Introduction

Selective and precise detection of explosive substances have gained much attention due to their obvious value to security efforts. Most commonly, nitroaromatic compounds such as 2,4,6-trinitrotoluene (TNT), 1,3,5-trinitrobenzene (TNB), 1,3-dinitrobenzene (DNB), 2,4-dinitrotoluene (DNT), and others [1] are used as ingredients in explosives and hence numerous sensing methods have been developed for the selective and accurate detection of nitroaromatics [2, 3]. However, existing methods such as gas chromatography coupled with mass spectrometry, surface enhanced Raman spectroscopy and nuclear quadrupole resonance, etc. [4], requires highly sophisticated instrumentation, a trained technician, lengthy sample preparation and time consuming procedures. Alternatively, turn-off or turn-on fluorescence based chemosensor methods for the detection of explosives have several advantages, such as high sensitivity, selectivity and ease of handling, when compared with the other conventional analytical methods. A variety of florescence sensor materials, such as transition metal complexes [5], supramolecular polymers [6], carbon nanotubes [7], quantum dots [8], dendrimers [8] and long chain conjugated polymer etc. [8] were more often used for the detection of nitroaromatics. On the other hand, stability, toxicity, lower sensitivity, lower selectivity and non-biodegradability of existing fluorescence sensor materials limit their application.

Metal-Organic frameworks (MOFs) are a class of porous hybrid materials constructed by organic linkers, transitions and non-transition metal cluster. Their distinctive properties such as tunable architecture, multi-functional sites, high porosity, large surface area etc. widens their applications in a variety of ways [9–12]. Among these, multi-functional sites including catalytically active sites, open metal sites and photo-responsive sites look promising for application in fluorescence chemical sensors [13]. Furthermore, different fluorescence emissions of MOFs can be synthesized by changing the organic linkers. Various novel metal-organic frameworks have been utilized as a fluorescence sensor material for the detection of nitroaromatic compounds. Qu et al. synthesized a Cd-based metal-organic framework for nitrobenzene detection [14]. Gole et al. fabricated a novel Zn-based electron deficient MOF for the selective fluorescence detection of nitroaromatic compounds [15]. Sun and his research group developed pentiptycene-based luminescent Cu (II) MOF for a highly sensitive and selective detection of electron deficient compounds [16]. Moreover, Su and co-workers synthesized the Cd–based 2D stable MOF for detection of -NO2 groups containing electron deficient species [17]. Generally, the sensing phenomena of fluorescent MOFs followed the oxidative quenching mechanism. Moreover, the fluorescent MOFs act as electron donors and nitroaromatics act as electron acceptors due to the presence of the electron-withdrawing -NO2 group.

A few studies reported that amine functionalized MOFs have excellent selectivity and sensitivity for the detection of nitroaromatic compounds compared to other MOFs, due to the Lewis basic sites of amine [18]. Lewis basic amine functionality of MOFs can lead to the binding of analyte molecules through hydrogen bonding and induce electron transfer between MOFs and analytes. In this report, amine functionalized Fe-based metal-organic framework NH2-MIL-101(Fe) was synthesized using 2-aminoterephthalic acid as an organic linker by means of a simple hydrothermal method. Notably, the photoluminescence quenching studies demonstrated that the prepared MOF is highly selective for electron deficient nitroaromatic compounds such as nitrobenzene, dinitrobenzene, 4-nitrophenol and 4-nitrotoluene. A quenching mechanism between guest and Lewis basic amine functionalities of MOF is also proposed. To the best of our knowledge, no reports exist concerning the ‘turn off’ fluorescence detection of nitroaromatic compounds using an NH2-MIL-101(Fe) metal-organic framework.

Experimental section

Chemical requirements

Nitroaromatic compounds such as nitrobenzene (NB), nitrotoluene (4-NT), nitrophenol (4-NP), 1, 3-dinitrobenzene (1,3-DNB) and 2-amino terephthalic acid were purchased from Alfa Aesar (https://www.alfa.com/en/catalog/A10293/). All the solvents were purchased from SRL chemicals Pvt. Ltd. India (http://www.srlchem.com/) and used without any further purification. Metal salts, such as CaCl2, Co (NO3)2, FeCl3, FeCl2, KCl, MnCl2, NaCl and NiCl2 were purchased form Sigma Aldrich (https://www.sigmaaldrich.com/), Loba Chemicals, India and SRL chemicals Pvt. Ltd. India (http://www.srlchem.com/).

Synthesis of NH2-MIL-101(Fe) metal-organic framework

The amine functionalized-NH2-MIL-101(Fe) was synthesized by following a previously reported method [19] with some modifications (Scheme 1). In a typical procedure, 1.6 g of FeCl3.H2O and 1.4 g of 2-aminoterephthalic acid were dissolved in DMF and the resulting solution was transferred into a 100 mL Teflon-lined autoclave and then heated at 150 °C for 12 h. After the hydrothermal treatment, the brown amine functionalized metal-organic framework was separated by centrifugation and washed with DMF and methanol and dried at 60 °C overnight. Furthermore, the resultant powder was suspended in 100 mL of methanol and stirred for 2 days to remove the residual DMF. Finally, the MOF was separated and dried at 150 °C for 12 h.

Scheme 1
scheme 1

A schematic illustration of NH2-MIL-101(Fe) MOF synthesis

Full size image

Fluorescence studies

For the fluorescence titration experiment, 5 mg of NH2-MIL-101(Fe) MOF was dispersed in 10 mL de-ionized water and 2 mL was poured into a quartz cuvette. The photoluminescence was measured by excitation at 360 nm (λex = 360 nm) and emission was observed at 430 nm. Similarly, all the photoluminescence titration experiments were carried out through the gradual incremental increase of analyte concentration in the ppm scale. The homogeneity of the reaction solution was maintained by magnetic stirring.

Choice of materials

Metal − organic frameworks (MOFs) are a class of porous materials composed of metal nodes/clusters and organic linkers. A variety of luminescent metal-organic frameworks have been used for sensing applications. Compared to quantum dots, conjugated conductive polymers, carbon nanotubes, MOFs are considered to be a good sensing platform in analytical science due to their unique properties such as high surface area, versatile framework compositions, and exposed active sites. Fe-based MOF are considered as simple material for the nitroaromatic compounds sensing. The synthesis process and precursor which is used for Fe-MOF synthesis is cost effective compared to all other MOFs, such as those based on Ti, La and Cd. To date, many strategies have been used in the attempt to apply MOFs in the fluorescent detection of nitroaromatic compounds. However, most of the reported MOFs are used for fluorescent detection in the organic or gas phases as MOFs have poor hydrolytic stability. In this study, an iron-based MOF was synthesized by a simple hydrothermal method and, importantly, the prepared MOF was used for the detection of nitroaromatic compounds in the aqueous-phase.

In addition, unlike other MOFs, light absorption of amine functionalized MOFs lies in the visible region. The organic linker such as 2- amino terephthalic acid can effectively harvest the light and transfer the photo-generated electrons to the Fe-oxo cluster. Moreover, the amine group of MOF facilitates H-bond formation with nitroaromatic compounds and thereby enhances quenching. Thus, a facile synthesis process, greater light absorption, the possibility of H-bond formation, water stability and low-cost precursors, makes NH2-MIL-101(Fe) a potential fluorescent probe for sensing applications.

Characterization techniques

Powder X-ray diffraction (PXRD) patterns were recorded with a PAN analytical X’pert powder diffractometer using Cu Kα radiation. The morphology of the synthesized MOFs was studied using field emission-scanning electron microscopy (FEI Quanta FEG 200 HR-SEM). Fourier Transform Infrared (FTIR) spectra were obtained using PerkinElmer (FTIR) spectrometer (Perkin-Elmer, USA). The photoluminescence studies were carried out using a spectrofluorometer FluoroMax (Horiba. Scientific Instrument Inc., USA). The time resolved spectroscopy and TGA measurements were recorded using DAS6 HORIBA instrument from Jobin Yvon - Japan and NETZSCH STA 449 F3 from Jupiter-India.

Results and discussion

The phase purity and crystalline nature of synthesized NH2-MIL-101(Fe) MOFs was analyzed by X-ray powder diffraction (XRD). Figure S1 shows that the characteristic diffraction peaks of the prepared samples are analogous to those previously reported for NH2-MIL-101(Fe) and no peaks corresponding to Fe2O3 were observed [19]. These peaks indicate the formation of phase pure NH2-MIL-101(Fe) MOF without any oxide impurities [20]. Moreover, the fact that there is no difference between the synthesized and the activated sample (i.e., the synthesized sample stirred in methanol for 3 days to remove the DMF solvent) demonstrates that, during the activation process, the structure of MOF is not changed or affected by the long period of mechanical stirring (Fig. S1).

In order to further confirm the formation and bonding of NH2-MIL-101(Fe) MOF, FTIR analysis was carried out. The results are depicted in supplementary information (Fig. S2). The characteristic stretching and bending vibrations of O-C-O bonds in the NH2-MIL-101(Fe) appeared in the region of 1600–1400 cm−1 [21]. The peaks appearing at 3249 and 3329 cm−1 can be attributed to N-H stretching vibrations. In addition, the aromatic ring C = C bond frequency was observed in the region of 1500 to 1600 cm−1 [22]. The peak identified at 545 cm−1 can mainly be attributed to Fe-O bonding of MOF (Fig. S2 inset) [19]. These FTIR results reveal the formation of metal-organic framework with 2-aminoterephthalic acid utilized as an organic linker, similar to previous reports [23]. Moreover, the FTIR spectra of synthesized MOF are analogous to those of activated NH2-MIL-101(Fe) MOF; indicating that the structure of MOF did not collapse during the simulation process and remained the same.

The morphology and elemental compositions of NH2-MIL-101(Fe) MOF were determined by means of FESEM and EDX analysis. FESEM images with different magnifications of MOF are shown in Fig. S3a–d. As showed in Fig. S3, the synthesized NH2-MIL-101(Fe) MOF shows hexagonal micro-spindle morphology with uniform particle distribution. Based on statistical histogram analysis (Fig. S3) the average particle size of NH2-MIL-101(Fe) was calculated to be 42.9 nm. Moreover, the elemental composition result shows the presence of elements including C, N, O and Fe. These observations clearly indicates that the MOF particles formed are discrete and not in an aggregated state, and also indicate the purity of the NH2-MIL-101(Fe) MOF prepared.

The absorption and emission spectra of NH2-MIL-101(Fe) MOF and organic linker (2-aminoterephthalic acid) were recorded and compared in order to further confirm MOF formation. The UV-visible absorption and emission spectra of both the free linker and NH2-MIL-101(Fe) MOF are depicted in Fig.1a, b. A sharp absorption peak appearing at 350 nm for the free organic linker may be due to π-π* and n- π* transitions. In contast, the absorption spectrum of NH2-MIL-101(Fe) completely deviated from that of the free organic linker and a broad absorption peak was observed. The broadening of UV-vis spectra may have been caused by greater electron transfer between the Fe-oxo cluster and 2- aminoterephthalic acid. In addition, along with π –π*, n –π* transition of organic linker, LCCT transition may also have occurred due to strong interactions between organic linkers and Fe-oxo clusters, resulting in a broadening of the the absorption spectra of NH2-MIL-101(Fe). This suggests that organic linkers interact strongly with Fe-oxo clusters and take part in MOF formation. Similarly, the PL spectra of the free organic linker and NH2-MIL-101(Fe) MOF were recorded at room temperature (excitation wavelength (λex) = 360 nm). The free organic linker and NH2-MIL-101(Fe) MOF exhibit emission peaks at 438 and 449 nm, respectively. The emission peak of NH2-MIL-101(Fe) MOF shifted to a higher wavelength compared to the linker. It is worth mentioning that the red shift in the emission peak reflects strong coordination between the metal cluster and the organic linker [24].

Fig. 1
figure 1

a UV-vis absorption spectra and b photoluminescence (PL) spectra (λex = 360 nm) of free organic linker and NH2-MIL-101(Fe) MOF

Full size image

Fluorescence sensing of nitroaromatic compounds with NH2-MIL-101 (Fe) MOF

In order to analyze the influence of solvents on fluorescence quenching of synthesized MOF, various solvents were used to record photoluminescence. As can be seen from Fig. 2, the fluorescence intensity of MOF depended considerably on the solvents in question. The relative order of % of quenching of luminescence with respect to different solvents is NB (96%) > CB (30%) > Toluene (29%) > Ethanol (17%) > IPA (14%) > DMF (9%) > water (0%). A drastic fluorescence quenching was observed upon the addition of nitrobenzene. Thus, the physical interaction of solvent molecules with MOF plays the vital role in influencing fluorescence intensity. This observation clearly demonstrated that in the case of MOF, the fluorescence intensity was “turned off” in the presence of electron deficient nitroaromatic compounds. To evaluate the selective sensing behavior of MOF towards nitrobenzene (NB), a series of fluorescence titrations were carried out with gradually increasing NB concentration whilst monitoring the emission response. The luminescence intensity gradually decreased with increasing NB concentration. However, 100% quenching was observed with 120 ppm of NB. This indicates that NB is closely attached to the MOF surface and facilitates much host-guest interaction [15, 25]. Hence, electron transfer reactions between electron donating MOF and electron deficient NB can take place during excitation, resulting in fluorescence quenching.

Fig. 2
figure 2

a Fluorescence spectra of NH2-MIL-101(Fe) MOF in the presence of various solvents (λex was 360 nm), b Plot of fluorescence quenching % with respect to different solvents obtained for NH2-MIL-101(Fe) MOF

Full size image
Fig. 3
figure 3

Fluorescence spectra obtained for the NH2-MIL-101(Fe) MOF upon addition of various concentrations of a nitrobenzene (a) nitrobenzene, (b) 4-nitrophenol, (c) 4-nitrotoluene and (d) 1,3-dinitrobenzene (λex was 360 nm)

Full size image
Fig. 4
figure 4

Stern–Volmer plots for the quenching of the luminescence of NH2-MIL-101(Fe) MOF upon addition of various concentrations of (a) nitrobenzene, (b) 4-nitrophenol, (c) 4-nitrotoluene and (d) 1,3-dinitrobenzene (λex was 360 nm)

Full size image
Fig. 5
figure 5

Lifetime decay profile of fluorescence spectra obtained for the NH2-MIL-101(Fe) MOF upon addition of various concentration of 4-NP

Full size image
Fig. 6
figure 6

a PL responses of NH2-MIL-101(Fe) MOF upon the addition of 1000 ppm of Ca2+, Co2+, Fe2+, Fe3+, K+, Mn2+ Na+ and Ni2; b 50 μM of H2O2 and aq NH3 (all the experiments were carried out three times (n = 3) and Error bars indicate standard deviations)

Full size image

Based on the above observation, the quenching of fluorescence is mainly attributed to the electron deficient nature of nitroaromatic compounds. Hence, the detection ability of NH2-MIL-101(Fe) MOF towards the electron deficient nitroaromatic compounds 4-nitrophenol, 1,3-dinitrobenzene and 4-nitrotoluene was investigated with identical experimental conditions. The results are given in Fig. 3. Interestingly, all nitroaromatic compounds showed significant quenching in fluorescence luminescence intensity. The order of quenching was as follows: 4-nitrophenol >4-nitrotoluene >1,3-dinitrobenzene > nitrobenzene. Complete fluorescence quenching was observed for 4-nitrophenol at 50 ppm. Whereas, 4-nitrotoluene, 1,3-dinitrobenzene and nitrobenzene showed complete quenching at 60 ppm, 100 ppm and 120 ppm respectively. These results clearly indicate that quenching intensity is not only dependent on the electron deficient nature of aromatic compounds but also on host-guest interactions. Among the different nitro aromatics, 4-nitrophenol shows 100% fluorescence quenching even at the low concentration of 50 ppm. This can be explained by the possibility for H-bond formation between guest (MOF) and host molecules (nitroaromatics) (Scheme 2) [26]. The H-atom present in the phenolic group of 4-nitrophenol can form an H-bond with NH2-MIL-101(Fe) MOF, resulting in rapid electron transfer between guest and host molecules, thereby decreasing the 60 ppm intensity. Besides 4-nitrophenol, 4-nitrotoluene also exhibits complete fluorescence quenching. The limit of detection for all the analytes was calculated from calibration plots (ESM; Fig. S6–S9) by means of the following formula [27]:

$$ \mathrm{LoD}\ \left(\mathrm{ppm}\right)=3\sigma /\mathrm{m}, $$
Scheme 2
scheme 2

Mechanism for the interaction for nitroaromatic compound with fluorescent NH2-MIL-101 (Fe) MOF

Full size image

where σ is standard deviation and m is slope. The LoDs of NB, 4-NP, 4-NT and 1,3-NB were calculated to be 32.2 ppm, 17.3 ppm, 9.8 ppm and 11.5 ppm, respectively, which further supports the high sensitivity of MOF towards electron deficient nitroaromatic compounds.

For the better understanding and quantification of the fluorescence quenching efficiency, the Stern-Volmer graph was plotted for all quenching studies (Fig. 4) using the equation:

$$ \left({\mathrm{F}}_0/\mathrm{F}\right)={\mathrm{K}}_{\mathrm{sv}}\left[\mathrm{A}\right]+1 $$

Where F0 and F are the fluorescence intensities before and after addition of the analyte, respectively, Ksv is the quenching constant (m−1), [A] is the concentration of the analyte. The Stern-Volmer plots for all analytes are linear at lower concentration levels. This linearity of Stern-Volmer plots reveals that the quenching occurred due to the static energy transfer process and not through dynamic mechanisms [28, 29]. The fluorescence quenching occurred due to photo-induced electron transfer (PET) between guest and host molecules. When, NH2-MIL-101(Fe) MOF was irradiated with light, the photo-excited electrons migrated to the lowest unoccupied molecular orbital (LUMO) of nitroaromatic compounds which is the low-lying π*-orbital and is stabilized by the -NO2 group through conjugation [30]. This stimulates electron transfer from the CB of MOF to the LUMO level of the analytes, thereby quenching the fluorescence intensity during excitation.

In order to confirm the quenching mechanism for the detection of 4-NP, the fluorescence lifetime measurement was carried out by monitoring fluorescence life time decay with the gradual addition of 4-NP to the MOF (Fig. 5). As can be seen from Fig. 5, the gradual addition of 4-NP to MOF resulted in only nominal change in the lifetime decay values. The average life time (τ) values of excited state carriers of MOF suspension at different concentration of 4-NP added MOF suspensions 0, 20, 40 and 60 μL of 4-NP were found to be 2.3, 2.2, 2.2 and 1.7 ns, respectively. The average life time of excited state carriers essentially remain unchanged after and before the addition of 4-NP. Thus, no change in the lifetime values of carriers strongly suggests that the fluorescence quenching follows static process rather than dynamic quenching process [31, 32].

Interference study

In order to test the selectivity of this method, interference studies were carried out with H2O2, NH3 and metal ions such as Ca2+, Co2+, Fe2+, Fe3+, K+, Mn2+ Na+ and Ni2+. The results are shown in Fig. 6. It can observed that no significant quenching was observed when adding high concentrations (50 μL: 1000 ppm) of metal ions, clearly indicating the good selectivity of MOF for the sensing process. Similarly, with H2O2 and NH3 the fluorescence intensity also remains unchanged. These results further confirm the selectivity of NH2-MIL-101(Fe) MOF. The efficiency of the present method was compared with that of the recently reported fluorescence based probes for nitroaromatic compounds. Table 1 shows the Ksv (Stern-Volmer constant) of each reported fluorescence detection method.

Table 1 An overview on recently reported nanomaterial-based fluorometric methods for the determination of nitroaromatic compounds
Full size table

Conclusion

In this study, we have reported on the detection of nitroaromatic compounds with NH2-MIL-101(Fe) MOF based on fluorescence quenching. The quenching in the fluorescence is mainly attributed to photo-induced electron transfer between guest and host molecules. Moreover, rapid quenching was observed for 4-nitrophenol due to inter-molecular H-bond formation; facilitating the rapid electron transfer reaction. The present work will stimulate interest towards developing Fe-based simple MOF materials for the selective and sensitive detection of nitroaromatics.