Tunable Metallogels Based on Bifunctional Ligands: Precursor Metallogels, Spinel Oxides, Dye and CO2 Adsorption.

A semisolid gel material is a gift of serendipity via various chemical interactions, and metal incorporation (metallogels) imparts diverse functional properties. In this work, we have synthesized four metallogels from tetrapodal and hexapodal carboxylic acid/amide-based low-molecular-weight gelators with Ni(II) and Cu(II) salts. These metallogels can be tuned to be a low-temperature precursor of porous spinel oxides. These xerogels exhibit impressive water soluble dye and carbon dioxide adsorption, which coupled with the tunability and facile synthesis of porous spinel oxides underscores their potential in environmental remediation and energy applications.

Introduction

In recent years, soft materials attracted plenty of attention due to their astonishing capabilities in the copious facets of life. The semisolid gel materials are a gift of serendipity via various chemical interactions.[1] These materials are loaded with immobilized solvents like dimethyl sulfoxide,[2]N,N-dimethyl formamide,[3,4] dichloromethane,[5] tetrahydrofuran,[6] water,[7] methanol, and ethanol[8] into their inorganic or organic gel matrixes.[9] The self-assembled semisolid gel materials derived from the low-molecular-weight gelator are called metallogels in the presence of a metal ion, otherwise termed as organogels.[10] Particularly, the design of metal-incorporated gel materials is challenging and appealing because metal imparts several functional properties into the material, like conductivity,[11,12] absorption/ion-exchange,[13−15] catalytic applications,[16,17] redox activity,[18] and other energy-related applications.[19,20] Noncovalent interactions[21] like π–π interactions, hydrophobic interactions,[22,23] hydrogen bonding, and electrostatic forces imparted the three-dimensional metallogel network[24,25] with immobilized[26] solvent in the gel network. In metallogels, the cross-linked network structure is stable and formed via the metal–ligand interaction that can be tuned by replacing the metal ions.[27] Metallogels find applications in fields such as light-emitting diodes,[28] sensors,[29] biomaterials,[30,31] and photonics.[32] The desolvated metallogels, viz., xerogels, are functionalized materials that can play an essential role as single-source precursors of metal oxides for energy applications and in environmental remediation.[13−15,33−35]

The search to diversify the energy dependency on nonrenewable fossil fuels to a renewable and sustainable source of energy is one of the most challenging issues currently because the combustion of fossil fuels results in the emission of carbon dioxide and other anthropogenic greenhouse gases. CO2 is considered as the main cause of the fluctuation of global climate.[36] Therefore, the reduction of carbon dioxide from the atmosphere is one of the the main concerns currently. The porous metal–organic materials are found as good sorbents for the CO2 at mild pressure and temperature. Also, the climate concerns demand usage of other green alternative ways of energy production. In this regard, the electrochemical energy storage devices[37−42] can be a good alternate option. Graphite has been used as the anode substance in the Li-ion batteries[43] because of its long cyclic stability,[44] low cost, and excellent kinetics.[45]

Ferrite spinels have been considered as alternatives of the graphite anode.[46−48] The reaction mechanism of charging and discharging that occurs in the ferrite spinels is entirely different from typical Li-ion insertion/extraction.[47,49,50] Moreover, we can potentially control the electrochemical performance of ferrite spinels, and also, it can be improved by tuning the morphology.[51] Among spinel ferrites, NiFe2O4 (NFO) and CuFe2O4 (CFO) have been developed and garnered great interest in the lithium-ion batteries[35,52] due to very high theoretical capacities of NFO (915 mA h/g)[53] and CFO (893 mA h/g)[51] compared to that of graphite (372 mA h/g).[43] Also, natural abundance and low cost make NFO and CFO spinels as promising alternate anode materials in lithium-ion batteries. The surface area and pore size of the NFO and CFO spinels highly influence the electrochemical activity by offering a large area for the electrochemical reactions, better cycling performance, and superior discharge capacities.[51]

Similarly, water pollution is also a primary global concern, and synthetic organic dyes (∼7 × 105 tons of global annual production) is one of the chief contributors[54] to it because they are carcinogenic,[55] mutagenic, and toxic.[56,57] Water-soluble dyes are used in the paper printing, leather, and textile industries.[58] The presence of dyes even at an extremely low level (less than 1 mg/L) is clearly visible, and the color is an indicator of the water quality for the general community. Therefore, removal of the color from wastewater is usually more significant than the elimination of organic substances.[59] Hence, there is scope to find out new adsorbents, which can be easily prepared and also do not required nonrenewable raw materials. In this account, the adsorption of dyes by the metallogel have gained interest.[60]

Herein, we have taken tetrapodal (LMWG1) and hexapodal (LMWG2) carboxylic acid/amide-based ligands. Both the tetrapodal and hexapodal ligands contain an amide functional group in their core and a carboxylic group at the terminal. Coordination of these ligands via their terminal carboxylic acid binding sites and intermediate amide binding sites with Ni/Cu(II) ions results in metallogels Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4. We have made metal oxides (NiO and CuO) by controlled thermal decomposition of precursor metallogels. More importantly, after getting the metal oxides successfully, we have tuned our synthesized metallogels to get NiFe2O4 and CuFe2O4 spinels. To the best of our knowledge, this is the first example of spinel ferrite synthesis from precursor metallogels. The synthesized spinel ferrites exhibit higher surface area as compared to previously reported spinels,[61−64] which are made by various synthetic methods.

We have also illustrated the application of Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels (the dried form of the gel) in the disposal of water containing dyes with high adsorption capacity. The adsorption kinetics, isotherm, and the plausible mechanism of adsorption by xerogels have been proposed.

Results and Discussion

The synthesized bifunctional carboxylic/amide based ligands were used for ligation with the metal ions. The presence of the carboxylic group makes it an excellent ligand for the metal ions, and the presence of the amide functionality and aromatic rings increase the possibility of having hydrogen bonding and π–π stacking interactions. These are the ideal condition for the formation of the metallogel. We have employed the bifunctional ligands, LMWG1 and LMWG2 (Figures S1–S3, see the Supporting Information), with Ni2+ and Cu2+ for the synthesis of four metallogel compositions Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 (Figure ).

Figure 1

Schematics of the synthesis of metallogels based on the tetra/hexapodal ligands.

The multifunctional (carboxylic acid/amide) LMWGs were designed in such a way that they possess selective ligating properties toward metal ions. We expected that this variation could be employed to synthesize the precursor metallogel for the synthesis of spinel oxide. We have tuned the synthesis of the metallogel by introducing a 1:2 (Cu/Ni:Fe) metal ratio for the precursor metallogel (Figure ).

Figure 2

Pictorial representation of the formation of precursor metallogels and the subsequent thermal decomposition to spinels.

Under a field-emission scanning electron microscope (FESEM), we have examined the molecular aggregations of the metallogels. The FESEM images of the Ni@MG1and Ni@MG2 reveals that they form a porous, spongy structure, and the transmission electron microscopy (TEM) studies indicate aggregation of the nanofiber-like structure (Figure c and Figure S4a, see the Supporting Information). The FESEM images of the Cu xerogels (Cu@MG3 and Cu@MG4) show a more compact structure compared to the Ni xerogels (Ni@MG1and Ni@MG2), and the TEM studies reveal that the gel network is formed by aggregation of nanoparticles rather than nanofibers as observed in Ni xerogels. (Figure d and Figure S4b, see the Supporting Information). The average particle size of Cu@MG3 was found to be around 45 nm, from the TEM image (Figure S4c). The selected-area electron diffraction (SAED) images revealed that the xerogels are amorphous (Figure e,f). The amorphous nature of the synthesized metallogels are also confirmed from the powder X-ray diffraction analysis (Figure S5).

Figure 3

SEM images of (a) Ni@MG2 and (b) Cu@MG3, TEM images of (c) Ni@MG2 and (d) Cu@MG3, and SAED images of (e) Ni@MG2 and (f) Cu@MG3.

Rheological studies were conducted to estimate the viscoelastic nature of the metallogel. The storage (G′) and the loss (G″) moduli were plotted against frequency for the metallogel samples (Figure a and Figure S6, see the Supporting Information). It was observed that in all the metallogel samples, G′ was found to be higher than G″ over the entire test frequency range; e.g., the G′ – G″ difference for the Ni@MG1 and Ni@MG2 were found to be 27254 and 38233 Pa at 0.1 Hz frequency. These results indicate that all the obtained metallogels are elastic, but in the case of copper-incorporated metallogels, the values of the G′ – G″ difference were 59731 and 111327 Pa for Cu@MG3 and Cu@MG4, respectively, at 0.1 Hz frequency.

Figure 4

(a) Dynamic frequency sweep vs gain modulus (G′) and loss modulus (G″) of Ni@MG1, Ni@MG2, and (b) TGA of the metallogels.

The morphology indicates that the xerogels might exhibit porosity. Therefore, N2 gas adsorption–desorption studies were performed on the xerogels. The permanent porosities of the Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels were affirmed through the reversible adsorption–desorption experiment at 77 K, which indicates a type-III adsorption isotherm with the mesoporous nature of the xerogels (Figure S7, see the Supporting Information). Furthermore, the BET surface areas and pore size distributions with pore volumes of the Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels were also calculated. The BET specific surface areas of xerogels were 13–38 m2/g with total pore volumes of 0.055–0.01 cm3/g (Table S1, see the Supporting Information). Based on our results, Ni@MG1 and Ni@MG2 have larger surface area than Cu@MG3 and Cu@MG4, which is in agreement with the microscopic studies and the dye adsorption properties of Ni@MG1 and Ni@MG2.

To examine the thermal stability of the xerogels, thermogravimetric analyses (TGA) were performed. The thermal behaviors of Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 were examined from room temperature to 800 °C. The TGA results show that all of the xerogels have weight loss before 150 °C, which is due to the removal of physisorbed moisture and coordinated solvent molecules. In the nickel-based metallogel, approximately 55 to 60% weight loss was observed in between 360 and 490 °C, but in the case of copper-based metallogels, there is two-step weight loss observed; initially, a loss of ∼36–37% in the range of ∼250–340 °C and then second weight loss of ∼18–20% around 400–520 °C may be due to the decomposition of xerogels (Figure b).

To identify the calcined product, Ni@MG2 and Cu@MG4 samples were heated at 500 °C for 3 h in a tube furnace. The PXRD patterns of calcined samples were found to match with the reported NiO and CuO ICSD reference numbers 92132 and 69758, respectively (Figure and Figure S8, see the Supporting Information).

Figure 5

Powder X-ray diffraction patterns of (a) Ni@MG2 xerogel, (b) NiFe2O4 precursor xerogel, (c) NiO reference from the ICSD database, (d) NiO as synthesized by thermal decomposition of the Ni@MG2 xerogel, (e) NiFe2O4 reference from the ICSD database, and (f) NiFe2O4 as synthesized by thermal decomposition of the precursor xerogel.

The thermogravimetric studies indicate that these metallogels could be employed to synthesize the precursor metallogel for the synthesis of spinel oxide at relatively low temperature with a high surface area. We have used a 1:2 (Cu/Ni:Fe) metal ratio for the synthesis of the precursor metallogel (Figure ). Also, the formation of the precursor metallogel and its elastic nature were confirmed by rheological studies where G′ was found to be higher than G″ over the entire test region (Figure S9, see the Supporting Information). Thermal decomposition of these precursor xerogels at 500 °C for 3 h resulted in the formation of the NiFe2O4 (NFO) and CuFe2O4 (CFO) spinels. We believe that this was possible due to the selective ligation of the carboxylic group (possibly the octahedral coordination) and the amide (possibly the tetrahedral coordination), which resulted in the spinel oxide instead of a mixture of binary oxides.

To confirm the structures of both NFO and CFO spinels, we have performed PXRD studies (Figure and Figure S8, see the Supporting Information). The NFO spinel synthesized from the metallogel crystallizes in the cubic space group Fd-3m with a = 8.337 Å and V = 579.47 Å3 (ICSD reference number 40040), and the PXRD pattern of NFO shows the existence of the pure nickel ferrite spinel phase only. The CFO spinel synthesized from the metallogel crystallizes in the cubic space group Fd-3m with a = 8.416 Å and V = 596.15 Å3 (ICSD reference number 188854), but there are some iron oxide (ICSD reference number 26410) peaks (Figure S8).

The SEM micrographs of NiO, CuO, NiFe2O4 (NFO) and CuFe2O4 (CFO) synthesized from metallogels, and precursor metallogels are shown in (Figures S10 and S11, see the Supporting Information). The morphologies of the metal oxide and mixed metal oxide are uniform, and all are spherical-shaped nanoparticles. Additionally, from the images, it is clearly seen that these nanoparticles are nearly closed packed; this would help in the electron transport at the time of photoelectrochemical and electrochemical applications.[65] We also carried out the energy-dispersive X-ray (EDX) study to get details about the elements present in the metal oxide and mixed metal oxide. In the case of nickel ferrite spinel, the ratio of atomic weight percent of nickel and iron metal was Ni:2.2 Fe, and for copper ferrite spinel, it was Cu:1.7Fe (Figure S11, see the Supporting Information).

The BET specific surface areas of NFO and CFO ferrite spinels were 276.31 and 201.13 m2/g, respectively, with average pore sizes of 1.459 and 0.638 nm, respectively. The total pore volumes of NFO and CFO were 0.57 and 0.406 cm3/g, respectively (Figure S12, see the Supporting Information). These synthesized ferrite spinels from the precursor metallogel exhibit high surface area as compared to many reported procedures (Table S2, see the Supporting Information).

The presence of the polar functional groups (amide linkage) prompted us to examine the adsorption behavior toward water-soluble organic dyes. The materials show fast and efficient absorption of methylene blue (MB) and congo red (CR) (Figure a–g, Figures S13 and S14, see the Supporting Information).

Figure 6

Ni@MG1 (a) before dye adsorption and after (b) CR and (c) MB dye adsorption. Photographs of (d) CR and (e) MB dye removal by Ni@MG1 at different time intervals. (f) UV–vis spectra of the adsorption of (f) CR and (g) MB dyes as a function of time for the sample Ni@MG1.

Adsorption Kinetics Studies

To understand the adsorption behavior of Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels, we have performed the adsorption kinetics experiments (Figures S13 and S14, see the Supporting Information). For better understanding, the linear forms of the two kinetic models pseudo-first-order and pseudo-second-order were employed to examine the adsorption kinetic behavior. The pseudo-first-order kinetic equation is represented aswhere q (mg/g) and qe (mg/g) stand for the amounts of dyes adsorbed at a given time interval t (min) and dyes adsorbed at equilibrium, respectively, and k1 (min–1) is the pseudo-first-order rate constant.

The pseudo-second-order adsorption rate equation is represented aswhere k2 (g/mg·min) is the pseudo-second-order rate constant.

The kinetic parameters calculated with the help of pseudo-first-order kinetic models and pseudo-second-order kinetic models are presented in Table (Figures S15–S18, see the Supporting Information). The obtained results from both models clearly indicate that the pseudo-second-order model seems to be best fitted because, for both CR and MB dyes, R2 values acquired from the pseudo-second-order model are far closer to 1. Furthermore, the qe, cal (calculated) values are much closer to qe, exp (experimental) values by applying the pseudo-second-order model in comparison to the pseudo-first-order model (Table ). From these results, we can say that the adsorption processes of CR and MB on Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels follow the pseudo-second-order model.

Table 1

Kinetic Parameters from Both Kinetic Model, Pseudo-First-Order and Pseudo-Second-Order, of the Adsorption of CR and MB on Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 Xerogels

			pseudo-first-order			pseudo-second-order
sample code	dye	qe, exp	qe, cal	k1	R2	qe, cal	k2	R2
Ni@MG1	CR	9.44	2.57	0.0222	0.8295	9.42	0.2439	0.9983
Ni@MG1	MB	9.82	2.13	0.0304	0.9015	9.84	0.0915	0.9998
Ni@MG2	CR	9.29	5.89	0.0147	0.9678	8.88	0.0155	0.9916
Ni@MG2	MB	9.75	1.13	0.0399	0.8453	9.76	0.4284	0.9999
Cu@MG3	CR	9.13	8.81	0.0004	0.8707	9.16	0.0098	0.9844
Cu@MG3	MB	9.91	4.14	0.0549	0.9062	10.08	0.0471	0.9997
Cu@MG4	CR	9.74	6.07	0.0073	0.8427	7.86	0.0162	0.9614
Cu@MG4	MB	8.57	4.14	0.0549	0.9715	6.79	0.0114	0.9761

Adsorption Isotherms

The equilibrium adsorption isotherm model has an equal importance because it helps to understand how the adsorbates interact with adsorbents. Langmuir adsorption isotherm and Freundlich adsorption isotherm models were taken into account to evaluate the adsorption behavior. The Langmuir isotherm model is based on the hypotheses that (a) all the adsorption sites are identical and each adsorption site can take only one molecule, (b) the adsorption energy is independent of the surface coverage, and (c) after adsorption, the adsorbate molecules cannot change their position.[66,67] The following equation represents the Langmuir isotherm modelwhere Ce (mg/L), kL (L/mg), qe (mg/g), and qmax(mg/g) stand for the equilibrium concentration, Langmuir adsorption constant, equilibrium adsorption capacity, and the maximum adsorption capacity, respectively. The values of qmax and kL can be calculated from the intercept and slope of the linear plot of Ce/qe versus Ce, respectively.

The Freundlich adsorption isotherm model talks about the multilayer adsorption on the energetically heterogeneous surface.[68] The following equation represents the Freundlich isotherm modelwhere 1/n stands for the heterogeneity factor and kF is the Freundlich constant. The Freundlich constant (kF) can be obtained by plotting lnqe versus lnCe. The isotherm and the data fitted by the Langmuir adsorption model for CR and MB on Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels are presented in Figure (Figure S19–S22, see the Supporting Information). All the parameters obtained from the adsorption isotherm models are listed in Table . R2 values (the correlation coefficient) demonstrated that the Langmuir model is best fitted in comparison to the Freundlich model.

Figure 7

Adsorption isotherms for (a) CR and (c) MB on Ni@MG1, Ni@MG2, Cu@MG3 and Cu@MG4 xerogels respectively. Langmuir isotherm model for the adsorption of (b) CR and (d) MB on Ni@MG1, Ni@MG2, Cu@MG3 and Cu@MG4 xerogels, respectively.

Table 2

Adsorption Isotherm Parameters for CR and MB Dyes on Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 Xerogelsa

	Langmuir model					Freundlich model
sample code	dye	qmax	∼qmax ± error	kL	R2	1/n	kF	R2
Ni@MG1	CR	1428.57	1428.57 ± 43.0	0.0094	0.9931	0.5394	45.0737	0.9691
Ni@MG1	MB	980.39	980.39 ± 10.0	0.0454	0.9983	0.3693	116.85	0.9483
Ni@MG2	CR	1191.8	1191.8 ± 66.0	0.0390	0.9807	0.5054	38.6327	0.9759
Ni@MG2	MB	704.22	704.22 ± 10.0	0.0208	0.9981	0.4111	53.3887	0.8748
Cu@MG3	CR	497.51	497.51 ± 13.0	0.0088	0.9957	0.4575	21.9046	0.9023
Cu@MG3	MB	529.10	529.10 ± 8.0	0.0498	0.9975	0.2834	87.5053	0.9034
Cu@MG4	CR	526.315	526.315 ± 11.0	0.0093	0.9971	0.4203	29.2272	0.9284
Cu@MG4	MB	458.7	458.7 ± 4.20	0.0652	0.9993	0.2570	90.3237	0.8344

Note: qmax stands for maximum adsorption capacity evaluated with the help of the Langmuir adsorption isotherm model.

Therefore, the adsorption mechanism of congo red and methylene blue dyes on Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 xerogels is well explained by the Langmuir rather than Freundlich isotherm model. Furthermore, the low values of 1/n (heterogeneity factor) also favored Langmuir-like adsorption in comparison to the Freundlich adsorption isotherm. Our adsorption capacity results are well compared with different kinds of adsorbents reported in the literature (Table S3, Supporting Information).

Dye Adsorption Mechanism

Several types of materials were utilized as adsorbents in the removal of toxic organic dyes from polluted waste. Usually, (i) the surface area of the material, (ii) different charge interactions involved in between the adsorbate and adsorbent, and (iii) the pore structures of the materials were reported as an essential cause for the dye adsorption.[69] Both the organic dyes (CR and MB) contain aromatic rings and surface charge in their structures. Therefore, there are possibilities that xerogels can adsorb the dye through electrostatic interactions and hydrogen bonding.[70] Among all, there is a chance to establish π–π interactions in between the π cloud of benzene rings of dyes and the adsorbent.[71] Particularly, our synthesized materials are derived from amide functional group containing ligands; amide is a fascinating group because the −C=O moiety of the amide functional group can behave as an electron donor and the −NH moiety can behave as an electron acceptor.[72] Therefore, we expected that the amide group also plays an essential role in the adsorption of both cationic and anionic dyes. The proposed mechanisms for adsorption of CR and MB dyes are presented in Figure . The nickel-based xerogel shows good adsorption capacity than copper-based xerogels, and this may be due to the larger surface area of the nickel-containing xerogel than copper-containing xerogels (Table S1 and Table S4, see the Supporting Information). Also, the porosity of the materials for the tetrapodal ligands is higher than that of the hexapodal ligands. It is to be noted that for the tetrapodal-ligand-based metallogels, the metal-to-ligand ratio was 2:1, whereas for the hexapodal ligand, the metal-to-ligand ratio was 3:1. In general, the adsorption capacities are higher for the material with higher surface area.

Figure 8

Schematic representation of the (a) congo red and (b) methylene blue dye adsorption mechanism. M stands for nickel and copper metal ions.

Furthermore, to confirm the interaction of the adsorbed dyes with the xerogel, FTIR analyses were performed (Figure S23, see the Supporting Information). In the case of CR, a 3470 cm–1 adsorption peak stands for the stretching of primary amine −N–H, which was slightly shifted and reduced after adsorption. The absorption band at 1611 cm–1 indicates the −N=N stretching and was also slightly shifted toward lower energy, whereas the peaks observed at 1226, 1124, and 1065 cm–1 correspond to the −S=O stretching and were shifted and also reduced after the adsorption. Moreover, the absorption peaks at 831, 750, and 696 cm–1 for the aromatic groups[73] in congo red were completely reduced after the adsorption. These results demonstrated that there are noncovalent interactions in between the congo red and Ni@MG1 that may be due to the participation of electrostatic interaction, π–π interaction, and hydrogen bonding.[74]

CO2 Uptake Analysis

The presence of the immobilized amide functional group and surface area of synthesized xerogels encourage us to study the CO2 adsorption. Before CO2 adsorption, we have activated the xerogel samples in the oven at 100 °C for 24 h, and then adsorption analysis were performed within the pressure range of 0 to 1 atm at 298 K. The study reveals that the synthesized materials exhibit excellent CO2 adsorption capacity in the family of metallogels (Figure , Table S5, see the Supporting Information). The adsorption and desorption paths of CO2 were the same in the case of nickel metallogels, and the capacities were lower than those of copper analogues. This may be attributed to the lower polarizing power of Ni(II) than Cu(II), which results from the Jahn–Teller distortion of the Cu(II) ions. The Cu@MG3 xerogel shows higher CO2 adsorption capacity than Cu@MG4 due to the higher surface area. However, in the copper-incorporated metallogel, the desorption path of CO2 does not follow the adsorption path; i.e., it shows hysteresis (Figure S24, see the Supporting Information). The results affirmed that on decreasing the external pressure, the adsorbed CO2 was not immediately released. The CO2 adsorption study indicates that the amide functionality, the choice of metal ion, and surface area have an effect on the overall capacity of the material.

Figure 9

CO2 adsorption capacity (mg/g) of (a) Ni@MG1 and Ni@MG2 and (b) Cu@MG3 and Cu@MG4.

Conclusions

Multifunctional metallogel synthesis from designed tetrapodal and hexapodal carboxylic acid-amide ligands, its water-soluble dye adsorption, and thermal decomposition to metal oxide have been accomplished. The synthesis can be tuned to prepare precursor metallogels for spinel oxides. The designed bifunctional carboxylic acid-amide based ligands have selective ligating properties toward metal ions. These variations in the ligation of the amide and the carboxylic acid are the cause of the formation of the spinel oxide instead of binary oxide. We have successfully synthesized ferrite (NiFe2O4 and CuFe2O4) spinels with high surface area (∼200–275 m2/g).

The equilibrium adsorption data of congo red and methylene blue on the xerogel-based adsorbents were estimated by Langmuir isotherm and Freundlich isotherm models. The Langmuir model shows a better fit, and the adsorption process conforms to the pseudo-second-order kinetic model. From our obtained results, we can say that the synthesized xerogels are promising candidates as adsorbents for congo red (497–1428 mg/g), methylene blue (458–980 mg/g), and carbon dioxide (22–40 mg/g, at 298 K and 1 atm).

Our work shows that these multifunctional metallogels can act as single-source precursors for spinel oxides and also as adsorbents for carbon dioxide and water-soluble synthetic organic dyes. Further work is necessary to see the versatility of the metallogel for the formation of spinel oxide and its potential application for the remediation and renewable energy application.

Experimental Section

Materials Used

1,3,5-Benzenetricarbonyl trichloride, 5-aminoisopthalic acid, and terephthaloyl chloride were bought from Sigma-Aldrich. 4-Dimethylaminopyridine and methylene blue were bought from Sisco Research Laboratories (SRL). Congo red was bought from Alfa Aesar, and all the metal salts and solvents were bought from Central Drug House (CDH) and utilized without further purification. The solvents N,N-dimethylformamide and N,N-dimethylacetamide (DMA) were used without further distillation.

Synthesis of LMWG1 and LMWG2 Ligands

The LMWG1 ligand was made with the help of the reported literature with minor modification.[75] In 15 mL of anhydrous N,N-dimethylacetamide, 5-aminoisophthalic acid (9.2 mmol, 1.66 g) and 4-dimethylaminopyridine (0.12 mmol, 0.014 g) were dissolved under a N2 atmosphere. Then terephthaloyl chloride (4.5 mmol, 0.913 g) was solubilized in 10 mL of anhydrous DMA and added drop wise to it. Under a N2 gas atmosphere, the mixture was stirred for 48 h. To get the precipitate of the product and to solubilize the excess 4-dimethylaminopyridine and 5-aminoisophthalic acid, 50 mL of 5% HCl was added. The LMWG2 ligand was sythesized with some modification of the reported literature.[76] We have used 1,3,5-benzenetricarbonyl trichloride (3 mmol, 0.79 g) in place of terephthaloyl chloride, and all the remaining things were the same. The obtained product was purified by washing with 50 mL of 5% HCl followed by 20 mL of water, 20 mL of methanol, and finally with ether. For LMWG1, 1H NMR (DMSO-d6, δ ppm): 13.32 (broad peak, 4H, COOH), 10.76 (2H, s, CONH), 8.70 (4H, s, ArH), 8.24 (2H, s, ArH), 8.17 (4H, s, ArH), yield = 1.89 g (85.32%), and for LMWG2, 1H NMR (DMSO-d6): δ 13.38 (broad peak, 6H, s, COOH), 10.95 (3H, s, CONH), 8.84 (3H, s, ArH), 8.73 (6H, s, ArH, 2), 8.26 (3H, s, ArH, 3), yield = 1.93 g (92.34%). The spectra of LMWG1 and LMWG2 ligands are shown in Figures S1–S3 (see the Supporting Information).

Metallogel Synthesis

In this work, we have adopted the reported synthetic technique for metallogel preparation.[77] In a typical synthesis, we have prepared the 0.1 mmol/mL solution by dissolving LMWG1 in DMA solvent. 0.5 mL (0.1 mmol/mL) of LMWG1 was placed separately in different vials, and 0.05–2.0 equiv. of nickel and copper acetate concerning the ligand LMWG1 concentration was also solublized in dimethylformamide and put separately in different vials. In the final step, the dimethylformamide solution of nickel and copper acetate was added to the solution of ligand LMWG1 so that the final volume was 1 mL in each glass vial. Then, after 10 min, by the glass vial inversion method, the formation of the metallogel was confirmed at room temperature. The same procedure was followed for the metallogels derived from ligand LMWG2 except that 0.05–3.0 equiv. of nickel and copper acetate was taken in different glass vials with respect to the ligand LMWG2 (0.1 mmol/mL) concentration (Figure ). The minimum gelation concentration for the LMWG1 ligand and metal was found to be in a 1:2 ratio, but for LMWG2, it was in a 1:3 ratio.

Spinel Precursor Metallogel Synthesis

To make a spinel precursor metallogel, 0.5 mL of 0.1 M LMWG2 was taken in glass vials, and 0.25 mL from each of the dimethylformamide solution of 0.2 M Ni(CH3COO)2·4H2O and 0.4 M Fe(NO3)3·9H2O were added into the 0.1 M LMWG2-containing glass vial. Then, after 30 min, formation of spinel precursor metallogels was confirmed through the glass vial inversion method. All the remaining things were the same to make the copper-containing spinel precursor metallogel except that 0.2 M Ni(CH3COO)2·4H2O was replaced with the 0.2 M Cu(CH3COO)2·H2O metal salt solution (Figure ).

Spinel Oxide Synthesis from the Precursor Metallogel

We have taken the dried spinel precursor metallogels Ni:2Fe@LMWG2 and Cu:2Fe@LMWG2 and heated them in a tube furnace at 500 °C for 3 h (Figure ).

Field-Emission Scanning Electron Microscope (FESEM)

The morphology of xerogels was analyzed using a field emission scanning electron microscope (ZEISS GEMINISem 500 with an energy-dispersive X-ray spectroscopy detector).

Transmission Electron Microscopy (TEM)

Metallogel samples were also analyzed using a transmission electron microscope (JEOL-JEM 2100) instrument operating at 120 kV.

Rheological Measurements

The rheological studies for all the metallogels were carried out using an AntonPaar (MCR 302) modulated compact rheometer. Samples were scanned at room temperature from 0.1 to 20 Hz on a parallel plate (9 mm diameter) at 1% strain to obtain the gain modulus (G′) and loss modulus (G″).

BET Surface Area Analysis

The N2 gas sorption studies were carried out to get the Brunauer–Emmett–Teller (BET) surface area using a Quantachrome Autosorb iQ2 analyzer. Generally, around 100 mg of xerogels was taken in the sample holder 9 mm in width. Before performing the experiment, xerogels were kept in a vacuum oven at 110 °C for 24 h; after that, samples were degassed at 120 °C for 6 to 7 h.

Thermogravimetric Analysis (TGA)

Thermogravimetric analysis was performed using an SDT Q600 (TA Instruments). The scan rate was 10 °C/min under nitrogen, with a flow rate of 100 mL/min.

Powder X-ray Diffraction Analysis (PXRD)

PXRD analysis was conducted using a Rigaku MiniFlex 600 diffractometer with Cu Kα radiation (λ = 1.5418 Å).

Fourier Transform Infrared (FTIR) Spectra

FTIR spectra of xerogel samples were measured in the range of 4000 to 400 cm–1 using a PerkinElmer Spectrum 400 in KBr mode.

UV–vis Spectra

A spectrophotometer (Shimadzu UV2500) recorded the UV–vis spectra using 1 cm path-length quartz cuvettes.

Adsorption Experiments

All adsorption experiments of the aqueous solutions of congo red and methylene blue were carried out at room temperature. 4 mg of adsorbent was dispersed in solutions (4 mL) of various concentrations under stirring until equilibrium was reached. 1 mL of the aliquot was taken out and then centrifuged for 3 min to measure the concentration of the supernatant with the help of a UV–vis spectrophotometer. To calculate adsorption capacity at equilibrium qe (mg/g), the following equation was usedwhere C0 (mg/L) stands for initial concentration, Ce (mg/L) stands for equilibrium concentration, V (L) represents the total volume of dye solution, and m (g) represents the adsorbent weight. To perform adsorption kinetics studies, 20 mg of each Ni@MG1, Ni@MG2, Cu@MG3, and Cu@MG4 materials were added to 20 mL of an aqueous solution of 10 mg/L MB and CR dyes, and then 1 mL of aliquots was taken out at certain time intervals to measure the concentration of the supernatant. To calculate the amount of dye adsorbed at a particular time t, (q), the following equation was used