2.1. Characterization of Catalysts
ZnAl LDH containing either carbonate or chloride anions was prepared in a single step by a modified urea method [25
]. The XRD spectra of the prepared ZnAlCl and ZnAlCO3
showed the typical patterns (Figure S1
). LDHs containing different anions, analyzed by field emission scanning electron microscopy (FE-SEM), exhibit different morphologies: ZnAlCO3
is constituted of hexagonal microcrystals assembled to form a sand-rose morphology with a main dimension of about 2 μm, while ZnAlCl consists of flat hexagonal microcrystals of diameter 2–5 μm and thickness of 200–300 nm. The distinct LDH morphologies are shown in Figure 1
The synthetized LDHs were used as supports of AgCl nanoparticles using a method developed in a previous work [24
], from which the basic principles will be recalled here. The method consisted of the addition of a silver nitrate solution to an aqueous dispersion of ZnAlCl. The presence of chloride anions provoked AgCl precipitation on the LDH surface, while the charge balance of LDH was assured by the nitrate anions. A sample named LDH1 was prepared starting from ZnAlCl; the addition of AgNO3
to a ZnAlCl dispersion containing NaBH4
(as reducing agent) gave a sample called LDH2. A sample called LDH3 was prepared starting from ZnAlCO3
, which was quickly washed with HCl 10−3
M in order to exchange the carbonate anions present on the lamella surface with chloride anions while the interlayer region remained unchanged (see experimental). In this work, no significant variations on the metal content were found, proving that the HCl treatment did not leach the LDH structure. The diffraction patterns of the prepared samples are reported in Figure 2
and show, besides the typical reflections of the LDH phase, additional reflections ascribable to AgCl. In particular, the peaks labeled with the dashed lines were assigned to the crystallographic planes (111
) and (222
) of the cubic phase of AgCl by comparison with a reference spectrum (PDF number: 85-1355). The interlayer distance of LDH1 and LDH3 was unchanged with respect to the starting material, suggesting that the exchange of the interlayer carbonate or chloride anions with nitrate was absent or very low; in the latter case, the formation of solid solutions in which the nitrate anions are solubilized in the chloride or carbonate phase could be supposed. Differently, in the sample LDH2, the reflection at 8.9 Å, typical of the nitrate phase of LDH, indicated that a large amount of interlayer chloride anions were substituted by nitrate anions (Figure 2
b). In this case, the presence of sodium borohydride in the ZnAlCl dispersion can favor the ion exchange between BH4−
]. When the AgNO3
is added, a part of the silver cations is reduced by the intercalated BH4−
and the positive charges of the lamellae are neutralized by nitrate anions of the silver salt. The formation of metallic silver was confirmed by the presence of the crystallographic planes (111
) and (200
) of its cubic phase in the XRD pattern.
reports the amount of silver in the samples, expressed as weight percentage, and the diameter of the AgCl NPs calculated by applying the Debye-Scherrer equation to the (111
) and (200
) reflections of the silver chloride cubic phase. The synthetic method used to obtain the composites seemed to affect the dimensions of the AgCl NPs that ranged from 40 to 100 nm. The smallest AgCl NPs were obtained starting from the LDH containing only the chloride anion on the surface.
shows the FE-SEM images of the composites. The composites are characterized by the presence of nanoparticles ascribable to AgCl on the surface of the LDH crystals, which in turn retain the original morphology: flat hexagonal and sand-rose crystals for the chloride and carbonate forms of LDH, respectively. AgCl size distribution, obtained by measuring 150 particles, is also reported in Figure 3
, while the mean diameters are indicated in Table 1
. The mean diameters of LDH1 and LDH2 were in good agreement with the values obtained by XRD; conversely, LDH3 showed larger particles than the calculated value of the crystal domains, suggesting a polycrystalline character of the particles.
The optical properties of the composites were studied by collecting UV–Vis reflectance spectra and comparing them with that of pure AgCl. As shown in Figure 4
, the samples displayed the characteristic absorption of the AgCl below 400 nm, in addition to the absorption band of LDH under 300 nm. In the LDH1 and LDH3 spectra, only the AgCl absorption was detectable, indicating that no Ag NPs were formed. The broad band in the Vis region of the LDH2 spectrum was ascribed to the surface plasmon resonance (SPR) of the silver nanoparticles [27
], in agreement with the XRD. The observed SPR covered a wide range of wavelengths due to the contribution of Ag NPs with a broad size distribution and Ag NP aggregates. Moreover, it was found to be centered at 440 nm, indicating the presence of Ag NPs with a diameter of about 30 nm [28
] according to the crystalline domain calculated by the Debye-Scherrer equation.
2.2. Photodegradation Experiments
The catalytic performance of the composites was investigated using photodegradation of Rhodamine B (RhB), a model dye, in aqueous solution under irradiation with a halogen lamp emitting radiation with λ > 350 nm. Table 2
reports the amount of catalyst per mL of RhB solution used in different experiments and the estimated half-life (t1/2
) of RhB.
To prove the synergic contribution of LDH and silver chloride, control experiments were performed as well. Degradation did not occur in the experiments without the photocatalyst. By using bare ZnAlCl and ZnAlCO3
, only modest catalytic activity was observed, even when physically mixed with AgCl (see Table S1 and Figure S2
Pure AgCl particles did show some activity in the photodegradation of RhB. The tested silver chloride was precipitated in the same conditions used to prepare the AgCl/LDH composite and the solid obtained was characterized by crystals of dimension 1–2 μm (Figure S3
). The temporal evolution of the spectral changes of RhB in the presence of pure AgCl is reported in Figure 5
A. The absorption spectrum of pure RhB shows a band centered at 554 nm, which decreases after contact with the catalyst in the dark due to adsorption of the dye on the solid surface (spectra at 0 min in Figure 5
A). During irradiation, the absorbance of the major absorption band gradually decreases and the position of its maximum moves toward the blue region; the maximum shifts from 554 nm (N
’-tetraethylated Rhodamine molecule, RhB) to 498 nm (Rhodamine) [29
] and the RhB degradation is completed after 80 min irradiation. The blue shift suggests that the degradation of RhB occurs in a stepwise manner in which the ethyl groups are gradually removed to form the de-ethylated species of RhB, which are then cleaved completely.
In order to get information about the contribution of de-ethylation and cleavage processes during RhB photodegradation, the relative concentration of the de-ethylated species and RhB (calculated as reported in the experimental section) was plotted as a function of irradiation time, as depicted in Figure 5
B. As shown in the figure, the concentration of RhB decreases over time, approaching zero after 50 min of irradiation, while the concentration of the de-ethylated species increases up to the maximum value after about 30 min. Then, cleavage of the chromophore group occurs, consuming the de-ethylated species and causing its concentration to decrease and approach zero at 70 min. On the basis of these findings, it can be stated that the degradation process starts mainly with RhB de-ethylation, which is followed by cleavage of the de-ethylated species.
This RhB photodegradation process has been observed in earlier studies [33
] and seems to depend on the adsorption modes of RhB on the surface of the catalyst. Generally, AgCl particles are negatively charged due to an excess of chloride anions on the surface; accordingly, RhB molecules are adsorbed on the AgCl surface through positively charged diethylamino groups. The reactive oxygen species, produced near the adsorbed RhB, preferentially attack the nearer organic groups, which are the amino groups in this case, thereby leading to de-ethylation.
We then performed kinetic experiments with the composites LDH1, LDH2 and LDH3 in the same conditions used for AgCl (mg Ag/mL RhB = 0.15). The temporal evolution of the RhB absorption band under irradiation (Figure 6
A–C) showed that the performance of LDH1, LDH2 and LDH3 was better than that of the pure AgCl, bleaching the RhB solution in 35, 45 and 15 min, respectively.
A shows that the wavelength of the absorption maximum (λmax
) is always centered at 554 nm, indicating that, in the discoloration mechanism, the cleavage process is predominant in the presence of LDH1. Conversely, in the kinetic experiment performed with LDH2, a distinct blue-shift of the λmax
to 539 nm occurred after 35 min of irradiation (Figure 6
B). In this case, the contribution of the de-ethylation process in the whole degradation procedure seems more consistent than for the previous catalyst. In order to get more quantitative information about the contribution of the de-ethylation and cleavage processes, the relative concentrations of the RhB, de-ethylated and cleaved species (calculated as reported in the experimental section) were plotted as a function of irradiation time in Figure 6
D–E for the experiments with LDH1 and LDH2. It can be observed that RhB degradation shows about the same time dependence with both catalysts and that de-ethylation and cleavage occur concomitantly, resulting in more efficient photodegradation processes in comparison with the process catalyzed by pure AgCl. Specifically, de-ethylation prevails over cleavage only during the first few minutes, while the concentration of cleaved species becomes significantly larger than that of de-ethylated species already after about 15 min. Moreover, cleavage is faster with LDH1 so that the concentration of de-ethylated species is always larger with LDH2.
The lack of enough experimental data for the composite LDH3 did not allow us to obtain a concentration trend for the species formed during photodegradation; however, it is possible to compare the data obtained after five minutes since the irradiation started with the corresponding data obtained with LDH1 and LDH2. The percentage of de-ethylated species in LDH3 (22.5%) was close to that of LDH1 (18%) and LDH2 (25%) catalysts, while the contribution of cleavage to the degradation process was very high (36%) in comparison to the other catalysts (7% for LDH1 and 12% for LDH2), lowering the RhB concentration to 42% of its initial value.
Although the present data did not allow us to establish whether de-ethylation and cleavage with LDH-based catalysts were consecutive or parallel reactions, it is clear that, in comparison with pure AgCl, the presence of ZnAl LDHs made cleavage much more competitive with respect to de-ethylation in the photodegradation process of RhB. Moreover, the presence of ZnAl LDHs appears to speed up the degradation kinetics of RhB, as shown by the RhB half-life (t1/2
, Table 2
), obtained from interpolation of the relative concentration data of RhB. The RhB t1/2
was halved in the presence of LDH1 and LDH2 compared with pure AgCl (12 and 9 min versus 22 min, respectively) and was lower than 5 min when LDH3 was used.
The LDH1 and LDH3 catalysts were recovered after the photocatalysis experiments and were analyzed with XRD (Figure 7
), revealing a small reflection assignable to metallic silver. This demonstrated that a part of the silver ions underwent a reduction during the irradiation, forming the [email protected]
The superior performance of the AgCl/LDH composites compared to pure AgCl may be attributed to the peculiar characteristics of the composites:
(i) RhB can be adsorbed on the composites through either positive amino groups or carboxylic groups interacting with the negative surface of AgCl or the positive surface of LDH, respectively. The orientation of RhB with the carboxylic groups on the surface of the catalyst favors the attack on the chromophore structure (cleavage) by the reactive species formed on the surface of the catalyst [33
]. Conversely, as described above, the adsorption of RhB with the amino groups favors the de-ethylation process. These two different adsorption modes may explain the simultaneous occurrence of the two degradation pathways of the dye.
(ii) The dimensions of AgCl particles grown on the LDH surface are in the nanometric scale, ranging from 40 to 100 nm, more than an order of magnitude smaller than pure AgCl (about 1 μm). Generally, the smaller the particle, the higher the photocatalytic activity due to the increase in the number of active sites [37
A comparison among the photoactivity of AgCl/LDH composites is useful to select the best catalyst and to get insight into the main factors contributing to the design of an efficient catalyst. LDH3’s superior activity may be explained considering the dimensions of AgCl grown on the LDH surface. The diameter of crystalline domains of AgCl particles in LDH3 is 2.3 times smaller than AgCl particles in LDH1 so that, for an equal amount of AgCl, the total surface of AgCl particles in LDH3 is about 2.3 times larger; this is reflected in the higher number of catalytic sites in LDH3. For proof of the effect of number catalytic sites on photocatalytic performance, a catalytic test with increased LDH1 content was performed (mg Ag/mL RhB = 0.30). The performance of LDH1 increased considerably, bleaching the RhB solution in 15 min, similar to the performance observed for LDH3 (see UV–Vis spectra in Figure S4
It was interesting to observe that, when the LDH1 concentration was doubled (mg Ag/mL RhB = 0.30), the percentage of RhB after 5 min of irradiation (41.9%, Table 3
) was about the same as that found for LDH3 (41.6%) with mg Ag/mL RhB = 0.15. However, the percentage of de-ethylated species (38.8%) in LDH1 was larger than that of LDH3 (22.5%), so that the contribution of cleavage in LDH1 was smaller (19.3%) in comparison with LDH3 (35.9%).
2.4. Mechanism of Photocatalytic Degradation
The identification of the reactive species involved in the degradation process of RhB allowed us to get insight into the reaction mechanism in the presence of our catalysts. To this end, experiments of photodegradation in the presence of radical scavengers, including benzoquinone (BQ) for the superoxide radical anion (O2·−
), isopropanol (IPA) for the hydroxyl radical (OH·
) and ethylenediaminetetraacetic acid disodium salt (Na2
-EDTA) for the positively charged vacancy or hole (h+
), were carried out. Figure 9
shows that the photoactivity of all of the catalysts was dramatically suppressed in the presence of BQ, indicating the superoxide anion was the main active species in the photochemical process.
On the other hand, photodegradation in the presence of LDH3 was hindered by EDTA, although to a lesser degree than BQ, indicating that h+ may also be involved in the process.
According to the literature, the superoxide anion can mainly be formed through a photosensitized mechanism [38
], wherein the first step is the absorption of light by the dye to induce an intramolecular π–π transition. Once RhB is excited, it transfers its electron to the inorganic matrix, forming an RhB radical cation [41
]. More specifically, the electron is expected to be accommodated in the conduction band (CB) of AgCl, for which the redox potential vs. normal hydrogen electrode (NHE) (−0.09 V [42
]) is higher than that of ZnAl LDH (−2.0 V [43
]). Subsequently, it can be transferred to an oxygen molecule adsorbed on AgCl to form the superoxide anion O2·−
, due to the higher potential of the O2
redox couple (−0.046 V vs. NHE [42
]). The superoxide anion can also diffuse along the LDH surface through an interchange reaction with the LDH anions and, finally, react with the radical cation of the dye (adsorbed either on AgCl or LDH), provoking degradation of the dye.
As mentioned in Section 2.2
, the position of the dye that is preferentially attacked by O2·−
is the group of the dye interacting with the surface of the catalyst; i.e., the carboxylic group or the amino group for dye adsorption on LDH or AgCl, respectively.
Moreover, due the presence of metallic silver in all the LDH samples, an SPR-mediated charge transfer process should be considered alongside the dye photosensitized mechanism [44
]. Silver plasmonic nanoparticles, in direct contact with the AgCl surface, act as a dye sensitizer, absorbing resonant photons that generate electron-hole pairs in Ag nanocrystals. The surface of AgCl particles is negatively charged due the presence of terminal Cl−
ions, thereby polarizing the electronic distribution of Ag NPs and moving the plasmon-excited electrons on the surface of the Ag NPs far from the Ag/AgCl interface and the holes to the surface of the AgCl NPs [18
]. The photogenerated electrons are trapped by O2
to form O2·−
, leading to dye degradation.
Finally, considering that AgCl has an indirect band gap of 3.25 eV [47
], electron–hole pairs can be generated by the absorption of photons with λ in the range 350–382 nm. This process, which is clearly not relevant for RhB degradation in the presence LDH1, could become significant for LDH3 because it has been reported that the carbonate anion reduces electron–hole recombination [48
2.5. Antibacterial Activity of Catalysts
In a previous work [24
], we found that AgCl/LDH composites exhibited very good antimicrobial activity against bacteria and fungi. Here, it was of interest to test the antibacterial activity of the best catalysts, LDH1 and LDH3, against Escherichia coli
, a typical water contaminant. Table 4
shows the minimum inhibitory concentration required to inhibit the growth of 90% of bacteria (MIC90); gentamicin and LDH were used as negative and positive controls, respectively.
The samples showed good antibacterial activity, even if they were lower than that of the positive control, gentamicin. The best antibacterial activity was expressed by LDH3, in agreement with the lower dimensions of the AgCl NPs, which correspond to greater surface area and thereby promote better contact with the bacteria [49
Kinetic experiments of the antibacterial activity of LDH1 and LDH3 in the dark and under irradiation (λ > 350 nm) were performed at the concentrations of MIC90 (Table 4
). The time–kill curves, reported in Figure 10
, showed that the number of live bacteria was already considerably reduced after 60 min in the presence of LDH1 and LDH3 under irradiation. It is noteworthy that the curves were superimposable with those of gentamicin in the dark, suggesting that the antibacterial activity of the catalysts under irradiation were comparable to those of the positive control. On the other hand, no reduction in bacteria was observed until 180 min for the catalysts in the dark or for the bacteria cultures both irradiated and in the dark in the absence of the catalysts. The high antibacterial activity of the catalysts under irradiation for a very short time can be explained by the formation of Ag clusters on AgCl (Figure 7
) and the production of reactive oxygen species (ROS), such as O2·−
, also observed in the photodegradation of RhB. The antibacterial activity of the composites is significantly enhanced due to a synergic effect of ROS and silver ions [50