with this behavior corresponding to LDH-type layered materials [13]. The diffraction lines from approximately 2θ(°)/d(Å) = 30/2.91766; 36/2.52937; 43/2.10491; 57/1.60640; and 63/1.46029 corresponding to the crystal planes (220), (311), (400), (511), and (440) of Fe₃O₄ are clearly observed. The diffractogram of the hydrotalcite deposited on the magnetic particles does not show other phases than those associated with the two components. As presented in the experimental section, LDHx@MPc@MNP (x = 1–5; m = Fe, Cu, Ni) represents Mg_0.75Al_0.25-LDH synthesized by traditional co-precipitation (LDH1), Mg_0.325Fe^II_0.325Fe^III_0.25-LDH (LDH2) synthesized by traditional co-precipitation, Mg_0.325Fe^II_0.325Fe^III_0.25-LDH synthesized by the mechanochemical method (LDH3), Mg0.325Fe0.325Al0.25-LDH synthesized by traditional co-precipitation (LDH4), or Mg0.325Fe0.325Al0.25-LDH synthesized by co-precipitation with TMAH (LDH5), on which the metallo-phthalocyanine (MPc) was drafted and attached with magnetic nanoparticles (MNPs).

In all Fe-LDH samples (LDH2 and LDH3), shown in Figure 2 and Figure 3, the Fe₃O₄ (ICDD 00-019-0629) lines are easily distinguishable (intense line for 311 and weaker ones for 220, 400, 511, 440), together with the characteristic lines of the LDHs, which highlights the oxidation of Fe²⁺ to Fe³⁺ similar with obtaining stable structures (FeO∙Fe₂O₃). This synthesis approach does not seem to allow protection against this oxidation, especially considering that all solid syntheses were carried out in an air atmosphere.

In comparison with LDH2, in the case of LDH3 (mechanochemical preparation), after the insertion of phthalocyanines and the deposition of magnetic particles, the LDH structure is visible, along with the lines of iron oxides, as can be seen in Figure 3b, which indicates its reconstruction during the rehydration of the mixed oxides.

For LDH4 or LDH5 samples, the Fe₃O₄ (ICDD 00-019-0629) lines (Figure 4 and Figure 5) are easily distinguishable (the intense line 311, together with weaker ones for 220, 400, 511, 440), which highlights the partial oxidation of Fe²⁺ to Fe³⁺, similar to obtaining stable structures (FeO∙Fe₂O₃). LDH5-based composites present a structure similar to LDH2, which corresponds to a mixture of oxides with less evident LDH arrangements.

The network parameter “a”,

Table 1. Cell parameters, IFS values, and crystallite size for prepared composites.

LDH Sample	Parameters		IFS 1 (Å)	I003/I006	I003/I110	FWHM003 (°)	2 Theta (°)	D 2 (nm)
	a (Å)	c (Å)/d (Å)
LDH1@CuPc@MNP 1	3.059	23.249/7.75	2.95	2.22	6.42	0.621	11.40	13
LDH1@FePc@MNP 1	3.059	23.304/7.77	2.97	1.96	4.61	0.860	11.38	9
LDH1@NiPc@MNP 1	3.057	23.215/7.73	2.94	2.06	7.30	0.706	11.44	11
LDH2@CuPc@MNP 1	3.054	23.104/7.70	2.90	1.00	1.00	0.319	35.52	26
LDH2@FePc@MNP 1	3.064	23.612/7.87	3.07	1.17	0.70	0.287	35.54	29
LDH2@NiPc@MNP 1	3.035	23.259/7.75	2.95	1.20	1.00	0.365	35.51	23
LDH3@CuPc@MNP 2	-	-	-	-	-	0.493	35.60	17
LDH3@FePc@MNP 2	-	-	-	-	-	0.461	35.57	18
LDH3@NiPc@MNP 2	-	-	-	-	-	0.492	35.58	17
LDH4@CuPc@MNP 1	2.94	23.561/7.85	3.05	0.85	0.31	0.360	10.48	22
LDH4@FePc@MNP 1	2.94	22.861/7.62	2.82	1.18	0.37	0.053	11.22	26
LDH4@NiPc@MNP 1	2.94	23.098/7.70	2.90	0.73	0.31	0.280	10.88	28
LDH5@CuPc@MNP 2	-	-	-	-	-	0.053	10.11	30
LDH5@FePc@MNP 2	-	-	-	-	-	0.099	10.21	35
LDH5@NiPc@MNP 2	-	-	-	-	-	0.235	10.16	33

¹ For LDH; IFS = interlayer free spacing. ² For iron oxide.

, indicating the average cation-cation distance in the layered network, is similar for all samples, so no change of the cations in the octahedral positions takes place. Moreover, partial replacement of Mg²⁺ from LDH1 with Fe²⁺ in LDH4 and LDH5 confirm the oxidation of Fe²⁺ at Fe³⁺ considering their ionic radius (Mg²⁺ = 0.72 Å; Fe²⁺ = 0.78 Å; Fe³⁺ = 0.64 Å) is presented already in the XRD pattern [24]. The same phenomenon is also observed also for the “c” parameter. The ratio between the intensity of the first harmonic and the second, I₀₀₃/I₀₀₆, indicates the non-existence of any other impurity phase inside the LDH lamellar layer. The crystallite size estimated using the Scherrer equation suggested a different crystallinity for the three types of LDHs, with the most crystalline samples being the LDH2. The IFS parameter, which represents the interplanar distance between two consecutive layers, is close in all cases, showing that phthalocyanine prefers a similar intercalation position regardless of the LDHs. Therefore, the phthalocyanine is positioned parallel to the hydrotalcite layers or anchored to its lamellar edges.

It should be noted that the sample obtained in the presence of TMAH (LDH5) shows the more intense diffraction lines compared with LDH4. The X-ray diffractograms of the LDH4@Cu, Fe, or NiPc@ MNPs and the calculation of the IFS parameter show that the anchoring of phthalocyanines occurs mostly on the edges of the hydroxide lamellae.

2.1.2. Diffuse-Reflectance UV-Vis Spectroscopy (DR-UV-Vis)

The DR-UV-Vis spectra of parent LDH1, LDH2, LDH3, LDH4, and LDH5 (Figure S1) prove the simultaneous presence of Fe²⁺ and Fe³⁺ species. Due to the presence of Fe³⁺ species, all samples exhibited a strong absorption around 500 nm [23], which decreases their band-gaps compared with the LDH without iron, increasing their absorption capabilities in the visible domain. These values are sensibly smaller compared with the ones reported in the literature [25]. However, several specific species can be distinguished, e.g., in the form of isolated Fe³⁺ (247–252 nm), Fe_xO_y oligomers (369–389 nm), and Fe₂O₃ particles (472–478 nm) [26]. The presence of Fe³⁺ in LDH4 and LDH5 was also confirmed by Raman spectroscopy (Figure S2).

The DR-UV-Vis spectra of the final composites (Figure 6) show a part of the strong absorption bands of phthalocyanine in the 350–450 nm region due to their B bands attributed to π–π* transitions in the macrocyclic ring [27], overlapping those of Fe₂O₃, and also in the 500–900 nm region due to their Q bands [28]. The blue color appearing at ligand π-π* transitions is characteristic of both the free and the encapsulated complexes.

The band around 700 nm, a Q band typical of metallo-phthalocyanines, represents the contribution of electron transfer between the orbitals of the aromatic 18-electron π system and the overlapping orbitals of the central metal [29]. Absorption has been confirmed for many phthalocyanine complexes and is related to the formation of singlet excitons [30,31]. The absorption maxima in the range 600–645 nm is characteristic of the presence of high-order aggregated or dimer phthalocyanines [32,33], while the Q band located around the 700 nm region is attributed to the presence of monomeric phthalocyanines. These bands are less evident for the composites based on LDHs containing iron (LDH2-LDH5).

2.1.3. Diffuse-Reflectance Infrared Spectroscopy (DRIFTS)

In Figure 7, a broad absorption band is observed in the 3700–3400 cm⁻¹ area corresponding to the vibrations of the OH groups, ν(O–H), while the band at 3000 cm⁻¹ is assigned the presence of hydrogen bonds between H₂O groups and the residual carbonate anion from co-precipitation, both located in the interplanar space [28]. The band at 1638–1650 cm⁻¹ corresponds to the H-O-H bending. The band located between 1100 and 650 cm⁻¹ is attributed to the vibrations of the carbonic groups, while the bands below the value of 600 cm⁻¹ correspond to M-O (M = Mg, Fe or Al) bonds.

The success of the preparation of the composites and the presence of phthalocyanine in them was also confirmed by the presence of specific bands in the infrared spectra [30]. The DRIFT spectra of the investigated materials (Figure 7) show the typical bands of the metal complexes, confirming their presence in the supported catalysts: 1287, 1333, 1418, 1468, 1495, and 1515 cm⁻¹ in the corresponding catalysts. For example, the band at approximately 1650 cm⁻¹ can also be assigned to the C=N bond, the band at 1120 cm⁻¹ can be assigned to the S=O bond, and the bands at 730, 1033, and 1090 cm⁻¹ can be attributed to the C-H bond. The very low intensity of these bands shows the existence of very small amounts of phthalocyanines in the chosen support.

2.1.4. Textural Properties

The textural properties of the composites are presented in

Table 2. Textural characteristics of the final composites.

Composite	Surface Area (m2 g−1)	Pore Volume (cm³ g−1)	Pore Size (nm)
LDH1_FePc_MNP	220	0.50	35, 54, 124
LDH1_NiPc_MNP	210	0.45	35, 57, 125
LDH1_CuPc_MNP	236	0.54	36, 55, 125
LDH2_FePc_MNP	91	0.22	39, 480
LDH2_NiPc_MNP	59	0.17	36, 504
LDH2_CuPc_MNP	84	0.22	36, 506
LDH3_FePc_MNP	90	0.28	38, 158, 295
LDH3_NiPc_MNP	73	0.24	36, 160, 293
LDH3_CuPc_MNP	79	0.26	36, 155, 300
LDH4_FePc_MNP	225	0.51	33, 93, 177
LDH4_NiPc_MNP	214	0.47	33, 100, 177
LDH4_CuPc_MNP	231	0.53	33,88,177
LDH5_FePc_MNP	133	0.32	36, 130
LDH5_NiPc_MNP	166	0.36	40, 155
LDH5_CuPc_MNP	159	0.34	37, 142

. The composites based on LDH2 and LDH3 present the lowest surface areas. This is a consequence of the structure and composition of the parent LDHs, those two having a predominant iron oxide composition with almost no evident LDH layered structure, as seen in XRD. The pore size is larger than that for the composites based on LDH1, LDH4, and LDH5. However, this can reflect the inter-particle voids.

The composites based on LDH1 and LDH4 present similar surface areas, although the latter present sensibly larger pores. At the same time, although TMAH can also act as a template molecule, besides being a synthesis agent, the surface area of the composites based on LDH5 is smaller to the one of the composites starting from the LDH4, having the same composition but prepared using NaOH as precipitating agent. In all cases, the surface area of the composites increased compared with the rehydrated initial LDHs, while the pore distribution became multimodal. The textural properties of the final composites, which also contain magnetic particles, are not influenced by the crystallite size of the parent LDHs.

2.1.5. Elemental Analysis

The amount of phthalocyanines calculated from the elemental analysis of selected samples,

Table 3. Amount of Cu or Ni-phthalocyanine determined in MNP@Cu or NiPc@ LDH4.

Composite	Amount Pc (mol/g Composite)
LDH1_FePc_MNP	6.5 × 10−5
LDH1_NiPc_MNP	6.6 × 10−5
LDH1_CuPc_MNP	6.7 × 10−5
LDH2_FePc_MNP	3.9 × 10−5
LDH2_NiPc_MNP	4.1 × 10−5
LDH2_CuPc_MNP	4.2 × 10−5
LDH3_FePc_MNP	3.8 × 10−5
LDH3_NiPc_MNP	4.1 × 10−5
LDH3_CuPc_MNP	4.3 × 10−5
LDH4_FePc_MNP	5.1 × 10−5
LDH4_NiPc_MNP	5.6 × 10−5
LDH4_CuPc_MNP	6.1 × 10−5
LDH5_FePc_MNP	4.2 × 10−5
LDH5_NiPc_MNP	4.4 × 10−5
LDH5_CuPc_MNP	4.7 × 10−5

, showed no significant differences for LDH1-, LDH4-, and LDH5-based samples, while for the composites derived from LDH2 or LDH3, the amount of photosensitizer was lower.

2.2. Antibiotic Removal Tests

The prepared samples were tested in the removal of two β-lactam antibiotics from water, namely amoxicillin and ampicillin (Figure 8).

Around 50% of the amoxicillin was removed after 10 min of stirring in dark for samples based on/starting from MgAl LDH and prepared by classical co-precipitation using NaOH as precipitating agent (LDH1 and LDH4), highlighting the adsorption ability of these composites (Figure 9). In contrast, LDH5, which was prepared in the presence of TMAH, exhibited a lower adsorption capability, being correlated with its textural properties. The composites containing FePc performed slightly worse than those containing CuPc or NiPc for all three starting LDHs.

Al three types of composites presented a lower adsorption capability of ampicillin, which can underline the importance of the antibiotic structure, with even only one hydroxyl group influencing this phenomenon. However, the photocatalytic contribution seems to have a slightly higher magnitude for amoxicillin.

The composites based on LDH2 and LDH3 presented, in general, lower activity than the other composites (Figure 10). The adsorption capability was lower. At the same time, the trend related to the nature of the substrate was preserved, the amoxicillin being removed in a larger amount.

No significant differences were observed between the two sets of composites, although the one prepared by the mechanochemical method exhibited a slightly higher activity.

If only the photocatalytic contribution to the removal process is considered (Figure S3), it can be concluded that in this step, phthalocyanine is responsible for the performance of the composites. The degree of removal depends mainly on the amount of MPc and it does not follow the variation in the band-gap of the parent LDHs.

There are two trends which can be observed by summarizing the results. First of all is the dependence of the removal degree on the nature of the composite. This is a consequence of the two phenomena involved: adsorption and the photocatalytic reaction. The different adsorption is mainly a consequence of the textural properties of the composites. At the same time, the literature states that for mesoporous supports with larger pores (above 200 nm), as it is the case for the composites based on LDH1 and LDH4, the amount of ampicillin (and this can be extended also to amoxicillin) adsorbed does not depend on the pore size and is only related to the surface charge density (σ) of the adsorbent surface [34]. In addition, in the case of LDHs, it was found that the adsorption uses a chemisorption mechanism, following the pseudo-second order [35,36]. However, for all the composites, the adsorption degree of amoxicillin was higher than for ampicillin, and this is the second trend observed. Hydrogen bonds, electrostatic repulsion, electrostatic attraction, competition of excess -OH and weak van der Waals forces are responsible for the adsorption [35,36]. The only difference in the structure of the two antibiotics is the presence of the hydroxyl group attached to the aromatic ring in the case of amoxicillin, which offers a more hydrophilic character to amoxicillin compared to ampicillin. This hydroxyl group offers an additional bonding capacity for amoxicillin. This hydroxyl group is also responsible for the higher degradation of amoxicillin compared with ampicillin, a trend which is in accordance with the literature [1,37]. In the same time, this hydroxyl group increases the electron density of the aromatic ring through resonance and inductive effects, making it more reactive towards oxidation.

In all the removal tests presented, although in some cases the removal of the antibiotic reached 100%, total decomposition of the antibiotic (mineralization) is not reached. We can assume that a longer irradiation time can achieve total mineralization or at least a higher degree of removal; however, from a practical point of view, this can be unfeasible. The literature presents two pathways of degradation: (i) hydroxylation and (ii) beta-lactam ring opening [38]. The FTIR-ATR spectra of amoxicillin solutions recorded every 30 min showed the continuous increase in the number of functional groups and bonds related to the degradation of the antibiotics (Figure S4). This is proof that the antibiotics are degraded to smaller molecules. However, the fact that the bands associated with these bonds are not decreasing after a certain point is an indication that these molecules are not totally degraded. Some authors concluded that the degradation intermediates still present toxicity due to the carboxyl groups generated during the degradation process [39].

3. Materials and Methods

3.1. Preparation of the Composites

Mg_0.75Al_0.25-LDH (LDH1) and Mg_0.325Fe^II_0.325Fe^III_0.25-LDH (LDH2) were synthesized using the traditional co-precipitation method, while Mg_0.325Fe^II_0.325Fe^III_0.25-LDH was synthesized using the non-traditional mechanochemical method (LDH3), using inorganic alkali for pH adjustment. In addition, Mg_0.325Fe_0.325Al_0.25-LDH was synthesized using co-precipitation by inorganic (LDH4) or organic alkali (LDH5).

The preparation was achieved by mixing two solutions. The first solution consisted of Mg(NO₃)₂∙6H₂O, FeCl₂∙4H₂O and Al(NO₃)₃∙9H₂O in a molar ratio of 0.325/0.325/0.25. The second solution contained 0.23 moles NaOH and 0.092 moles Na₂CO₃ (1M in Na₂CO₃) for LDH1, LDH2, and LDH4 or Tetra Methyl Ammonium Hydroxide (TMAH), 25% in water, for LDH5. Both solutions were mixed at 600 rpm and pH 10 at room temperature. The solutions were mixed with a feed rate of 60 mL∙h⁻¹ each. At the end of the precipitation, the suspension was aged in air for 18 h at 80 °C, then cooled to room temperature and filtered, washed with bi-distilled water until pH 7 and, finally, dried for 24 h in air at 120 °C (LDH1, LDH2, LDH4, LDH5). The synthesis in the presence of TMAH required a significant decrease, of almost 10 times, in the amount of bi-distilled water used for washing compared to the synthesis with inorganic alkali (NaOH). The non-traditional mechanochemical method for obtaining LDH3 was performed by a direct mortaring of the aforementioned precursors for 1 h at a pH of 10, and then the obtained solid was washed and filtered.

For the insertion of phthalocyanine in the LDH structure, the latter’s memory effect was exploited. Thus, the hydrotalcites were calcined at 460 °C after which they were cooled in a desiccator. The hydrotalcites were placed directly in a solution of Cu or Ni-phthalocyanine 3,4′,4″,4′′′-tetrasulfonic acid and left to stir for 12 h. After, the phthalocyanine-containing hydrotalcite was filtered off and dried at 120 °C for 24 h in an air atmosphere (LDH1@Fe, Cu or NiPc, LDH2@Fe, Cu or NiPc, LDH3@Fe, Cu or NiPc, LDH4@Fe, Cu or NiPc, LDH5@Fe, Cu or NiPc). For the synthesis of LDH@Fe, Ni or CuPc@ MNP, FeCl₂∙4H₂O, and Fe(NO₃)₃∙9H₂O were added to a suspension of Ni or CuPc@ LDH (for a mass ratio MPc@Fe-LDH/Fe₃O₄ = 3/1). The suspension was heated to 70 °C, after which the iron was precipitated by adding 30 mL of NaOH (0.15M). The obtained material was removed with the help of a magnet, washed with water and dried at 70 °C for 12 h. (LDH1@Fe, Cu or NiPc@MNP, LDH2@Fe, Cu or NiPc@MNP, LDH3@Fe, Cu or NiPc@MNP, LDH4@Fe, Cu or NiPc@MNP, LDH5@Fe, Cu or NiPc@MNP). Figure 11 shows Cu-phthalocyanine 3,4′,4″, 4′′′-tetra sulfonic acid as model for the phthalocyanines used.

3.2. Characterization Techniques

3.2.1. X-ray Diffraction (XRD)

XRD data were collected with the Shimadzu (Kyoto, Japan) XRD-7000 diffractometer, with the monochromatic Cu Kα radiation (λ = 1.5406 Å, 40 kV, 40 mA) at a scanning rate of 0.1 °/min in the range 2θ = 10–90°. The size of the crystallites was estimated using Scherrer’s equation.

3.2.2. Diffuse-Reflectance UV-Vis Spectroscopy (DR-UV-Vis)

The DR-UV-Vis spectra were collected using a Specord 250 (Analytic Jena, Jena, Germany) spectrometer equipped with an integrating sphere as a measuring device in the reflectance mode. MgO was used as reference material.

3.2.3. Diffuse Reflectance Infrared Spectroscopy (DRIFT)

Infrared spectra were collected with a Brucker (Billerica, MA, USA) Tensor II equipped with a Harrick Praying Mantis Diffuse Reflectance attachment. A Perkin-Elmer Spectrum Two with Attenuated Total Reflection Fourier (ATR) cell with diamond crystal was also used to collect the spectra. The final spectra are the average of 20 scans with a resolution of 4 cm⁻¹.

3.2.4. Textural Analysis

The textural characteristics of the catalysts (surface area, pore volume, and pore size) were determined from adsorption-desorption isotherms of nitrogen at −196 °C using a Micromeritics (Norcross, GA, USA) ASAP 2020 Surface Area and Porosity Analyzer. The samples were outgassed for 12 h at 120 °C before nitrogen adsorption.

3.2.5. Elemental Analysis

The elemental analysis (CHNS) of the final composites was performed using a EuroVector (Pavia, Italy) EuroEA EA3000 Series analyzer.

3.3. Antibiotic Removal Tests

The irradiation of the composites has been carried out using a sunlight simulator Sciencetech (London, ON, Canada) SF150-A Small Collimated Beam Solar Simulator (Air Mass AM1.5G Filter and Light-Tight Reaction Chamber). The advance of the reaction was monitored by liquid chromatography using an Agilent (Santa Clara, CA, USA) HPLC equipped with a Agilent Zorbax SB-C18 column (4.6 × 150 mm, 5 microns); the mobile phase was 25 mM KH₂PO₄:ACN = 60:40, under a flow of 0.5 mL min⁻¹, column temperature 60 °C, and with DAD detection at 204 nm. For the experiments, 15 mL of a prepared 0.17 mM solution of antibiotic and 60 mg of catalyst were added into quartz test tubes. Before turning on the lamp, the samples were kept in the dark, under stirring, for 10 min to reach the absorption-desorption equilibrium.

4. Conclusions

A series of composites was prepared starting from five types of LDHs (co-precipitated in the presence of inorganic or organic alkali and the mechanochemical method), which were then exchanged with three types of metallo-phthalocyanines. In the end, magnetic nanoparticles were attached. LDHs containing Fe showed a partial oxidation from Fe²⁺ to Fe³⁺. The final composites presented a multimodal pore distribution. Samples containing evident LDH structures performed better in general than the ones containing iron oxide mixtures, the composites being more active towards amoxicillin removal compared with ampicillin. The nature of the phthalocyanine did not have such a great influence, although some differences in the activity were observed. The removal was a combination between adsorption and photocatalytic degradation. However, FTIR-ATR analysis of the antibiotic solutions at different reaction times suggest that the total decomposition is not reached, or at least not to an important extent, after the two hours of irradiation. The best composites for this application were those based on LDH4 (MgFe²⁺Al prepared by co-precipitation in the presence of NaOH), which presented high adsorption capacity in 10 min and, at the same time, high photocatalytic activity for both amoxicillin and ampicillin. Although the composite based on LDH1 (MgAl prepared by co-precipitation in the presence of NaOH) presented similar performance in the removal of amoxicillin, it performed poorly for the removal of ampicillin. This difference is the result of both adsorption and photocatalytic decomposition, which are affected by the structure of the target molecules. The use of TMAH as template and precipitating agent does not lead to an increase in performance for the LDH5-based (Mg_0.325Fe_0.325Al_0.25-LDH prepared by co-precipitation) composites due to the oxidation of Fe³⁺ and formation of mixed oxides during rehydration. The poor performance of the composites based on Fe²⁺ and Fe³⁺ simultaneous doping of LDH (LDH2 and LDH3), in comparison with the composites doped only with Fe²⁺ (LDH4 and LDH5), is related mainly to their structure and composition. The first are mixtures of iron oxides, compared with the real LDH structure of the latter. However, the solid prepared by mechanochemical reaction (LDH3) presents some degree of layered structure as seen by XRD. This is also reflected in the textural properties of the material.