Metal-organic framework-based nanomaterials as opto-electrochemical sensors for the detection of antibiotics and hormones: A review

Introduction

Pharmaceuticals, in particular antibiotics, have become ground-breaking drugs in the medical field for treating a variety of infectious diseases in both humans and livestock. Antibiotics are widely utilized in animal feed to enhance growth and production, which contaminates human food products. Particularly inexpensive quinolones, a class of synthetic antibiotics, are frequently used to treat bacterial infections in animals, particularly fish, cattle, and poultry. However, the presence of these antibiotic residues in foods derived from animals, such as eggs, milk, meat, and fats, can have a number of negative impacts.

Approximately 70 tons of generic and 50 tons of proprietary quinolones are consumed annually in the United States, Japan, South Korea, and the European Union (EU). Recently in China, annual quinolone usage in animal feeds was thought to be in the range of 500 tons, and human consumption was around 1400 tons [1]. Due to their complicated structural makeup, the majority of antibiotics are eliminated unaltered in urine and faeces, which ultimately contaminate natural water sources and soil [2-6]. In environmental samples, antibiotics are currently being found at levels between nanograms and micrograms per litre. Antibiotic abuse and overuse also have serious consequences for the environment and human health. The rise of zombie-like antibiotic-resistant bacteria, which have been linked to severe allergic reactions in people, has been caused in part by the considerable fraction of microbes that have been resistant to a few particular antibiotics [4-7]. Additionally, excessive antibiotic residues in the environment might constitute a major concern by producing conditions including gonorrhoea, tuberculosis, and pneumonia, which complicate their treatment [7,8].

Furthermore, the environment is becoming more and more contaminated with both artificial steroids (e.g., 17α-ethinylestradiol, gestodene, trimegestone, antiandrogens, and synthetic estradiol) and natural hormones (e.g., testosterone, progesterone, estradiol, triiodothyronine, thyroxine, and melatonin) [9-11]. Synthetic hormones are often used to accelerate plant and poultry growth, as well as to boost the production of milk in cattle and other animals [9]. In the world today, the use of synthetic hormones for oral contraception, bodybuilding, and weightlifting has increased at an unprecedented rate. These synthetic steroid hormones are endocrine disruptor substances (EDSs) because they have the potential to interfere with physiological functions and negatively impact both the health of the organism and its offspring [12]. Figure 1 shows the several sources and routes by which hormone and antibiotic residues enter the environment.

EDSs are of concern because low quantities can have detrimental effects on the endocrine system and may change an animal’s rate of growth, development, and reproduction [9,12-14]. Even though they are only present in trace amounts in the environment, pharmaceutical substances such as antibiotics and hormones are considered emerging pollutants. Antibiotic overuse and misuse that leads to antimicrobial resistance pose an urgent threat to global public health, killing more than 36,000 people in the United States and being linked to over five million deaths globally in 2019, according to the USA Centers for Disease Control and Prevention report [15].

This is also in line with the World Health Organization’s assertion that “antibiotic resistance” poses a significant financial and societal risk to public health [16]. In fact, according to a 2015 WHO projection, if the current trend in the abuse of antibiotics persists, 300 million people will die prematurely globally over the next 28 years [17]. Considering these environmental and health concerns, a number of regulatory bodies and nations, including the EU, have prohibited the use of chemicals and pharmaceuticals with a hormonal action for the promotion of growth or fattening of livestock through a number of directives due to the endocrine disruptive potential of steroids and natural hormones. Since the COVID-19 pandemic, there have been allegedly higher levels of antibiotics and hormones found in effluents due to increased global antibiotic consumption and the usage of various hormones that promote health [18,19].

Therefore, research efforts have been concentrated on monitoring and detecting antibiotics and hormones in environmental, clinical, food, and biological samples due to the bioaccumulation, persistence, ecological, and health risks associated with them. However, to detect these emerging contaminants, analytical techniques that are sensitive and selective enough must be developed due to their incredibly low concentrations [2,6-9]. Colourimetry, chromatography, enzyme-linked immunoassay (ELISA), radioimmunoassay (RIA), surface-enhanced Raman spectroscopy (SERS), and capillary electrophoresis are common analytical techniques used to qualitatively or quantitatively determine pharmaceuticals in various matrices because they are sensitive (Figure 2), have a significant tolerable limit of detection (LOD), and are selective in many cases. However, they have a number of shortcomings.

For example, in order to increase the sensitivity of the established immunoassays, Mitchell and Lowe [20] and Wu et al. [21] used the ELISA technique for the analysis of testosterone and progesterone, respectively. Even though this sensitive ELISA approach provided the necessary sensitivity for hormone detection in biological samples, it had a number of drawbacks, including limited specificity, high cost, cross-reactivity, and significant kit variability. The conventional ELISA method necessitates a laborious, multistep separation procedure. A fluorescence polarisation immunoassay, which does not require a separation step, is one development that has been made to get around this restriction [22]. Furthermore, chromatographic techniques such as thin-layer chromatography (TLC), liquid chromatography-coupled mass spectrometry (LC–MS), high-performance liquid chromatography (HPLC), and gas chromatography coupled with mass spectrometry (GC–MS) have demonstrated greater selectivity and lower variability when compared to immunoassays for the measurement of antibiotics and steroid hormones in various complex samples [22-26].

Despite their sensitivity and reproducibility, chromatographic techniques require expensive, bulky equipment, excessive amounts of solvents, extended times for sample preparation and extraction stages, and a variety of stationary phases, making them more difficult to use in labs with limited resources. Additionally, the mobile phase utilized in the separation of antibiotics and hormones affects the chromatography’s ability to detect substances [23]. The advantages of SERS are its rapid detection time, affordable detection cost, and ease of use. It also has a number of drawbacks, such as the necessity for technical experts with the necessary training, the limited selectivity and reusability of substrates, and the degradation of substrates with time, which reduces the signal.

Several portable electrochemical and optical (chemiluminescence, fluorescence, and electrochemiluminescence) sensors have been developed for antibiotic and hormone detection analyses to address the limitations of the standard analytical methods. These sensors’ advantages include high sensitivity, low cost, rapid analysis time, simplified operation without the need for complicated and expensive instrumentation, in situ analyte monitoring, and potential miniaturization. Portability, miniaturisation, and fast signal responses are just a few of the breakthroughs in sensor design made possible by nanomaterials.

Nanomaterials are becoming a key component of the analytical procedures required for pharmaceutical, environmental, food safety, and health analyses. They have exceptional physico-chemical and opto-electronic properties, a high surface area-to-volume ratio, and their surfaces are easy to functionalize. Additionally, compared to their bulk counterparts, nanomaterials are particularly sensitive to changes in surface chemistry, enabling nanosensors to achieve extremely low detection limits. Numerous nanomaterials shown in Figure 3 have different functionalities, including high conductivity, good catalytic activity, and optical and plasmonic properties, making them attractive candidates for opto-electrochemical sensing platforms [1,7-10,27-33]. Reliable and sensitive nanomaterial-enabled portable sensors can be quickly deployed to resource-constrained sites to offer rapid and cost-effective monitoring and detection of different analytes without the need for bulky, expensive instrumentation or highly skilled technical experts.

Given the many benefits of nanomaterials, it is anticipated that the incorporation of nanotechnology in sensing would lead to the development of a diagnostic tool for detecting hormone and antibiotic residues in diverse single or multiple matrices. Other researchers have published a number of nanomaterial-based (quantum dots, carbon-based, and metal-based) sensors for the detection of various analytes [1,2,7,9,27-38].

A review paper that offers a comprehensive analysis of current developments based on metal-organic framework (MOF) opto-electrochemical nanosensors for the detection of hormones and antibiotics is still missing, though. This review focuses on a variety of sensing applications that use MOFs as well as the synergistic mechanisms of MOF hybrids or composites that improve sensing performance. It provides a brief summary and discussion of MOF synthesis methods. This review begins by summarising the importance of monitoring and detecting hormones and antibiotics in various matrices. The traditional analytical methods employed for detecting pharmaceutical residues are then briefly described. Comprehensive reviews and discussions were conducted on the mechanisms and performance of opto-electrochemical platforms using MOFs and their hybrids as opto-electrochemical sensors for the efficient detection of hormones and antibiotics in various matrices. Finally, challenges and future perspectives are presented. Therefore, it is anticipated that the concepts presented in this review will stimulate further investigation into MOF-based materials for opto-electrochemical detection of various other analytes (explosives, viruses, and various other emerging contaminants).

Review

Opto-electrochemical sensors: mechanisms and challenges

It is crucial to assess the performance of sensors during development using standard metrics such as selectivity, sensitivity, the limit of detection (LOD), and response time. Selectivity refers to a sensor’s capacity to respond to a narrow range of target analytes while resisting interaction with other non-target species; this is often used to determine the accuracy of the results. The LOD of an analyte is the lowest concentration at which it can be consistently detected by an analytical procedure.

A measurable signal that can be statistically distinguished from the background or a blank signal must be produced by this concentration (C_LOD) [39]. Sensitivity (S) is defined as the ability to change the measured signals (optical or electrical) in response to a change in the amount of analyte. This has a close connection to a sensor’s LOD; the LOD decreases as the sensitivity increases. Generally, it is acknowledged that the C_LOD can be stated as a function of S and S_b (the standard deviation of a group of blank signals produced via consecutive tests devoid of an analyte): LOD = 3.3·S_b/S where S = slope of the calibration curve. Also, the slope of the calibration curve can be used to define sensor sensitivity (Figure 4).

The minimum analyte concentration that can be reliably and precisely quantified is expressed by the term “limit of quantification” (LOQ). For estimation, a level of 10·S_b/S is recommended. The kinetics of both chemical recognition and signal transduction are correlated with sensor response time. An ideal sensor, when taking into account the aforementioned key parameters, should be specific for the target analytes, sensitive to changes in analyte concentrations, have a rapid response time, have a long lifespan of at least several months, and be small (miniaturised) with the potential for low-cost manufacturing.

Optical sensing: fluorescent sensors

Optical sensors are light-based analytical devices based on the alteration in the measurement of light wavelengths following the interaction of the analyte with the molecular recognition element (Figure 5). Biorecognition elements and signal transducers (chemiluminescence, interferometry, surface plasmon resonance, luminescence, colourimetry, or surface-enhanced Raman spectroscopy), are the key components of an optical sensor. Analyte concentration, existence, and other relevant physical attributes are determined from the optical signals. In recent years, interest in optical methods of hormone and antibiotic detection has grown due to their rapid response times, simplicity of use, and high sensitivity [32-38]. In contrast to other techniques such as colourimetry and surface plasmon resonance, this review will exclusively concentrate on the luminescence (particularly fluorescence) sensing mechanism.

The basis for optical sensing is the luminescence mechanism, which is the spontaneous emission in the optical range of ultraviolet, visible, or infrared light by a substance without being induced by heat. The emission is caused by electrons transitioning from higher-energy molecular orbitals to lower-energy ones, typically the ground state or the lowest empty molecular orbitals. Luminescence may be caused by intrinsic defects, a particular moiety within the compound (metal or ligand), impurity-induced defects, or it may exist in pure crystals or molecules. According to the manner of the substance excitation, several distinct forms of luminescence are differentiated, as shown in Figure 6. A molecule, nanostructure, or atom must be able to absorb light radiation, resulting in electronic excitation, for photoluminescence to occur, whether it be fluorescence or phosphorescence. The molecule-bound electron in the fluorescence mechanism absorbs a photon and is activated after the analyte interacts with the molecular recognition element. The transition from the ground state (S₀) to the excited state (S_n, n = 1, 2, ...) occurs in femtoseconds [39].

Depending on the wavelength of the absorbed photon, the excited state of the electron may result in the electron occupying any one of a number of possible vibrational levels. In a rapid (nano- to microseconds) transition from the lowest excited state (S₁) to the ground state (S₀), the excited electrons relax radiatively through a combination of steps. It is conceivable for the emission to relax to a range of vibrational levels of the S₀, which gives rise to a bandwidth of potential photon wavelengths. The electron will have lost some of the initial excitation energy through vibrational relaxation, causing the emitted photon to have a longer wavelength and lower energy. Other processes that do not involve light emission can be used to relax the excited state S₁. These are non-radiative processes that hinder fluorescence emission by competing with it and lowering its effectiveness [39-41]. The shift of the fluorescence spectrum to longer wavelengths with respect to the excitation spectrum is called the Stokes shift.

Fluorescence relaxation processes are all spin-neutral (spin-allowed), and the electron’s spin orientation is always preserved. In contrast, phosphorescence is a separate phenomenon. The electron spin is inverted (spin-forbidden) as a result of a rapid (femto- to microseconds) intersystem crossing from a singlet S₁ to an energetically advantageous excited triplet T₁ state. After a long delay (a few milliseconds to a few hundred seconds), relaxation to the singlet S₀ may take place with the emission of a photon, known as phosphorescence.

Since the majority of these emerging contaminants (antibiotics and hormones) are non-fluorescent, several luminescent or fluorescent materials have been utilised to monitor their levels in different matrices. The choice of the sensor material is crucial to achieving efficient sensing of the target analyte for luminescence-based sensors. Although luminescent sensors have been made using a variety of organic fluorophores and phosphors, the drawbacks of conventional organic dyes for developing luminescent sensors include their toxicity, ease of aggregation, photobleachability, and low capacity for adsorption of the target analyte.

Numerous luminescent materials, including semiconductors, metal complexes, metal-based fluorescent nanoparticles, MOFs, and inorganic phosphors doped with lanthanides, have been thoroughly researched to address these drawbacks. In recent years, theoretical and applied research has focused heavily on luminescent MOFs as an alternative sensing material for fluorescent sensors. These MOFs have an easy-to-functionalize surface, a tunable pore size, intrinsic luminescence, and a desirable adsorption capacity that can enhance MOF–target analyte interactions and transduce these interactions into measurable optical responses. A nanoscale MOF (In-sbdc) with a significant quantum yield of 13% and stable emission in water, for instance, was synthesised by Liu et al. [42] using In³⁺ (metal node) and trans-4,4′-stilbenedicarboxylate (ligand). In-sbdc showed a sensitive response to a variety of tetracycline antibiotics, with detection limits of 0.28–0.30 μM, according to the scientists. Additionally, it was claimed that the antibiotics’ ability to quench In-sbdc was based on an energy transfer mechanism in the sensing system.

A water-stable two-dimensional lanthanide-based MOF (Ln-MOF) was synthesised by Ren et al. [38] in a different study to serve as a reversible luminescent sensor for the detection of sulfamethazine (SMZ) antibiotics. According to the authors, the limit of detection for SMZ is 0.655 μM, and the Ln-MOF luminescence is strongly quenched, with a quenching constant of 4.60 × 10⁴ M⁻¹. They proposed two potential mechanisms for quenching (inner-filter effect and electron transfer). The overlap between the antibiotic’s absorption spectrum and the Ln-MOF’s excitation/emission spectrum is thought to be the cause of the inner-filter effect, as shown in Figure 7. The second process was linked to an electron transfer from the L-MOF’s conduction band to the antibiotic’s lowest unoccupied molecular orbital. In a related study, Zhang et al. [43] constructed a fluorescent aptasensor to detect the hormone 17-estradiol in urine, water, and milk samples. They were able to detect at a limit of 0.35 nM and determined that the detection process relied on turn-on Förster resonance energy transfer.

The principles of fluorescence quenching (“turn-off”) [37-39] or fluorescence enhancement (“turn-on”) mechanisms [44,45] form the foundation of the majority of luminescent sensors. Static quenching and dynamic quenching are the two basic categories under which luminescence quenching is classified. Dynamic quenching, which is described by the Stern–Volmer equation (Equation 1), results from the interaction and subsequent collision between analyte and fluorophore.

The luminescence intensities prior to and following the addition of the quencher (analyte), respectively, are represented by F₀ and F. The molar concentration of the analyte is [C], and K_d is the Stern–Volmer dynamic quenching constant. The concentration of the quencher will have a linear relationship with the plot of F₀/F vs [C]. If more than one kind of quenching mechanism, such as dynamic quenching and static quenching, is involved in the process, the plot can shift away from linearity and bend upward or downward. Following that, Equation 1 can be represented as indicated in Equation 2, and by rearrangement, Equation 2 yields Equation 3 as follows:

where

K_d and K_s can be determined from slope and intercept of a straight line that is produced by plotting K_m against [C]. In the event that the Stern–Volmer graphs diverge downward, two fluorophore populations are present, but only one of them is accessible to the analyte. Equation 1 is modified in these circumstances to produce Equation 5, where F is the total fluorescence and Fx and Fy are the accessible fluorophore and non-accessible fluorophore fluorescence, respectively. If one fluorophore is accessible to the analyte (x) and the other is hidden (y), in this case, Equation 6 will give the luminescence.

The fluorophore–analyte complex forms a non-fluorescent ground state, which causes static quenching. The type of quenching mechanism is significantly influenced by the system temperature, concentration of analyte, and viscosity. When the temperature increases, the Stern–Volmer quenching constant (K_sv) values for the dynamic mechanism increase, whereas for the static mechanism, they decrease. This is consistent with the report of Sheta and co-workers [10]. According to the authors, the K_sv values of the Cu-MOF-based nanosensor were directly correlated with the temperature, indicating that the quenching mechanism between the triiodothyronine hormone and the optical sensor is dynamic. Additionally, the lifetime decay of the fluorophore is unaffected by changes in analyte concentration for static luminescence quenching. The luminescence lifespan in the case of dynamic quenching, however, varies depending on whether the quencher is present or not. The luminescence mechanisms are briefly discussed below, however, the detailed principles of these mechanisms are not covered in this review:

Fluorescence (Förster) resonance energy transfer (FRET): In the late 1940s, Theodor Förster put forth the theory, which is based on energy transfer. FRET occurs when an electronically excited fluorophore (donor) transmits its excitation energy to a nearby analyte (acceptor) within 10 nm in a non-radiative manner. In general, the degree of overlap between the fluorophore’s fluorescence emission spectrum and the analyte’s absorption spectrum, the relative orientation of the fluorophore and analyte dipoles, and the distance between them affect the rate of energy transfer [46]. In order to measure the hormone 17β-estradiol in complex sample matrices (milk, urine, or environmental water), Zhang et al. [43] developed a practical FRET-based turn-on fluorescence aptasensor with high selectivity, a low detection limit of 0.35 nM, and an efficient recovery rate of 92.4 to 120.6%. Zhou et al. [47] developed two- and three-dimensional Zn-MOFs with bis-ligand coordination for sensing fluorescent antibiotics (e.g., cefixime, lactams, chloramphenicol, or sulfonamides). The MOFs were found to be efficient, selective, and sensitive toward the antibiotics by their fluorescence quenching behaviour, and it was found that there were distinct overlaps between the analyte’s emission bands and the MOFs’ emission band at 322 nm. The FRET mechanism was therefore thought to be key in detecting these analytes.

Inner filter effect (IFE): The IFE was formerly thought of as a fluorescence measurement error. It has gained widespread use in luminescent sensing in recent years as a significant non-irradiation energy conversion technology. IFE is observed when the analyte (acceptor) absorbs the fluorophore’s (donor) excitation or emission light. The IFE-based fluorescence technique is simpler and more versatile than FRET since it does not require covalent bonding between the analyte and the fluorophore or intermolecular interaction [44,48,49]. Although energy transfer occurs in both IFE and FRET, the process can be distinguished by the fluorescence lifetime. In contrast to FRET, where the fluorescence lifetime changes, the fluorescence lifetime in the IFE remains constant both before and after the addition of an analyte to the fluorophore (independent of the fluorescent intensity) [46-51].

The IFE is further categorised into primary and secondary types. The primary IFE is the absorption of the fluorophore’s excitation by the analyte, the secondary IFE is the absorption of the fluorophore’s emission [50]. Yazhini et al. recently developed a highly stable inner filter effect-based Zn-luminescent MOF for the selective detection of tetracycline [48]. They were able to attain a very low detection limit of 0.11 μM and a Stern–Volmer quenching constant value of 1.2 × 10⁴ M⁻¹, which demonstrated the accuracy and precision of the MOF’s sensitivity towards the detection of antibiotics. They noticed that the MOF’s (fluorophore) excitation peaks are superimposed over the antibiotic’s (acceptor) UV–visible spectrum. Moreover, the estimated average fluorescence lifetimes before and after the addition of tetracycline were 1.39 and 1.36 ns, respectively, and remained virtually constant. This finding strongly shows that fluorescence quenching is caused by an antibiotic’s internal filter effect and confirms that FRET does not occur. A multicolour fluorescent tetracycline nanosensor (Figure 8) with an ultra-high sensitivity and a detection limit of 7.1 nM in real samples (river and lake water, honey, and milk) was reported by Zhang et al. [52] in another study.

The scientists developed a portable, simple, and affordable optical sensor based on test paper with fluorophore immobilisation for a quick and visible detection of antibiotics (Figure 8a). It was investigated and established how tetracycline-induced quenching occurred. As seen in Figure 8b, the fluorophore’s excitation spectrum and the absorbance band of tetracycline around 357 nm partially overlapped. Moreover, the addition of tetracycline barely affects the fluorescence lifetime of the fluorophore (Figure 8c), which changes from 7.42 to 6.90 ns. These findings confirmed that the steady quenching process based on IFE was responsible for the tetracycline-induced fluorescence quenching.

Photoinduced electron transfer (PET): PET is an excitation-induced electron transfer between analytes (electron acceptors) and a fluorophore (an electron donor). Typically, PET results in photoquenching due to an internal redox reaction during the electron deactivation process. For effective quenching, a complex driven by either hydrophobic, van der Waals, or π–π-stacking interactions is generated between the electron donor and the electron acceptor with a separation on a sub-nanometre length scale. The complexation of the electron donor and electron acceptor results in a change in the electron energy levels, which in turn affects the fluorescence signals. Although this complex is capable of returning to the ground state without emitting radiation, exciplex emission is occasionally seen. To detect nitrofuran antibiotics (such as metronidazole, ornidazole, nitrofurantoin, and ronidazole) in water, Fan et al. [53] synthesised a chemically stable zinc-based MOF that functions as a multi-responsive luminescent sensor. With the addition of the antibiotics, the MOF’s luminescence intensity decreased, with LOD values ranging from 19 to 26 ppb. The luminescent MOF-based sensor displayed good anti-interference capability and was highly selective for nitrofuran antibiotics in aqueous solutions. The photoinduced electron transfer was thought to be a key factor in quenching. The electron deficiency of all the antibiotics, which allowed the excited electrons to transfer from the fluorophore’s conduction-band orbital to the analyses’ lowest unoccupied molecular orbitals, was thought to be the cause of the PET-induced quenching. As can be seen in Figure 9, the antibiotics have substantially lower LUMO energy levels than the fluorophore ligands, which accounts for the excellent quenching effects of PET on the luminescence of the MOF sensor [54].

Photoinduced charge transfer (PCT): The PCT mechanism relies on the exchange of electrons between acceptor (analyte) and the donor (fluorophore), which results in the alteration of the fluorescence signals. A partial charge transfer of a fully conjugated system occurs in optical PCT sensors. This mechanism involves the complexation of donor and acceptor, which changes the electron energy levels and the fluorescence signals. While PET sensors have the electron donor moiety separated from the fluorophore by a spacer, PCT sensors typically feature an integrated receptor and fluorophore [46].

Intramolecular charge transfer (ICT): When the fluorophore contains both electron-withdrawing and electron-donating groups, ICT, an electron transfer process, takes place. In contrast to PET, the electronic states produced by this method are “charge-separated” states. The emission and absorption spectra make it simple to discriminate between PET and ICT. Although there is a significant quenching of luminescence intensity in PET, there is no visible spectral shift. Contrarily, ICT yields environment-dependent changes in luminescence intensity along with sizable modifications in the excitation and emission spectra.

Electrochemical sensors

Recent years have seen a significant increase in the use of electrochemical sensors [1,7,9,28-33,55-60] that use nanostructured materials as potent analytical tools because of their advantages in terms of portability, affordability, high sensitivity, and ease of fabrication. Through functions such as active large surface area, rapid electrode kinetics, and efficient catalytic activity, the amplification of electrochemical signals based on nanostructured materials has great potential to enhance both the selectivity and sensitivity of electrochemical sensors [55,56].

Even though various sensor systems based on nanomaterials have been published in the literature, it is still difficult to incorporate them into common in situ devices. The most often employed nanomaterials for electrochemical sensors are divided into four categories based on their chemical makeup: (i) metal oxides and metal-based materials (including MOF), (ii) dendrimers and polymer-based, (iii) carbonaceous materials, and (iv) hybrids or composites. Because of their exceptionally high thermal and electrical conductivities, effective catalytic properties, high chemical stability, rapid rate of electron transfer, adequate surface area, and favourable piezoelectric, electronic and gravimetric properties, metal and metal oxide materials with sizes less than 100 nm are the main choice for the design of electrochemical sensors. These characteristics all combine to improve the electrochemical process.

A sensing or working electrode that acts as a transducer, an electrolyte, a diffusion barrier, and a reference counter electrode are the common components of electrochemical sensors (Figure 10).

The target analyte interacts with the recognition layer at the sensing electrode surface to produce an electrical signal that contains the analytical information. Chemical reactions (redox reactions) on the electrode surface are converted by the physicochemical transducer into electrical signals that may be easily identified and shown by electrical equipment. Electrochemical sensor-based techniques can be classified as conductometric, potentiometric, voltammetric, or amperometric, depending on the electrical signal that needs to be measured [56]. Conductivity is measured using conductometric sensors at various frequencies. In potentiometric sensors, a local equilibrium is created at the sensor–analyte interface, and when no current is present, the composition or concentration of the analyte is determined from the potential difference (voltage) between the working and the reference electrode in the form of an electrical signal. The potential of the working electrode is a function of the analyte concentration in the solution. Amperometric sensors use a potential applied between a working and a reference electrode to produce the reduction or oxidation of an electroactive species and then measure the resulting current [61].

In voltammetric sensors that typically have two or three electrodes (the working electrode, the counter electrode, and the reference electrode), the current is measured as a function of the applied voltage at the working electrode. Currently, cyclic voltammetry (CV), differential pulse voltammetry (DPV), square wave voltammetry (SWV), linear sweep voltammetry (LSV), and stripping voltammetry are well-established voltammetric techniques that are frequently used for electrochemical sensing of antibiotic and hormone residues [62]. However, other methods for the identification of these emerging contaminants, such as chronoamperometry and electrochemical impedance spectroscopy, have also attracted a lot of interest [7]. Electrochemical biosensors are created by functionalizing the nanomaterial on the working electrode with biomolecules (e.g., nucleic acids, enzymes, proteins, aptamers, and immunoglobulins) to realise an analyte-specific reaction. Because of their modification with biomolecules, some electrochemical biosensors have shown higher specificity and selectivity than unmodified electrochemical sensors; nevertheless, they have shorter lifetimes and poorer levels of stability due to the rapid degradation of the biomolecules.

Extensive research has been focused on the development of different high-performance electrode materials via modification of the materials’ surface with functional groups or doping with various metals or via the formation of nanocomposites. Because of their large surface area, ordered structure, tunable physico-chemical properties, and good absorbability, MOFs exhibit high potential among the various materials used for electrochemical sensor electrodes [63].

Strong interactions between the functional groups of MOFs and the target biomolecules via electrostatic forces, stacking, and/or hydrogen bonding, which lead to high accumulation of the target analyte, are another factor that supports the development of electrochemical sensors. However, because of the high proportion of organic ligands, most MOFs have poor electrical conductivity, which lowers their electrochemical detection performance [64]. Researchers have focused on various research efforts to improve the conductivity and amplify the electrical signals of MOFs by combining them with other highly conductive materials (such as carbon materials, metal nanoparticles, or metal oxides) [63-69]. This is motivated by their large surface area, which can facilitate the loading of nanoparticles. Additionally, MOFs have been converted into their electrochemically active derivatives, such as mesoporous carbon composites and porous metal oxides, to achieve an improved electrochemical performance [63,67].

In a recent study, Tang et al. [29] used an ultrasonication and reduction process to combine nanostructured Ag nanoparticles with a MOF (ZIF-67) to fabricate nanopinna-based composite electrochemical sensors for acetaminophen detection. The results showed a wide linear detection range with a detection limit of 0.05 μM. Due to the synergistic effects of the wide porosity and high specific surface area of the MOF and the outstanding catalytic activity and high conductivity of Ag nanoparticles, the hybridisation improved the electrochemical performance of the sensor. A novel electrochemical aptasensor based on Fe MOFs was reported by Song and co-workers [70]. In order to increase the sensor’s sensitivity to the detection of antibiotics, the MOF was further modified using carbon nanofibres and gold nanoparticles using a variety of techniques. In the presence of seven other antibiotics, the electrochemical sensor showed good selectivity and stability, with the lowest detection limit of 0.01 nM. Tetracycline concentrations in samples of lake and tap water were precisely quantified by the sensor when its performance was tested using real samples. An electrochemical sensor for the detection of tetracycline based on a glassy carbon electrode modified with MIL-53 (Fe) was published by Chen and co-workers [71]. The authors claim that the electrochemical sensor has a good linear relationship for tetracycline detection in the range of 0.0643–1.53 mol L⁻¹ and that it has a high anti-interference ability.

It is important to note that a number of MOF-based materials have been used as opto-electrochemical sensors for the dual detection of hormones and antibiotics using hybrid optical and electrochemical methods. As two non-interfering and mutually independent signals are obtained, the likelihood of inaccurate results is reduced. For example, Rani et al. [50] synthesised a gold nanoparticle-decorated amine-functionalized Zr-MOF through a solvothermal synthesis process in order to detect nitrofurazone antibiotics via dual-mode (opto-electrochemical) sensing. Utilizing both the fluorescence of the Zr-MOF and the electrochemical properties of nitrofurazone antibiotics, a more accurate signal and diversified results were obtained. With a LOD of 3 × 10⁻⁹ mol·L⁻¹ in the electrochemical mode and 5.5 × 10⁻⁹ mol·L⁻¹ in fluorescence sensing, the prepared sensor demonstrated a wide linearity range for the detection of the antibiotic.

Metal-organic frameworks: properties and synthesis

In recent years, there has been a lot of interest in metal-organic frameworks (MOFs) [72-75], which are extremely porous materials. They are highly crystalline materials with extremely low densities (0.1–1 g·cm⁻³), the highest specific surface areas (165–7800 m²·g⁻¹) currently known, tunable surface properties, good mechanical and thermal stability, intriguing and controllable morphologies, and uniform yet tunable pores that are created by the self-assembly of metal ions or clusters and organic ligands (i.e., “linkers” or “struts”) through coordination bonding [30,67,76-78]. The development of the first MOF and covalent organic framework (COF) is credited to Omar Yaghi of Berkeley University of California. In particular, Yaghi reported in 1995 on the synthesis and crystallisation of the first MOF in which metal ions are connected by carboxylate linkers (ligands) [67], as shown in Figure 11.

The metal ions act as nodes, connecting the ligands’ arms to create a repetitive, cage-like structure. The internal surface area of MOFs is incredibly vast because of their hollow structure. However, the chemical stability of the older MOFs was not very good, especially those that contained pure tetrahedral divalent metal units.

However, an incredibly stable and highly functional MOF was published in 1999 by Yaghi and co-workers [79], and several other MOFs have been synthesized and reported by various other researchers since then. MOFs are also referred to as “coordination polymers”. Secondary building units (SBU), which determine the final topology and hence the properties of the MOF framework, are coordination complexes generated between the donor atoms of the ligands and the metal ions. By regulating how many ligands can bind to the metal and in which direction, the coordination preference of the metal, for instance, affects the stability, size, and nature of the formed pores.

Electronic and optical properties of MOFs

In addition to the physicochemical characteristics of MOFs already described, the following is a brief outline of some of the desirable qualities of MOFs that are required for developing opto-electrochemical sensors.

Electronic properties: Electrostatic potential, density of states, electron density, bandgap, and conductivity are some of a MOF’s crucial electrical properties. Because of the typically insulating properties of the organic linkers and low levels of π–π-orbital conjugation, significant electrical conductivity is uncommon in MOFs. Even though MOFs’ electronic characteristics have not received much attention, their potential as electrically conductive porous materials has only recently come to light. Recent research has demonstrated that the nature of metal clusters, their size, and the kind of organic linkers all affect the MOFs’ electronic properties [80,81]. To clarify the electrical properties of MOFs, Kuc et al. [80] used tight-binding simulations based on density functional theory with periodic boundary conditions. In comparison to the building units, they noticed that MOFs have a charge distribution that remains constant, and their electronic properties show a wide range of bandgap energies categorized as insulators or semiconductors. The authors pointed out that metal clusters (for example, isoreticular MOFs) essentially define the overall electronic properties of MOFs and provide MOFs with the characteristics of a wide-bandgap semiconductor like ZnO. The size of the organic linker and the hybridization of the central atom of the linker both affect the bandgap values of MOFs, which range from 1.0 to 5.5 eV. Organic linkers with more conjugated carbon atoms tend to have smaller bandgaps. Additionally, the bandgap narrows as the number of sp²-hybridised carbon atoms in the linker increases because more π states contribute to the overall band structure. Additionally, computational analyses have shown that the functionality of the linker, such as nitro, carboxylic, or amine, can affect the electrical characteristics of MOFs, particularly the bandgap. The p orbital interactions of the functional group with aromatic carbon atoms, which might produce localization of donor states close to the aromatic ring, were suggested as the source of the bandgap modulation due to changes in the functionality of the organic linker [82]. Due to MOFs’ low resistivity and rapid charge carrier mobility, some researchers [80-83] have recently suggested that MOFs occasionally exhibit superconductive behaviour. The presence of metallic state bands, which correspond to π-type crystal orbitals centred on ligand atoms and metal ions, as well as the availability of the most extensive mobility pathways between the sheets, clarified the conductive behaviour even further. Through variation in temperature resistivity, Clough et al. [83] demonstrated band-like metallic conductivity in cobalt-based MOF. The ferroelectric characteristics of MOFs have not yet been extensively investigated experimentally. However, calculations based on density functional theory have been used to examine aspects of MOFs, including ferroelectric polarisation. Particularly appealing multiferroic and ferroelectric properties have been associated with MOFs containing formate ligands [84]. The incorporation of guest molecules (dopants) into the porous network of MOFs, the use of a distinct linker, and the creation of MOF composites with other conducting materials (carbon-based, metal, and mixed metal oxide nanoparticles) are just a few experimental strategies that have been suggested to further improve the electronic properties of MOFs.

Luminescence properties: The fact that MOFs exhibit not just fluorescence but also phosphorescence and scintillation has drawn attention to their optical capabilities for some time. Due to their hybrid composition, MOF materials are capable of a variety of emission phenomena (Figure 12) that are uncommon in other material classes. For example, MOFs show metal-based emission (typically the lanthanides) and antennae effects, exciplex and excimer emission, analyte-based sensitization and emission, and ligand-based luminescence [85]. The MOF structure, which determines the bonding geometry, the metal electronic configuration, and the accessible states of the metal ions in the framework versus the HOMO–LUMO gap of the organic ligand(s), all have an impact on the emission that results from MOFs.

Metal-based luminescence: The incorporation of lanthanide elements into the MOF structure results in the most frequent occurrence of metal-based luminescence [85-87]. This process is known as ligand-to-metal charge transfer (LMCT). The lanthanide series consists of fifteen elements, ranging in atomic number from 57 for lanthanum to 71 for lutetium. The chemically related elements scandium and yttrium are also included in this group of fifteen, which are commonly referred to as rare earth elements. All lanthanoids, with the exception of Eu²⁺ and Ce⁴⁺, exist in the trivalent oxidation state (Ln³⁺). Ln³⁺ has a firmly seated 4f orbital in its ground state electronic configuration ([Xe]4fⁿ, where n = 0–14), and the electrons in this orbital are protected by a wider radial expansion of its filled 5s² and 5p⁶ subshells, which create 4f orbitals as inner orbitals. As a result, the spin–orbit interaction, which is the basis for their intriguing photophysical features, dominates the extremely minor and less significant crystal field influence on the deeply seated 4f orbitals. For lanthanoids, three different forms of electronic transitions are possible, namely the 4f→5d transition, broad charge transfer transitions, which are intense but broad, and sharp narrow bands (4f→4f transitions), which are very weak in intensity. Compared to the 4f→5d and the charge transfer transitions, which are Laporte-allowed transitions (high probability of occurrence), the 4f→4f transitions are Laporte-prohibited transitions (low probability of occurrence). However, due to parity selection principles, the direct excitation of lanthanoids into excited 4f states is practically limited, suppressing the lanthanoid’s sharp luminescence and leading to poor quantum yields and weak absorbance. Strong vibronic coupling between the desired lanthanoid and a highly absorbing ligand is a popular method of resolving this issue since it allows for direct energy transfer from the more accessible excited states of the ligand to the suitable metal energy level. The “antenna effect”, which is the name given to this coupling, causes a significant rise in luminescence [85]. Particularly, the lanthanide ion emits its distinctive luminescence as a result of being indirectly activated by energy or electron transfer through its surrounding ligands. A major benefit of this method is its significant Stokes shift, which makes it easier to distinguish between lanthanide luminescence and its excitation light and ligand-centred luminescence. Second, it significantly increases the luminescence quantum yield at room temperature conditions [87,88]. Although nearly all lanthanide elements display photoluminescent capabilities, only two lanthanoid ions, Eu³⁺ and Tb³⁺, have intense visible luminescence in red and green, respectively. These two ions are the ones that are most frequently used when synthesising MOFs for optical sensing applications. Note that phosphorescence, which is typically relatively weak at ambient temperature due to solvent quenching and self-quenching of the long-lived excited state, constitutes the majority of lanthanide luminescence. The possibility of solvent quenching and self-quenching, however, has been virtually eliminated in MOFs since lanthanide ions are entrenched in the network of organic ligands. So strong is the luminescence produced by lanthanides in MOFs that it can even be seen with the naked eye. Also, actinide-containing MOFs exhibit luminescent behaviour [86]. The metal ion can have different levels of impact on the emission depending on the electronic structure of the metal and the relative energy of the metal and linker orbitals. To develop luminescent MOFs, a variety of transition metals have also been combined with different ligands. Note that the linker, not the metal, is often the focal point of luminescence from such MOFs [85]. Because ligand–metal transitions (d→d) may strongly reabsorb or quench the fluorescence produced by the organic linker via energy or electron transfer through the partially filled d orbitals, the emission from transition metals with unpaired electrons is often not strong. But MOFs containing transition-metal ions devoid of unpaired electrons, particularly those with d¹⁰ configurations, can produce remarkable linker-based luminescence [85-87].

Linker-based luminescence: In MOF chemistry, a variety of organic linkers have been used, many of which contain rigid backbones that have been substituted with multiple carboxylate groups for metal coordination. As linkers in MOFs, organic compounds with fused π rings and strong conjugation are frequently used. Due to their stiffness, they are frequently both very emissive and intensely absorbing, which can result in a variety of intriguing luminescence features in the material. In MOFs, ligand-localised emission, ligand-to-metal charge transfer, and metal-to-ligand charge transfer are among the typical mechanisms of linker-based luminescence. Terephthalic acid, trimesic acid, 5-sulfosalicylic acid, 4-aminobenzoic acid, and nitrilotriacetic acid are a few of the commonly used linkers. Note that the scope of this review does not include a thorough discussion of the linker types. Depending on the type of analyte and the composition of the MOF, luminescence intensity in MOFs can either be quenched or enhanced.

Due to their exceptional characteristics, MOFs have found usage in a variety of fields, including sensors, gas adsorption, energy storage, drug delivery, catalysis, water treatment, and bio-medical imaging [89-101]. Numerous early MOFs produced from divalent metals displayed excellent porosity but were inappropriate for usage in water, moisture, or acidic or basic environments due to stability problems, which significantly restricted their further use and commercialization. When exposed to atmospheric moisture, MOF-5, for instance, eventually degrades [95]. As a result, in recent years, there has been an increase in interest in the chemical stability of MOFs. The stability of MOFs in various environments has begun to be addressed, and researchers are working to create more robust framework structures. The stability of MOFs is thought to be a result of the comparatively brittle coordination bridges that sustain the framework structures. For example, carboxylate-based ligands are considered hard bases and are capable of forming stable MOFs with high-valence metal ions (Cr³⁺, Fe³⁺, Al³⁺, Zr⁴⁺, and Ti⁴⁺), resulting in a more stable framework. Because of their exceptional stability in water and even in acidic or basic environments, high-valence metal-based MOFs like the MIL (Material Institut Lavoisier) series, including MIL-53 (synthesised in 2002), and the Zr⁴⁺-based MOFs have drawn a lot of attention. This trend has continued in recent years, with an increasing number of stable MOFs being synthesised and reported. Due to the development of more stable MOFs, the sensing of different analytes from industrial, food, water, clinical, and environmental samples is now included in the recent expansion of MOF-based material applications.

Because of their high sensitivity, extensive porosity, controllable architectures, facile functionalization, large surface area, and low detection limit, MOF-based sensors have received a lot of interest. Additionally, MOFs are a perfect choice for opto-electrochemical sensors because of their stability and the variety of interactions that are made possible by the inclusion of both inorganic and organic moieties in the structure [28,34-37,50,67]. For the development of optical sensors, MOFs are enormously beneficial since their organic ligands with conjugated or aromatic moieties can generate fluorescence when exposed to radiation. Also, the metal ions (such as inorganic clusters and lanthanides) can produce photoluminescence [34-37].

The uses of MOFs in electrochemical sensing have only been lightly investigated because the majority of MOFs exhibit insulating properties. The potential of MOFs as electrochemical sensing platforms has been shown in a number of recent studies. The following factors contribute to MOFs’ electrochemical sensing abilities: (i) They have vast surface areas, highly active catalytic capacities, and unsaturated metal coordination sites, which make them ideal materials for coating electrodes used in sensing applications. (ii) MOFs benefit from the mass transfer of the analytes due to their high porosity, which can effectively amplify the electrical signals and increase detection sensitivity. (iii) Through the use of size exclusion effects, MOF-based matrices can exhibit good selectivity towards particular analytes due to the tunable shape and size of the accessible cavities and channels within the framework [67,76]. MOFs thus provide intriguing and noteworthy benefits over other materials, considering their electronic and luminescent properties.

Electrochemical sensors display good stability and rapid response over a suitable humidity range. However, under extreme conditions, the electrochemical sensor lifetime is influenced by temperature and humidity. Because of the structural features of MOFs, the electrochemical sensing capabilities of MOFs are not considerably impacted by humidity. By exposing more active sites during the interaction with water molecules, MOFs’ increased porosity and larger specific surface area could limit the effects of humid environments. In practical settings, the key factor contributing to the decreased resistance of the majority of MOF-based sensors under high humidity is an increase in their conductivity. Increased humidity causes the adsorbed water molecules to form an interconnected water film on the MOF surface through hydrogen bonding, which promotes proton conduction and raises the sensing materials’ conductivity. Therefore, MOFs are capable to exhibit significant conductivity changes under various humidity environments by ligand choice and structural modification.

The inclusion of hydrophilic ligands and metal oxides into MOFs has been the focus of study to further increase the stability of MOFs under humid environments. For instance, metal–oxygen groups such as Ti–O and functional amino groups (NH₂) are hydrophilic, and MOFs based on these hydrophilic groups are anticipated to exhibit a decrease in the electrochemical impedance when the humidity levels rise. Additionally, Cu-based MOFs with high hydrophilicity resulting from the existence of uncoordinated active sites are helpful in the preparation of stable electrochemical sensors that can function well in extremely humid environments. Al-based MOFs, such as CAU-10 and MIL-96, are renowned for their great stability and strong hydrophilicity and are consequently anticipated to function well in humid environments.

Drawbacks of MOFs as opto-electrochemical sensors and recent advances

The classic MOFs have some limitations when utilized as opto-electrochemical sensors, despite their high specific surface area and high porosity. Here, some of these weaknesses are briefly reviewed, and new developments to address these shortcomings are also provided.

Poor specificity for analytes: Conventional MOFs lack the specificity for a single analyte. Non-specificity and subsequent activities from other electroactive analytes or analytes with fluorescence capabilities in the sampling fluid will compromise the accuracy of the results. To increase the specificity of MOF materials for the recognition of a particular analyte, numerous strategies have recently been developed. These strategies include the development of functional MOFs and their combination with bionic enzymatic lock-and-key structures, as well as the introduction of molecular recognition technology such as molecularly imprinted polymers, bioconjugation pairs, aptamers, and antibodies onto MOF structures. In several aspects, MOFs exhibit outstanding superiority over other supporting materials when used as matrices for immobilizing macromolecular biological components such as enzymes or aptamers. This method results in excellent specificity and selectivity of MOF-based opto-electrochemical sensors. Nevertheless, understanding the matrix and immobilization procedures is crucial to maximizing the activity and stability of the immobilized macromolecular entity because the physical and chemical characteristics of these biological entities may change during the immobilization process.

Low electrical conductivity in conventional MOFs: Conventional MOFs are well known for having poor electrical conductivity and low carrier mobility, which severely restricts their potential as electrode materials in electrochemical sensors for real-world applications. The main cause of the poor electrical conductivity of MOFs is the widespread use of carboxylate linkers. The electronegativity of the carboxylate oxygen atoms is so high (3.5) that electrons need a higher voltage to pass through the organic linkers. Because of this, the metal d orbitals and oxygen atoms do not overlap well. In order to solve these challenges, research interest in numerous MOF materials with exceptional electrical conductivity and high carrier mobility has increased significantly around the globe. To create such MOFs, a variety of methodologies and strategies have been employed. The formation of MOF-based nanocomposites with carbon-based materials, various metallic/metal oxide nanostructures (e.g., nanoparticles, nano-arrays, nanopillars, and monoliths) are examples of the techniques used to develop conductive MOF-based materials. Additionally, conductive metal nodes (such as Ag⁺, Cu²⁺, and Ni²⁺), redox-active linkers (such as catecholate, 2,3,6,7,10,11-hexaiminotriphenylene, hexaaminobenzene, and ortho-disubstituted benzene) conducting molecules are incorporated into the pores of MOFs [67-73]. MOFs made of different metals, also known as mixed metal MOFs, have recently been produced to dramatically increase the electrical conductivity. Because of the different oxidation potentials and related electron configurations, the bonding between different metal cations in MOFs can improve the electrocatalytic sensing effectiveness of the MOF and boost electrical conductivity.

Small linear dynamic range of bulk MOFs: A linear dynamic range, or the range of concentrations where the analytical signals are precisely correlated to the concentration of the analyte in the solution being tested, is one of the crucial factors to evaluate the performance of a robust and reliable sensor. The physicochemical processes on conventional MOFs’ surfaces are weakened because they are typically large and uneven bulk materials, which may limit their performance when employed as opto-electrochemical sensors. As shown in Tables 1–4, some of the bulk MOFs exhibit small linear dynamic ranges. Various research efforts have been made to enhance the functionality of MOFs as opto-electrochemical sensors and expand their linear dynamic range and analyte detection limits. Downsizing bulk MOFs to submicro- or nanoscale sizes is one of these approaches. When compared to bulk MOFs, micro-/nanoscale MOFs (such as nanowires, nanoarrays, nanosheets, nanorods, nanoclusters, and nanofibres) exhibit both the intrinsic properties as well as additional physicochemical characteristics, including less diffusion barriers, shorter diffusion distances for analyte transport, more readily accessible surface active sites, improved conductivity, and higher catalytic efficiency. Also, the number of defects on the outer surfaces of nanoscale hierarchical MOFs is higher. In electrochemical sensing, these characteristics of nanostructured MOFs either enhance their electrocatalytic performance or boost the reaction rate involving analytes bigger than the MOF’s pores.

General synthesis methods for MOFs

The key factors encouraging the use of MOFs in opto-electrochemical chemo- and biosensing (together with biodegradability, non-toxicity, and facile post-modification) are the positive features already mentioned. A variety of MOFs have been produced using a number of different techniques, and each technique has advantages and disadvantages. The topology, physicochemical characteristics, size, and shape of MOF crystals in micro- and nanoscale regimes are all controlled using various synthetic techniques. To increase product yields, reduce synthesis time, and achieve desirable physicochemical properties, modifications in compositional and process factors are used. The source of metal ions, solution pH, solvent, reactant concentration and molar ratio are compositional parameters that can be changed to control the size and shape of MOF crystals. The crystal size and shape can also be greatly influenced by process variables such as temperature, time, heating source, and pressure. The scope of this review paper does not include detailed synthesis mechanisms, characterisation of MOFs, or a review of each MOF synthesis method. It is strongly advised to read the review articles by Lee et al. [100] and Stock and Biswas [101] for a more comprehensive grasp of MOF synthesis methodologies.

MOFs have mostly been synthesised by conventional electrical heating via small-scale hydrothermal or solvothermal synthesis procedures, which can take hours to several days until complete crystallisation. These techniques are especially well suited for the composition-controlled development of large, high-quality, and extremely crystalline MOFs. Additionally, crystalline phases that are unstable at the melting point can be created using these techniques. In comparison to other traditional synthesis methods, MOFs produced using hydrothermal or solvothermal methods typically have higher chemical purities.

These methods have drawbacks, such as the requirement for expensive autoclaves and the inability to watch the crystal grow if a steel tube is employed, in addition to the lengthy synthesis duration. Various alternative synthesis techniques, including mechanochemical, electrochemical, microwave-assisted, sonochemical, microfluidic, ionothermal, and dry-gel conversion approaches, have since been used to speed up the synthesis time and generate smaller, more homogeneous crystals [67,76,96,97]. Each MOF synthesis method is briefly described in Figure 13, and Lee et al.’s review work [100] includes a detailed demonstration of synthesis methods with faster crystallization periods.

When using the microwave synthesis approach, a high-frequency magnetic field is produced, which can swiftly cause internal heating effects within molecules and cause the temperature of the reaction system to rise steadily and uniformly [77,102] to start chemical reactions. This method can generate MOFs with highly porous structures and lower particle sizes due to rapid crystallisation without causing localised overheating. By using a microwave-assisted technique at 120 °C in a microwave reactor with a microwave power of 100 W, Liu et al. [103] synthesised two-dimensional UiO-67 nanosheets. According to Jhung et al. [98], for instance, the synthesis of MIL-100-Cr using a microwave technique at 220 °C sped up the reaction by 20 times with comparable yield, physicochemical, and textural qualities when compared to the traditional hydrothermal procedure at the same temperature. The combination of the precursors’ quick dissolution and the oxygen–metal networks’ condensation acceleration was thought to be the cause of the rapid synthesis. Abbasi et al. compared a mechanochemical synthesis technique with the sonochemical synthesis of Cu-MOF utilising N,N-dimethylformamide (DMF) as the solvent [99]. The product yield, specific surface area, and average particle size for the ultrasound-synthesised Cu-MOF are 84.5%, 370.1 m²·g⁻¹, and about 50 nm, respectively, while they are 76.1%, 1033.8 m²·g⁻¹, and above 100 nm, respectively, for the microwave-synthesised Cu-MOF.

The synthesis of MOFs using the sonochemical method is quick and continuous, and the rate of synthesis is significantly higher than that of conventional hydrothermal/solvothermal methods. The mechanical effects that the ultrasonic waves have on the medium/solvent as they move through it during the synthesis can help in solvent emulsification, gel liquefaction, and solid dispersion. The MOFs produced by ultrasonic synthesis have homogeneous particle sizes. 2,6-pyridine dicarboxylic acid and zinc nitrate and were used to synthesise Zn-MOF using the ultrasonic-assisted reverse micelle process by Zeraati and co-workers [104]. For the mechanochemical synthesis process, it entails a low-cost mechanical grinding of organic ligands and metal salts, which lessens the effect of solvent volatilization in the environment. The consumption of organic solvents is drastically reduced when thermal energy is replaced with mechanical energy. There are three different ways to make MOFs using the mechanochemical technique, namely neat grinding, which requires no solvent, liquid-assisted grinding, which is rapid but needs a catalytic amount of solvent, and ion- and liquid-assisted grinding, which needs a catalytic solvent and a tiny amount of salt to accelerate MOF crystallisation [67]. Using the mechanochemical synthesis method, Khosroshahi et al. [105] synthesised a heterostructure MOF composite by manually pulverising MOF-808 and NiFe₂O₄ in a mortar.

Indirect bipolar electrodeposition, cathode synthesis, anode synthesis, and electrophoresis deposition can all be used to synthesise MOFs electrochemically. Rapid synthesis, benign reaction conditions, and good mesoporous structure are advantages of this approach despite its low yield and propensity for undesirable by-products. MOFs can be made continuously and uniformly, the particle shape may be controlled, and there is a minimal solvent demand [77]. For the electrochemical detection of the insecticide carbendazim in samples of vegetables and fruits, Peng et al. [106] used the electrodeposition approach to quickly and effectively produce a glassy carbon electrode covered with Co-MOFs and carbon nanohorns utilising 1,4-benzenedicarboxylic acid as the organic linker and Co(NO₃)₂·6H₂O as the metal ion source.

Also, during the synthesis of MOFs, the type of solvent used affects the nature of the MOF significantly because of its polarity, protolysis properties, and solubility of the organic ligands. The most often employed solvents include water, ionic liquids, deep eutectic solvents, ethanol, methanol, N,N-dimethylacetamide, and N,N-dimethylformamide. Some solvents function throughout the synthesis as ligands, while others serve as both a solvent and a structure-directing agent. To get around issues with varying solubilities of the different starting materials, mixtures of solvents have also been used. According to studies by Loiseau et al. [96] and Senkovska et al. [97] the influence of the solvent was seen for isoreticular MOFs based on the MIL-53 topology, identified as Al-MIL-69 or DUT-4. Using 2,6-naphthalene dicarboxylic acid as a rigid ligand and water as the solvent, Loiseau et al. [96] hydrothermally synthesised Al-MIL-69 at 210 °C for 16 h. The identical metal salt and linker were then employed by Senkovska et al. [97] in a non-aqueous solvent (DMF) at 120 °C for 24 h using a solvothermal method. While the synthesis in DMF as a solvent produced an open-pore MOF that exhibited no flexibility, the synthesis in water produced a nonporous closed structure that could not be opened.

Note that researchers are concentrating on the development of more rapid, continuous, and practical techniques for MOF synthesis in order to meet commercial, specialised application, and industrial needs. Recent research has focused in particular on using optimal synthesis methods to produce a desirable MOF using computational methods.

MOF-based electrochemical sensors for antibiotics and hormones

Due to their distinctive structural benefits, periodic network architectures, and unsaturated metal coordination sites, MOFs are ideal electrochemical sensing platforms, as was previously mentioned. As a result of these characteristics, MOFs have an enhanced catalytic capacity and can be used to effectively coat electrochemical sensor-based electrodes. Particularly, pristine MOFs with exceptional electroactivities (Cr-MOF, Cu-MOF, Co-MOF, and Ni-MOF) have been directly used to detect various pharmaceutical substances in diverse media. The conductivity and dispersibility of the original MOFs can be increased by the addition of carbon nanomaterials, which enhance the sensors’ electroactivity [76]. Pure MOFs and their composites can be further transformed into MOF-derived carbonaceous and/or metallic nanomaterials and their corresponding composites through the application of a controlled thermochemical process [107], which combines the intrinsic characteristics of MOFs with those of the metals/carbon, making these materials suitable for use as robust and high-performance electrodes for sensors. By integrating functional species into MOFs, it is also possible to develop novel MOF-based electrochemical sensors [108]. These new MOF-based electrochemical sensors will have improved multifunctional properties and selectivity for targeted pharmaceuticals. The active nitro groups found in some regularly used antibiotics, including nitrofurazone, chloramphenicol, and metronidazole, support redox reactions. These antibiotics have the ability to produce redox signals in electrochemical detection when used with MOF-based electrodes.

For example, Rani et al. [50] synthesised a zirconium-based MOF (UiO-66-NH₂) utilising a solvothermal technique for opto-electrochemical sensing of nitrofurazone. The MOF electrochemical sensing platform demonstrated a good recovery in the range of 97.7–98.4% in lake water samples and a broad linear concentration range for nitrofurazone of 1 × 10⁻⁸ to 5 × 10⁻⁵ mol/L. Due to nitrofurazone’s electrochemical activity, its redox activity on MOF-based electrodes was observed as a redox signal. Two reduction peaks and two equivalent oxidation peaks were observed in the cyclic voltammogram of the antibiotic, according to the authors. A one-electron reduction of the nitro group in nitrofurazone to a nitro radical was thought to be the cause of the initial cathodic peak. As a result, a free nitro radical was created, and it was further reduced to a hydroxylamine derivative. It was determined that the nitro radical’s oxidation to nitrofurazone and the oxidation of the hydroxylamine derivative to a nitroso derivative were the causes of two oxidation peaks in the reverse scan. The detection limits for chloramphenicol and metronidazole antibiotics were determined to be 31 and 165 nM, respectively, in a study by Baikeli et al. [109] utilising an iron-doped metal-organic framework (ZIF-8) calcined nanoporous electrode modified with glassy carbon. Nitrogen in the structures of chloramphenicol and metronidazole exerts a redox effect. The Fe dopant in ZIF-8 MOF provides an electron to the nitro group, which is then coupled with the proton donor to generate a reduction product and to produce the electrochemical reaction signal.

A new electrochemical aptasensor based on NH₂-MIL-101(Fe) MOFs for tetracycline detection was reported by Song and co-workers [70]. The sensor showed a lowest detection limit of 0.01 nM. Investigations showed that the repeatability was good, with the relative standard deviation of ten subsequent measurements being just about 2.3%. Tetracycline recovery rates in actual samples of Donghua University lake and tap water ranged from 89.7 to 102.8%. To produce highly effective electrochemical aptasensors for the detection of ultrasmall traces of penicillin, He et al. [110] synthesised an Ag-based MOF aptasensor. The Ag₂SiF₆-MOF-based electrochemical sensor exhibits exceptional penicillin sensitivity and selectivity, with an LOD of 0.849 pg·mL⁻¹. The relative standard deviation values are less than 5.0%, and the penicillin recovery rates from the raw milk sample ranged from 99.3 to 104.8%. The findings suggest that the newly developed aptasensor has good accuracy for detecting penicillin in raw milk. Table 1 lists recently published electrochemical MOF-based sensors for detecting antibiotics in various samples, and a review of the findings, significant trends, detection techniques, and technical remarks are provided.

Numerous MOF-based electrochemical sensors have also been used to detect hormones at various concentrations. Some of these hormones have electrochemical properties that make it easier for MOFs to detect them, while analyte recognition molecules are typically added to MOF materials to enhance the sensing and selectivity of the sensor materials. For instance, Li et al. [129] reported the development of a copper-based MOF electrochemical sensor for the detection of dopamine, a significant neurotransmitter. The as-synthesized MOF electrode sensor showed strong selectivity for dopamine in real samples when operated under optimal conditions, with a detection limit as low as 1.5 × 10⁻⁷ M and a linear response range of 5.0 × 10⁻⁷ to 1.0 × 10⁻⁴ M. By immobilizing a monoclonal antibody specific for thyroxine onto a Cu-MOF modified with polyaniline, Mradula et al. [130] reported a label-free electrochemical immunosensor for the detection of the hormone thyroxine. The MOF-composite immunosensor demonstrated high selectivity, a rapid detection time of 20 min, a detection limit of 0.33–0.17 pM, and a linear range of 10–10⁵ pM. In order to detect the follicle stimulating hormone (FSH) antigen, Palanisamy et al. [131] created amine-functionalized Fe-containing MOFs and then bound them with FSH antibodies. According to the study, the estimated LOD for the MOF-based immunosensor material for FSH was 11.5 and 11.6 fg/mL for serum and buffered solutions, respectively. MOF-based electrochemical immunosensors are fast, precise, portable, and typically disposable, requiring very little sample volume for automated hormone detection as compared to sophisticated, expensive, and time-consuming methodologies and advanced equipment. Recent studies on MOF-based electrochemical sensors for the detection of various types of hormones are reviewed, summarized, and provided in Table 2. These studies include important parametric conditions, detection mechanisms, and significant results.

MOF-based fluorescent sensors for antibiotics and hormones

Some key intrinsic properties of MOFs, particularly those with lanthanide metal nodes, include higher quantum yields, a larger Stokes shift, and a longer lifetime. These characteristics make MOFs attractive fluorescence sensors for detecting various analytes. Recently, increasing research efforts have been focused on the fabrication of novel fluorescent MOFs for detecting trace amounts of hormones and antibiotics. Organic ligands commonly serve as recognition sites for the precise detection of analytes and give rise to the fluorescence of MOFs [153-158]. By lowering non-radiative relaxation or improving the ligand-to-metal charge transfer activity, organic ligand modifications that functionalize MOFs yield high fluorescence emissions [155].

Additionally, the immobilisation of hydrophilic groups (such as RCOO⁻, –NH₂ and RSO₃⁻) onto the organic ligands might improve the water solubility and structural stability of MOFs, expanding the range of potential uses for them in aqueous environments. For instance, by synthesising a Eu-doped NH₂-MIL-53(Al) nanostructured material, Chen et al. [156] developed an antibiotic detection probe with fluorescence. Tetracycline was added to the sensor solution, and the antenna effect produced by coordinating tetracycline with Eu³⁺ enhanced the fluorescence intensity at 616 nm, the characteristic emission peak of Eu³⁺, whereas the fluorescence of the MOF at 433 nm was quenched by Förster resonance energy transfer between the tetracycline–Eu³⁺ complex and the MOF. As a result, tetracycline was detected effectively thanks to this shift in the fluorescence signal ratio. Table 3 and Table 4 highlight recent studies using MOF as fluorescence sensors for the detection of hormones and antibiotics. These tables review, enumerate, and provide information on the detection procedures, types of MOFs, and real samples in which the analytes have been identified.

Opto-electrochemical nanostructured sensors: practical challenges and future perspectives

The use of opto-electrochemical sensors in numerous environmental, agro-industrial, and biomedical applications has great potential. Despite the enormous success of glucose sensors, much more effort is still needed to develop opto-electrochemical sensors that are commercially feasible or market-ready. For instance, the COVID-19 pandemic has shown how crucial and urgent it is to have reliable, affordable, and rapid diagnostic sensing tools. Opto-electrochemical sensors have a number of advantages over other currently used sensing approaches or techniques. They can be mass-produced and are compact enough to fit into portable devices. The MOF opto-electrochemical sensors have shown a number of promising results, and in particular, they have produced good results when applied to real samples. Nevertheless, there are still obstacles preventing their commercialization and other challenges. The following is a brief discussion of some of these challenges, recent developments, and potential future advances.

It is important to note that while many authors have used opto-electrochemical sensors to detect analytes in actual samples, very few authors have evaluated the LOD and sensing capabilities of these sensors against more established techniques such as HPLC, AAS, and ELISA, to confirm their sensitivity and selectivity. Additionally, the majority of the published studies base their validation of the sensor performance solely on the real samples produced in the lab of the author without comparing those performances to those of actual samples produced by other researchers in other labs. As a result of the data being based solely on the author’s laboratory investigations, the commercialization of such sensors is constrained. Therefore, for the sake of practicality, findings from conventional analytical methods and actual samples from other labs under numerous different conditions should also be included when reporting the sensor performances.

The application of “actual samples” is arguably the most significant test to confirm the reliability of a sensor. It is impossible to certify a sensor as a trustworthy sensing tool if it is not stable or functioning in real samples. Numerous species that easily adsorb onto surfaces are frequently present in real samples. Since non-specific adsorption tends to dramatically reduce the repeatability, specificity, and sensitivity of the sensors, it has been one of the key obstacles to using electrochemical sensors in practical applications. With electrochemical sensors, sweat, blood serum, human serum, urine, blood serum, saliva, interstitial and tear fluids are the most often utilized real samples.

All of these practical samples are affected by the matrix effect, which tends to negatively obstruct the detection of a particular analyte and reduce recovery values and sensor sensitivity [189-192]. For instance, saliva samples must frequently be filtered or diluted in order to be utilized as actual samples because they are complicated mixtures. The biggest issue with urine samples is the variety of pH ranges that can affect the position of peak potential and the height of current intensity, and create stability problems [191]. Tear fluid has also recently attracted a lot of attention due to its reduced complexity and ease of access for non-invasive sampling procedures, however the pH value can fluctuate, the sample volume is small, and the composition of tears shed in response to emotion and irritation may differ.

Researchers routinely dilute real samples to lower the interference effect below a tolerable threshold, which aids in overcoming the matrix effect. However, the sample is further from reality the more diluted it gets. Without sample dilution or processing, sensors should ideally be able to perform well with pure actual samples such as whole blood [189]. Researchers are addressing this issue when it comes to point-of-care sensing platforms and diagnostic sensors with cutting-edge materials and techniques to enhance sensor performance when utilized with actual samples. In particular, the development of non-charged and hydrophilic layers has been employed to obstruct matrix adsorption on the surface of the electrochemical sensors using both active and passive approaches. Lichtenberg et al. provided a comprehensive review of these techniques [192]. Although molecularly imprinted polymers have been used to modify electro-active sensor materials such as MOFs to increase the specificity and selectivity of the target analytes for antibiotics and hormones, future research should also concentrate on the development of simple-to-prepare reagents capable of suppressing the matrix effects in actual samples.

A fluorescence labelling procedure is initially needed for fluorescence-based nanostructured sensors in order to detect analytes such as antibiotics and hormones that are not fluorescent by nature. Linking the fluorescent tags to the biomolecules may require additional sample preparation steps. To avoid additional sample preparation stages, some studies have concentrated on the integration of optical sensing platforms with microfluidic channels that contain sample filtration/separation, target labelling, mixing, and washing. Prefabricating nanostructured materials with recognition moieties for particular analyte detection is another strategy. For instance, MOFs with strong fluorescence properties can be modified using aptamers or molecularly imprinted polymers in disposable, low-cost materials such as filter paper. This method can greatly cut down on time-consuming procedures and improve the system’s effectiveness.

The operation of the potentiostat for an electrochemical sensor still requires external instruments. A laptop or PC, for instance, is frequently needed to operate the potentiostat or process the data. This is a challenge, particularly when deploying the electrochemical sensor for routine analysis in the field. Single-board computers, such as Raspberry Pi, Arduino, and smartphones, can be used as a practical solution to tackle challenging tasks simultaneously, such as data analysis, image processing, and controlling peripheral devices. Additionally, by enabling real-time analyte detection for numerous other applications, including healthcare and environmental monitoring, the portable electrochemical sensor device will perform better when connected to the Internet of Things.

Obtaining a low limit of detection is another key problem that needs to be addressed for the development of electrochemical sensors that are commercially viable or market-ready. When developing a commercially viable electrochemical sensor, the LOD is a crucial parameter. This metric represents the smallest concentration or amount of a particular analyte that may be accurately identified while maintaining a respectable signal-to-noise ratio. Because analytes are frequently present in real samples only in trace amounts, developing a sensor with a low LOD is essential. LODs with values as low as picomole and femtomole levels have been reached in the case of some ultra-sensitive electrochemical sensors due to the development of nanomaterial-modified surfaces.

Researchers have developed many nanostructured materials to modify the electrodes. Since drop casting techniques are frequently not as reproducible as one would want, it is still difficult to modify electrodes with nanostructured materials based on these techniques. The conformation and topology of these nanomaterials, for instance, may change between each sensor since it is challenging to manage the immobilization of nanoparticles with varied populations of size and shape when making large quantities of sensor electrodes. Since it is not practical to evaluate each sensor made in mass-production facilities, sensor-to-sensor repeatability is crucial during the manufacturing process. Drop casting frequently results in uneven coatings of different thicknesses. Although spraying/spray coating creates uniform coatings with the ability to vary thickness, the majority of commonly used spin/spray coating devices are still expensive and not made for the electrodes used in electrochemical sensors that have smaller surfaces, such as disposable screen-printed electrodes. Future research should focus on developing simpler, more affordable mechanized immobilization and coating systems that can be used to modify commercially viable electrodes with uniform and consistent nanostructured surfaces.

Notably, the stability of sensors has also proven challenging, restricting the use of these devices under extreme conditions and in remote places. As a result, it is essential to develop sensors that can function for a considerable amount of time. Sensors are frequently distinguished by their shelf-life. Due to problems with aggregation and flaking of layers modified by nanoparticles, long-term stability may become a significant concern when utilizing nanomaterials. It is anticipated that the high surface area and stability of MOFs as hybrid electrode materials will enable the production of remarkable sensors with superior stability and minimal loss in analytical performance by combining the excellent electrical conductivity of inorganic nanomaterials with the analyte recognition capabilities of molecularly imprinted polymers and aptamers.

Poor electrical conductivity and low analyte specificity are the main drawbacks of conventional MOF-based electrochemical sensors. Hybrids and the development of nanocomposite materials based on MOFs are recent advancements to deal with these problems. For instance, scientists are developing multimetallic MOFs. Some newly developed bimetallic MOFs have been used gradually in the electrochemical sensor industry because they offer electroanalytical capabilities superior to those of conventional MOFs. Jalal et al. [193] showed that bimetallic MOFs exhibit greater electrochemical activity than monometallic MOFs for electrochemical detection of the antibiotic doxorubicin because of the synergistic interaction between the metal centres and the electrochemically active ligands.

An additional strategy is the development of MOF-based nanocarriers containing molecular recognition components (such as proteins, enzymes, and antibodies) and signals amplification components (metal nanoparticles or enzymes). The great porosity of MOFs makes it possible for them to hold a lot of active molecules, and the numerous surface functional groups open up the possibility of biomolecular modification via hydrogen bonding, π–π stacking, covalent bonding, and other intermolecular interactions. Excellent specificity, precision, and stability are provided for the detection of analytes on the complex matrix by the addition of specialized recognition components. Additionally, the incorporation of bimetallic nanoparticles in MOF structures could boost the efficiency of electron transfer at the electrode interface in addition to offering numerous catalytic sites.

Notably, the challenges with opto-electrochemical sensors described above apply not only to the detection of hormones or antibiotics but also to other applications in agriculture, forensic science, food safety, environmental monitoring, and defence and military applications. Fortunately, the discovery made possible by the use of diverse nanostructured materials constitutes a substantial advancement with implications for all the sectors indicated above. More research should concentrate on developing market-ready products using lab-developed sensors that have undergone extensive field testing and stability investigations over a period of many months.

Conclusion

For the development of reliable high-performance sensors for the monitoring and detection of trace analytes such as pharmaceuticals and their metabolites in complex matrices, high sensitivity and low detection limits are in fact the essential requirements. The focus of this review is on opto-electrochemical sensing materials based on functional metal-organic framework nanostructured materials for the detection of various pharmaceuticals, in particular, antibiotics and hormones, in complex matrices. Numerous opto-electrochemical sensing applications are made possible by the high porosity, tunable topology, high quantum yield, fluorescence capabilities, and ease of functionalization of MOFs. Their distinctive characteristics, including electrical, optical, and chemical properties, are also reviewed and critically discussed. Numerous optical sensing systems are reviewed and discussed, with a focus on those connected to fluorescence sensing and electrochemical processes.

The challenges with opto-electrochemical sensors based on MOFs are highlighted. Also mentioned are recent developments that aim to ease these technical difficulties. In this regard, it is demonstrated that the recognition potential, electrocatalytic, fluorescence emission, and analytical performance of MOF sensors can all be enhanced by the incorporation of synthetic or improved recognition elements such as molecularly imprinted polymers, aptamers, and mixed nanoparticles. Summaries of previously published opto-electrochemical sensors based on MOFS and their analytical parameters are provided in Tables 1–4. We critically evaluate and discuss key findings on mechanisms, synthesis approaches, actual samples used, and ligands and reagents for the synthesis of the MOFs.

It should be noted that MOFs and their derivatives are not just useful for detecting hormones and antibiotics; rather, their inherent properties have shown that they can be applied to a variety of fields, including healthcare (point-of-care), the environment, food quality and safety, and agriculture. Also, they have a very promising future in the practical detection of explosives. It is envisaged that future research will combine MOFs with other innovative nanostructured materials, such as Mxenes, transition metal chalcogenides, quantum dots, and artificial nanozymes, to develop hybrid materials with synergistically enhanced characteristics and stability.