Advances of Nano-Enabled ZnFe₂O₄ Based-Gas Sensors for VOC Detection and Their Potential Applications: A Review

Abstract

The demand for reliable gas sensing technologies in chemical, manufacturing, environmental, and occupational sites has increased in the last few decades following the global volatile gas sensor market, which is expected to grow further beyond 2025. Currently, several types of sensors have been employed for applications in different fields. Optical sensors are widely implemented in mining and environmental monitoring. Conventional food testing methods are utilized for the detection of any chemical or microbial agent in the food industry. Although robust and sensitive, most sensing technologies are expensive, labor-intensive, and necessitate the use of time-consuming gas sampling pretreatment steps, and these issues impede the achievement of quick, simple detection, portable, and cost-effective gas monitoring. For this reason, researchers around the world are investigating the possibility of using gas sensors as a promising technology that has the potential to alleviate industrial safety concerns. As a highly sensitive semiconducting metal oxide, gas sensors based on ZnFe₂O₄ have the potential to ensure environmental and occupational safety in real time. This review introduces and highlights recent developments in ZnFe₂O₄ gas sensors for application in different fields. The challenges limiting the wide application of the ZnFe₂O₄ sensor are outlined. Furthermore, this review discusses the common strategies adopted to improve the sensing properties of ZnFe₂O₄ for gas detection. Finally, future perspectives on further improvements of ZnFe₂O₄ sensing properties are discussed, and integration of ZnFe₂O₄ sensors into electronic noses to tackle the selectivity issue and how they can feature on the Internet of Things is outlined.

1. Introduction

The detection of different target gases using gas sensors has drawn a lot of attention in recent years. This is mainly due to certain industrial activities in the mining and energy sectors that release various toxic, flammable, and explosive gases (e.g., CH₄, H₂S, CO₂), thus threatening human health and the surrounding environment [1]. As such, environmental monitoring through the detection of gases by reliable sensors remains crucially important. The detection of naturally emitted gases is essential to process and quality control. Furthermore, chemical reactions with gaseous emissions in food production and aging have posed a sharp rise in the demand for gas sensors to analyze complex mixtures. A survey conducted by Grand View Research in 2018 showed that more than 20% of the market share of Europe’s Volatile Organic Compound (VOC) gas sensor, by application, was in the food and beverage sector [2]. Furthermore, the global market size for VOC gas sensors was valued at USD 141.7 million in 2018 and is expected to increase by 4.0% in 2025 [2]. This creates opportunities for sensors targeting VOCs.

In general, sensors are devices that transform a chemical reaction of an analyte or a form of physical property of a system into an analytically useful signal. They operate mainly under two basic functional units. A receptor part is the unit that transforms chemical information into measurable responses, and a transducer takes the response and converts it to a useful signal [3]. The receptor part of sensors operates under three principles, including physical, chemical, and biochemical principles. These principles describe the operating procedure with which the sensors are classified. The physical principle is based on the measurement of temperature, mass, electrical properties, absorbance, etc. On the other hand, the chemical sensor is based on chemical reactions with an analyte to give a certain signal, while the biochemical one gives out a signal due to a biochemical process and is commonly known as a biosensor [4,5]. For this reason, sensors have been used in different fields, including environmental control and monitoring [6], biomedical [7], operation and maintenance in civil construction [8], the food sector [4], and other production industries [9,10,11]. For instance, optical sensors have been widely implemented in the mining and environmental monitoring sectors [12]. In the food industry, conventional methods have been used to detect and analyze several VOCs released by food. These include gas chromatographic techniques, high-performance liquid chromatography (HPLC), polymerase chain reaction (PCR), as well as immunological methods, i.e., enzyme-linked immunosorbent assays (ELISAs) [13,14,15,16,17]. Although robust and sensitive, most sensing technologies are expensive, labor-intensive, and require time-consuming gas sampling pretreatment steps, and these issues impede the achievement of quick, simple detection, portable, and cost-effective gas monitoring. For instance, Mustafa et al. [18] described a novel approach to designing a calorimetric-based biosensor to measure hypoxanthine levels in degraded fish. Carvalho et al. [19] developed a texture sensor capable of differentiating fruits according to skin texture and determining fruit ripeness. However, several sensors come with disadvantages. To obtain more accurate results using the textural sensor developed by Carvalho et al. [19], the fruits should be rubbed at different locations, as textures differ with location, requiring labor and time. In contrast, gas sensors based on variations in electrical properties have several advantages over other sensors. They offer long-term detection at low concentrations without the need for complex measurement techniques. Of particular interest, SMOs have the unique property of reversible gas interaction with pre-adsorbed ambient oxygen, which significantly changes the electrical resistance measured as the output signal. Hence, SMO-based gas sensors have attracted a lot of attention, considering that they show great potential for gas-sensing applications. This type of sensor offers low cost, easy fabrication, simplicity of measurement, and enhanced sensing performance [20]. Moreover, SMO-based gas sensors offer a wide range of applicability. Different SMO materials, including binary oxides (e.g., ZnO, TiO₂, SnO₂, CuO, etc.) and ternary oxides (e.g., LaCeCoO₃, NiFe₂O₄, etc.), are available for choice to be used in the design and fabrication of gas sensors as active sensing layers. Among them, spinel-type ferrites with an AB₂O₄ formula are sensitive to a wide range of gases, such as H₂S, VOCs, NH₃, NO₂, and H₂O, giving them application possibilities in a wide range of fields. This is owing to their unique electronic and magnetic structures that can be tuned to suit a particular application by the redistribution of cations in the structure. However, a variation in the stoichiometry can cause instability and gas sensing limitations. Several strategies have been developed and applied to improve spinel-ferrite-based sensors to maintain their possibilities for a wide range of applications. Among several spinel ferrite materials, zinc ferrite (ZnFe₂O₄) has been extensively studied for its potential in gas sensing owing to its excellent chemical stability, low coercivity, and high electromagnetic performance [21,22,23,24,25]. A comparison of commonly used and investigated SMOs for VOC detection shows that ZnFe₂O₄ and its composites are promising materials for manufacturing as a VOC gas sensor (Table 1). Moreover, the selective properties of ZnFe₂O₄ sensors towards VOCs make them a good candidate for gas sensing applications in the food sector, occupational sites, chemical, and production industries.

Although ZnFe₂O₄ sensors show high sensitivity and selectivity to several groups of target gases, their selectivity to specific gas species and high operating temperatures remain a challenge that hinders their wide-range application. Researchers have attempted to improve the selectivity of ZnFe₂O₄ towards a particular VOC through surface decoration with metal nanoparticles, including Ag, Au, and Pt, and the design of novel hybrid nanocomposites (i.e., ZnO/ZnFe₂O₄) [26,27]. Furthermore, to tackle the selectivity problems of SMO-based sensors, Industry 4.0 technologies are implemented for more reliable and sustainable devices that automate and modernize sensor applications for all industrial needs.

Therefore, in this review, the recent advances and applicability of ZnFe₂O₄ gas sensors have been explored. We start by briefly explaining the structure and describing the accepted gas sensing principle of a spinel-type ZnFe₂O₄ based sensor. We further outline the challenges that hinder practical applications of ZnFe₂O₄ based sensors and indicate the important gas sensing properties in various industries. Different strategies used to develop and overcome the limitations of practical applications of SMO-based gas sensors in general are also discussed. Finally, different industrial applications of ZnFe₂O₄ based sensors and machine learning techniques that can be adopted to improve the selectivity and applicability of sensors are described.

Table 1. Gas sensors employed for VOC detection.

Material	Target Gas	Response (Ra/Rg)/Concentration (ppm)	Limit of Detection (ppm)	Response/Recovery Times (s)	Operating Temperature (°C)	Ref
ZnFe2O4	Acetone	192/90	-	3/21	120	[28]
ZnO/ZnFe2O4	Acetone	92.9/90	0.18	7.7/17.3	120	[29]
ZnFe2O4/SnO2	Acetone	120/100	0.10	30/197	210	[30]
Sn-ZnO/ZnFe2O4	Triethylamine	28.1/10	<0.2	-/7	270	[31]
Au-ZnO	Isoprene	1371/1	0.006	-	350	[32]
ZnO-CuO	Acetone	11.1/1	0.04	-	200	[33]
Pd-SnO2	Ethylene	11.1/100	0.05	1/-	250	[34]
SnO2-CuO	Ethanol	8/100	-	4/10	320	[35]
SnO2	Ethanol	59.6/40	-	105/100	150	[36]
CuO/WO3	Xylene	6.3/50	-	5.5/16	260	[37]
TiO2	Acetone	12/100	-	3/421	320	[38]
CoTiO3/TiO2	Benzene	33.4/50	0.1	49/9	RT	[39]
TiO2	Ethanol	2.2/100	-	-	RT	[40]

Table 1. Gas sensors employed for VOC detection.

Material	Target Gas	Response (Ra/Rg)/Concentration (ppm)	Limit of Detection (ppm)	Response/Recovery Times (s)	Operating Temperature (°C)	Ref
ZnFe2O4	Acetone	192/90	-	3/21	120	[28]
ZnO/ZnFe2O4	Acetone	92.9/90	0.18	7.7/17.3	120	[29]
ZnFe2O4/SnO2	Acetone	120/100	0.10	30/197	210	[30]
Sn-ZnO/ZnFe2O4	Triethylamine	28.1/10	<0.2	-/7	270	[31]
Au-ZnO	Isoprene	1371/1	0.006	-	350	[32]
ZnO-CuO	Acetone	11.1/1	0.04	-	200	[33]
Pd-SnO2	Ethylene	11.1/100	0.05	1/-	250	[34]
SnO2-CuO	Ethanol	8/100	-	4/10	320	[35]
SnO2	Ethanol	59.6/40	-	105/100	150	[36]
CuO/WO3	Xylene	6.3/50	-	5.5/16	260	[37]
TiO2	Acetone	12/100	-	3/421	320	[38]
CoTiO3/TiO2	Benzene	33.4/50	0.1	49/9	RT	[39]
TiO2	Ethanol	2.2/100	-	-	RT	[40]

2. Gas Sensors

2.1. Semiconductor Metal Oxides (SMO) Based Gas Sensors

An SMO-based sensor has a sensing element with a planar arrangement, consisting of a sensitive layer deposited over a substrate with electrodes for the measurement of the electrical properties. The sensing layer is generally heated by a heater located at the back of the substrate. SMO-based sensors are often referred to as chemiresistive sensors due to the changes in electrical resistance in response to changes in the nearby atmospheric environment. This phenomenon was first demonstrated in 1962 by Seiyama using ZnO [41]. Since then, sensitive layers composed of various binary, ternary, and composite SMOs have been studied for their ability to detect gases, making them more relevant in various applications.

The wide range of SMOs employed for gas sensing are characterized as either n-type, with electrons as the majority charge carriers, or p-type, with holes as the majority charge carriers. Unfortunately, most SMO materials exhibit limitations such as cross-sensitivity to untargeted gases, drift, high humidity sensitivity, thermal stability, and longer response times. This makes the choice of a suitable sensor for each application difficult. ZnFe₂O₄ based sensors are reported to be sensitive to a variety of gases, such as H₂S, VOCs, NH₃, NO₂, and H₂O. This places it in a position to be explored for a wide range of applications. Regardless of performance, an ideal SMO should possess unique sensing properties that are suitable for a particular application.

2.2. ZnFe₂O₄ Spinel Structure and Gas Sensing Mechanism

Since general gas detection involves a change in electrical properties, it is important to understand the structure and conductance of spinel ferrites to fully grasp the sensing mechanism of ZnFe₂O₄ based sensors. A spinel structure contains sublattices with a general formula of AB₂O₄. The bulk spinel structure of ZnFe₂O₄ is typical. Therefore, all the Fe³⁺ ions prefer to occupy the octahedral (B) sites, while divalent and non-magnetic Zn²⁺ ions prefer to occupy the tetrahedral (A) sites. The distribution of Zn²⁺ ions and Fe³⁺ ions at sites A and B is altered in nanocrystalline ZnFe₂O₄ spinel ferrite, resulting in the mixed structure described as follows: $\left(Z{n}_{1-x}^{2+}F{e}_x^{3+}\right)\left[Z{n}_x^{2+}F{e}_{2-x}^{3+}\right]O_4$ where x is the inversion parameter. This inversion of cations introduces interesting magnetic and electrical properties [42]. The concentration of charge carriers is regulated by the non-stoichiometry of cations sitting at the B-site. Therefore, the conductivity is due to the hopping of charge carriers (electrons or holes) between cations with different charge states residing in the B-site. Thus, conductance in ZnFe₂O₄ occurs by the hopping of electrons between $F{e}^{3+}$ and $F{e}^{2+}\left(F{e}^{3+}+e^{-}\leftrightarrow F{e}^{2+}\right)$ in the octahedral site [43,44] through the so called Verwey conduction model proposed by E.J. Verwey et al. [45] in 1947. Although a high concentration of Fe²⁺ in the B-site results in high conductance, it is also important to note that n-type spinel ferrite gas sensors have a slightly lower stoichiometric composition and contain couples of Fe²⁺ and Fe³⁺ in the octahedral site even prior to oxygen chemisorption [20].

In line with the well-accepted SMO-based sensor mechanism, oxygen species are adsorbed on the surface of the material. At elevated temperatures, the oxygen molecules (O₂) convert to atomic oxygen, thus retracting electrons from the surface of the material as described by Equations (1)–(4). Following the above discussion, chemisorbed oxygen on the surface of ZnFe₂O₄ will retract electrons from the Fe²⁺ oxidising it to Fe³⁺, consequently increasing resistance, as illustrated in Figure 1. The interaction of chemisorbed oxygen with a reducing gas then leads to the release of electrons back into the surface, which reverses the process, and a change in resistance is recorded according to Equation (5). The reduction reaction reduces the Fe³⁺ to Fe²⁺ which increases the conductivity of ZnFe₂O₄. The change in the conductivity of the ZnFe₂O₄ is measured as the sensor response.

$$O_2\left(gas\right)\leftrightarrow O_2\left(ads\right)$$

(1)

$$O_2\left(ads\right)+e^{-}\leftrightarrow O_2^{-}\left(ads\right)T<147°\mathrm{C}$$

(2)

$$O_2^{-}\left(ads\right)+e^{-}\leftrightarrow 2O^{-}\left(ads\right)147°\mathrm{C}<T<397°\mathrm{C}$$

(3)

$$O^{-}\left(ads\right)+e^{-}\leftrightarrow O^{2-}\left(ads\right)T>397°\mathrm{C}$$

(4)

$$Targetgas\left(ads\right)+O^{-}\to Gasproducts+e^{-}\hspace{1em}147°\mathrm{C}<T<397°\mathrm{C}$$

(5)

3. Challenges and Limitations of ZnFe₂O₄ Sensors

Patents on preparation methods and prototypes of ZnFe₂O₄ sensors have been filed; however, there are not commercially available ZnFe₂O₄-based gas sensors. This could be due to several challenging issues hindering practical applications that are not yet resolved. There are several requirements that a sensor must meet for a specific application. For instance, in an agricultural setting, the environment surrounding raw and processed materials is inherently complex. Gas sensors can be employed to monitor environmental parameters in a changing climate, allowing producers to conduct efficient irrigation and pest control. Therefore, the gas sensors in those environments should be able to function under extreme pressures and temperatures [47]. Similar to the food industry, sensors need to meet certain standards to be employed for a particular application. Such standards can be evaluated by accessing the sensing properties of the developed SMO. Like most SMO-based sensors, ZnFe₂O₄ based sensors have limitations that hinder their practical applications, including: (i) operating temperature, which is one of the important parameters for identifying the optimal sensing performance of a gas sensor. The operating temperatures of ZnFe₂O₄ based sensors are reported to be between 100 °C and 400 °C. Such high temperatures can cause thermal instability within the ZnFe₂O₄ structure, affecting the long-term stability of the sensor. Furthermore, gas sensors that operate at high operating temperatures suffer from degradation of electrode contacts, causing drastic changes in sensor properties and sensor design challenges [48]. (ii) Secondly, the performance of a sensor in harsh, humid conditions is an important aspect for commercial gas sensors. ZnFe₂O₄ sensors are extremely sensitive to humidity and have been reported to be good humidity sensors. However, this is an issue for the detection of VOCs in applications such as diabetes diagnosis or monitoring of food products in refrigeration systems where moisture cannot be avoided. (iii) Thirdly, the gas selectivity of the ZnFe₂O₄ sensors is another sensing property that needs careful consideration. The selectivity of a sensing material is evaluated by the ratio of the response toward a particular gas to that of another. SMO materials are not selective in nature, but they can be modified to react more to a specific target gas. ZnFe₂O₄ already shows good sensing characteristics and is highly selective for H₂S and VOCs [49]. The challenge it faces is the selectivity towards a specific gas among a complex mixture of VOCs.

4. Strategies for the Enhancement of ZnFe₂O₄ Sensing Properties

To date, several studies have been conducted investigating the sensing properties of ZnFe₂O₄ towards VOCs, including ethanol [50], acetone [25], formaldehyde [51], BTX [52], etc. As outlined, most of the investigated VOCs are found in almost all industries, and this positions ZnFe₂O₄ as a potential gas sensor for wide application. To meet the requirements for application in these fields, ZnFe₂O₄ gas sensors should be able to detect very low concentrations of VOC and demonstrate high selectivity. Research and development using different strategies has been conducted in the past few decades targeting specific gases.

4.1. Effects of Zn²⁺ Metal Ion Substitution

The types of cations, as well as the compositions and site occupancies, determine the physical, chemical, and electrochemical properties of spinel-type ferrite-based metal oxides because they can accommodate a wide range of cations with more than one oxidation state between the tetrahedral and octahedral sites [53,54]. Metal-ions such as Ni²⁺ and Cu²⁺ can substitute Zn²⁺ in the lattice structure of ZnFe₂O₄. Due to the different properties each metal-ion possesses, substitution of Zn²⁺ by other metal-ions can fine tune the properties of ZnFe₂O₄ to suit different applications [55,56,57,58,59]. Depending on the size of the substituted metal-ion, the structural features of ZnFe₂O₄ can be altered with possible changes in the lattice structure, crystal size, and morphology. For instance, Shen et al. [60] substituted Zn²⁺ with Co²⁺ using the formula Co_1−xZn_xFe₂O₄ with 0.0 ≤ x ≤ 0.5 and demonstrated that the ‘a’ lattice parameter decreased with an increase in Zn content until x = 0.3, where the lattice parameter increased dramatically thereafter. This dependence of the lattice parameter on the substitution of metal-ions was also observed by Chakrabarty et al. [61], where they substituted Zn²⁺ with Ni²⁺. In their work, they showed that both the ‘a’ lattice parameter and crystal size decreased with an increase in Ni²⁺ concentration, with crystal size changing by 1.86 nanometers at full substitution.

During substitution of Zn²⁺ in the tetrahedral site, redistribution of cations from the tetrahedral site to the octahedral site may occur, with the substituted metal-ion occupying the octahedral position, migrating the Fe³⁺ to the tetrahedral site [62]. This alters the conductivity of the spinel and, ultimately, its sensing properties. A common substitution in gas sensing is the Zn²⁺ by Ni²⁺ of ZnFe₂O₄. This substitution has been reported to reduce the operating temperature of the sensor by invoking two possible adsorption pathways at the tetrahedral and octahedral sites [63]. Furthermore, Ni²⁺ ions preferably occupy octahedral sites and participate in spinel conductivity [55]. A higher concentration of Ni²⁺ in the octahedral site promotes greater oxygen adsorption, which oxidizes Ni²⁺ to Ni³⁺ to maintain electrical neutrality, resulting in more surface oxygen reacting with the target gas and a higher response [20]. In a previous report, the substitution of Zn ions by 10% Ni ions resulted in a unique electronic structure with the correct charge concentration, enhancing the sensing performance [64]. However, to achieve much better sensing properties in Ni²⁺ substituted ZnFe₂O₄, it is necessary to control the right amount of Fe²⁺ and Ni²⁺ in the octahedral site, and research around this requires further development. Furthermore, different metal-ion substitutions have previously been demonstrated to enhance sensing properties.

VOCs. For example, Wu et al. [65] prepared copper-substituted ZnFe₂O₄ hollow micro-spheres using facile solvothermal and annealing technology. As displayed in Figure 2, the Cu²⁺ substituted ZnFe₂O₄ ferrites showed a high response to a very low concentration of acetone (0.8 ppm). The enhanced sensing properties of the Cu²⁺ substituted ZnFe₂O₄ sensors were attributed to the effects of Cu²⁺ on the lattice cation distribution, which resulted in a higher electron depletion layer and adsorbing capacity. It is important to note that not all substitutions enhance sensing properties. For instance, substitution of Cu²⁺ in ZnFe₂O₄ significantly enhances its sensing properties, while substitution of Zn²⁺ into CuFe₂O₄ nanostructures reduces the response. However, the relation between sensing properties and the number of substituted metal-ions varies with different target gases [63].

4.2. Effects of Structural Morphology

Metal-ions exposed to different facets have certain adsorption interactions with gas molecules. Due to facets being flat surfaces on a specific geometric shape or morphology, the surface structure, corners, and edges of the SMO have a stronger relationship to the exposed facet [66]. Based on this, different gas molecules prefer to adsorb in different formations on particular surfaces of ZnFe₂O₄ [67]. To enhance the sensing properties of ZnFe₂O_4, numerous efforts have been dedicated to synthesizing different structural morphologies such as nanotubes, nanospheres, and nanofibers that will expose the preferred facet for gas adsorption. However, the effects of surface geometry have not been extensively studied in gas sensing due to challenges pertaining to morphology manipulation. On another note, SMOs with a high surface area and low density are preferred for gas sensing, as they generally exhibit good sensitivity [68]. The surface area depends on morphology as a result of variations in the surface-to-volume ratio. Furthermore, the porosity accompanying the surface area allows access to different active sites and facets, which have consequences for sensitivity and selectivity [69]. The effect of porosity on ZnFe₂O₄ sensing properties was demonstrated by Zhang et al. [68], where the porous structure allowed the detected gas to diffuse through both the outer and inner surfaces of the tubes (Figure 3). This morphology and the small grains that make up the unique interconnected channel structure improved the detection behavior of ZnFe₂O₄. Due to the dense structure of a compact film, the ZnFe₂O₄ nanoparticle-based sensor had a relatively low surface-to-volume ratio and could only detect gases effectively using a thin layer of the film near the surface.

In another study conducted by Zhou et al. [70], ZnFe₂O₄ porous nanospheres displayed excellent selectivity towards 30 ppm acetone and a remarkable response of 11.8, which was 2.5 times higher than that of ZnFe₂O₄ nanoparticles. The porous structure of the nanospheres that exposes the active surface of different facets was found to be the main reason for the observed remarkable sensing performance. On the other hand, Dong et al. [71] reported on the synthesis of monodisperse ZnFe₂O₄ nanospheres for the detection of toluene. Due to the large specific surface area, they could detect a low concentration of toluene, attaining a response of 1.41 at 1 ppm. Such enhancement in the sensing performance of porous ZnFe₂O₄ was attributed to diffusion and the target gas being exposed to a higher active surface area.

Certain facets are more reactive than others due to active sites that are more distinguished by the arrangement of surface atoms. To confirm this concept, Zou et al. [67] applied density functional theory (DFT) to study the adsorption sites and the interaction mechanism of the NH₃ molecule on the surface of ZnFe₂O₄. It was concluded that NH₃ molecules preferred to adsorb with the formation of a H₃N-Zn bond on the surface (110) of the spinel ZnFe₂O₄. Thus, the adsorption of target gas molecules depends on the preferred surface and active sites of high-energy surfaces. Studies directly focusing on determining high-energy facets of ZnFe₂O₄ through experimental and computational studies are still limited; however, research has been focused on preparing morphologies that can expose such facets to enhance the sensing performances. ZnFe₂O₄ hollow spheres that showed excellent selectivity were prepared by Yang et al. [72] for the detection of ethylene glycol. Li et al. [73] synthesized porous ZnFe₂O₄ nanorods with a networked pore structure for the detection of acetone. Table 2 compares the performance of different ZnFe₂O₄ sensors with different morphologies, while Figure 4 displays some of the morphologies synthesized from the literature. Qu et al. [74] conducted a clear comparison of the effect of morphology on ZnFe₂O₄, which emphasizes the importance of morphology on the sensing properties of ZnFe₂O₄, the response and selectivity from solid, yolk-shell to double-shell, which were attributed to porous structures (Figure 5).

Figure 4. SEM micrographs of different morphologies of (a) porous ZnFe₂O₄ olives. (Reproduced with permission from [75], Copyright 2019, Elsevier). (b) porous ZnFe₂O₄ spheres. (Reproduced with permission from [70], Copyright 2014, Elsevier). (c) ZnFe₂O₄ tubes (Reproduced with permission from [68], Copyright 2007, Elsevier). and (d) ZnO/ZnFe₂O₄ nanocubes. (Reproduced with permission from [76], Copyright 2017, RSC Advances).

Figure 4. SEM micrographs of different morphologies of (a) porous ZnFe₂O₄ olives. (Reproduced with permission from [75], Copyright 2019, Elsevier). (b) porous ZnFe₂O₄ spheres. (Reproduced with permission from [70], Copyright 2014, Elsevier). (c) ZnFe₂O₄ tubes (Reproduced with permission from [68], Copyright 2007, Elsevier). and (d) ZnO/ZnFe₂O₄ nanocubes. (Reproduced with permission from [76], Copyright 2017, RSC Advances).

Table 2. ZnFe₂O₄ based sensors with different morphologies and their corresponding specific surface area morphologies and performance responses towards different VOCs.

ZnFe2O4 Morphology	Surface Area (m2g−1)	Target Gas	Concentration (ppm)	Reproducibility (Cycles)/Stabilty (Days)	Ra/Rg	Ref.
Mesoporous nanostructures	103.6	Acetone	100	-/30	11.6	[77]
Porous Olives	96.5	Ethanol	200	8/7	223	[75]
Porous nanospheres	59.0	Acetone	30	3/30	11.8	[70]
Monodisperse nanospheres	87.40	Toluene	1/100	5/20	1.41/9.98	[71]
Hollow urchin-like core-shell spheres	50.62	Toluene	100	6/30	79	[78]
Porous nanospheres	59.0	Acetone	100	3/30	42.1	[70]
Porous nanorods	82.01	Acetone	100	-/30	52.8	[73]
Hollow spheres	48.1	Ethylene glycol	100	-/8 weeks	35.5	[72]

Table 2. ZnFe₂O₄ based sensors with different morphologies and their corresponding specific surface area morphologies and performance responses towards different VOCs.

ZnFe2O4 Morphology	Surface Area (m2g−1)	Target Gas	Concentration (ppm)	Reproducibility (Cycles)/Stabilty (Days)	Ra/Rg	Ref.
Mesoporous nanostructures	103.6	Acetone	100	-/30	11.6	[77]
Porous Olives	96.5	Ethanol	200	8/7	223	[75]
Porous nanospheres	59.0	Acetone	30	3/30	11.8	[70]
Monodisperse nanospheres	87.40	Toluene	1/100	5/20	1.41/9.98	[71]
Hollow urchin-like core-shell spheres	50.62	Toluene	100	6/30	79	[78]
Porous nanospheres	59.0	Acetone	100	3/30	42.1	[70]
Porous nanorods	82.01	Acetone	100	-/30	52.8	[73]
Hollow spheres	48.1	Ethylene glycol	100	-/8 weeks	35.5	[72]

Figure 5. (a) Comparison of sensor responses of ZnFe₂O₄ double-shell, york-shell, and solid microsphere-based sensors to 20 ppm acetone as a function of the operating temperature and (b) ZnFe₂O₄ double-shell, york-shell, and solid microsphere-based sensors to various gases at 206 °C. Reproduced with permission from [74], Copyright 2018, Elsevier.

Figure 5. (a) Comparison of sensor responses of ZnFe₂O₄ double-shell, york-shell, and solid microsphere-based sensors to 20 ppm acetone as a function of the operating temperature and (b) ZnFe₂O₄ double-shell, york-shell, and solid microsphere-based sensors to various gases at 206 °C. Reproduced with permission from [74], Copyright 2018, Elsevier.

4.3. Effects of Generated Surface Defects

For spinel structures, cation mixing can create centers of local charges that serve as electron or hole traps that lead to a point defect (also called an antisite defect) [79]. This is often obtained after a heat treatment process where the spinel structure normally undergoes an order-disorder transition. Two of the most common defects that can occur in ZnFe₂O₄ are Schottky and Frenkel defects. The Schottky defect involves simultaneous movement of the cation and anion to the surface, while the Frenkel defect occurs either on the cation sub-lattice or on the anion sub-lattice. These two defects are created independently, and one usually dominates the other. El-Sayed [80] investigated the electrical conductivity of Ni_1−xZn_xFe₂O₄ (x = 0.1, 0.3, 0.5, 0.7, and 0.9). The type of defect was identified by carefully measuring the relative variation in densities ΔD/D₀ and lattice constant Δa/a₀, where D₀ denotes the true density calculated by XRD analysis and ΔD represents the difference between the true and obtained bulk densities of the prepared samples. If ΔD/D₀ > Δa/a₀, then the Schottky defect dominates, while the Frenkel defect dominates when ΔD/D₀ is approximately equal to Δa/a₀ [80]. In this study, the produced material showed a Schottky defect dominating for all values of x.

Structural defects are regarded as the most important parameter in gas sensing, as they influence the chemisorption characteristics of the sensor, leading to enhanced sensing performance. As described above, it is understood that the gas sensing of ZnFe₂O₄ nanostructures is a result of resistance modulation that comes from electron hoping between iron ions located in the octahedral site (Fe³⁺ + e⁻ ↔ Fe²⁺). A higher amount of Fe²⁺ in the octahedral site can increase oxygen chemisorption, which then results in higher resistance modulation. Thus, maintaining the right concentration of Fe²⁺ in the octahedral site without undermining the structural stability of the ZnFe₂O₄ FCC structure is crucial to enhancing its sensing properties. Fe²⁺ can be regulated by synthesizing ZnFe₂O₄ with the non-stoichiometry of iron ions. Sutka et al. [81] prepared a non-stoichiometry Fe excess ZnFe₂O₄, and when compared to the stoichiometric ZnFe₂O₄, the sensitivity towards VOCs increased by threefold. Fe²⁺ can also be set by controlling surface defects; in particular, oxygen vacancies can be compensated by reducing Fe³⁺ to Fe²⁺ which then results in higher oxygen chemisorption. The presence of oxygen vacancies can also narrow the band gap of ZnFe₂O₄ by raising the position of the valence band, as illustrated in Figure 6. Such a narrow band gap can result in more electron transfer, which in turn results in a higher gas response.

4.4. Effects of Surface Doping

In an attempt to produce sensing materials that exhibit improved sensor performance, researchers have focused their attention on improving the chemical composition through surface decoration. Surface decoration involves the addition of catalytically active sites to the surface of the sensing material. Such active sites will enhance the sensing performance by favoring certain interactions of gas species with the surface and reducing the response times of the sensors [83]. The most common surface decoration species are noble metal nanoparticles (i.e., Au, Ag, Pt, Pd, etc.) due to their multiple sensitization effects [84]. Two sensitization mechanisms can occur that are influenced by surface decorations, including chemical and electronic sensitization. In electronic sensitization, noble metal nanoparticles incorporated on the surface act as a strong electron acceptor, thus increasing the amount of Fe²⁺ in the octahedral and increasing the depletion around the interface. In chemical sensitization, incorporated noble metal nanoparticles provide preferred adsorption sites, which create active sites that are spilled over onto the ZnFe₂O₄ surface to react with target gases. Therefore, the observed sensing enhancement in ZnFe₂O₄ through surface decoration is often attributed to both electronic and chemical sensitization. However, chemical sensitization is a more logical phenomenon for the improvement of sensing properties, given that catalytic reactions on SMO surfaces serve as the receptor function of SMO-based gas sensors and that catalytically active metal nanoparticles are used as additives to strengthen these reactions to increase sensor selectivity and sensitivity of the sensors [85,86].

Au-decorated ZnFe₂O₄ was recently prepared by Li et al. [26] using the hydrothermal method for the detection of chlorobenzene. In their study, it was found that the Au decorated ZnFe₂O₄ yolk-shell sphere revealed an improved sensing performance compared to the pristine ZnFe₂O₄ yolk-shell sphere owing to both electronic and chemical sensitizations induced by Au addition to the surface of ZnFe₂O₄ yolk-shell spheres. The addition of Au particles particularly resulted in a response of 90.9 toward 10 ppm of chlorobenzene, a detection limit of 100 ppb, and outstanding selectivity. It was then concluded based on this finding that the Au-decorated ZnFe₂O₄ yolk-shell sphere-based sensor can be used for monitoring chlorobenzene in the food industry. Nemufulwi et al. [87] evaluated the effects of Au on ZnFe₂O₄ nanoparticles. It was also discovered that the addition of Au metal nanoparticles to the ZnFe₂O₄ surface enhanced the sensing behavior of nanostructured ZnFe₂O₄. In another study, Lv et al. [88] doped ZnFe₂O₄ with an Sb metalloid. They achieved a response of 35.5 from the 0.5% Sb-doped ZnFe₂O₄, which was twice as high as that of pure ZnFe₂O₄. Doping was discovered to increase the specific surface area in the Sb-doped ZnFe₂O₄ samples, increasing the number of active adsorption sites for the target gases. Furthermore, an appropriate amount of element Sb on the surface of ZnFe₂O₄ increased the concentration of deficient and surface chemisorbed oxygen, further improving the sensing performance. In this case, the element Sb played a significant role by acting as a catalyst that strengthened surface reactions by increasing the amount of chemisorbed oxygen.

4.5. Heterostructures

SMOs come in two varieties: p-type semiconductors, where holes make up the majority of the charge carriers, and n-type semiconductors, where electrons make up the majority of the charge carriers. Numerous research studies have demonstrated that composite metal-oxides can be used to enhance selectivity, sensitivity, and other crucial sensing qualities. Employing SMOs with similar or dissimilar conductivity, metal heterostructures can be formed, whereby different types of SMOs can form a physical interface [89]. Depending on the bandgap of the chosen SMO to form an interface with ZnFe₂O₄, most of the heterojunctions formed are in the context of type I and type II band-gap alignment. A type I configuration is characterized by straddling band alignment, where at the heterojunction the edges of the valance and conduction band are located within the base SMO or vice versa, forming a nested type I heterostructure, as shown in Figure 7. In type II heterostructures, the adjacent domains of the SMOs form a staggered alignment at the heterojunction. The valence and conduction bands of the added SMO are either higher or lower than those of the base SMO. Heterostructures have attracted attention due to their potential to enhance sensing sensitivity by modulating resistance through heterojunction barriers [90].

Several structural architectures that are used to produce heterojunction between two SMOs have been developed, i.e., those with simple mixtures that can be randomly distributed throughout the materials (e.g., graphene-ZnFe₂O₄) [91], a base material with an SMO deposited on its surface (e.g., CuO@ZnFe₂O₄) [92], and a well-defined partition between the two SMOs (e.g., ZnO/ZnFe₂O₄) [76]. The most common interface formed for gas sensing applications is the p-n heterojunction [93]. However, in each of the structural architectures involving the n-type ZnFe₂O₄ nanostructure, several heterojunctions can be formed to improve the sensing properties. Enhanced sensor performance can arise from charge transfer and the creation of a charge depletion layer when two distinct SMOs are in close electrical contact and the fermi levels across the interface are equilibrated to the same energy [93]. The charge transfer was perfectly illustrated by W. Li et al. [94] in the case of the ZnO-ZnFe₂O₄ heterojunction, as illustrated in Figure 5. To separate the excited electron-hole pairs on the ZnFe₂O₄ side of the heterointerfaces, the excited electrons of ZnFe₂O₄ are driven to the conduction band of ZnO. However, excited holes in the ZnO valence band are preferentially captured and accumulated in the ZnFe₂O₄ valence band, which has the most negative valence band of ZnFe₂O₄. Therefore, a new potential barrier is created between the ZnO-ZnFe₂O₄ interfaces. In the literature survey conducted here, ZnFe₂O₄ nanostructures have been used more often to create the n-n heterojunction for gas-sensing applications. Wang et al. [95] prepared ZnO/ZnFe₂O₄ with hollow nanocages that showed enhanced sensing performance to 100 ppm acetone as compared to ZnO nanocages and ZnFe₂O₄ nanospheres. Furthermore, the ZnO/ZnFe₂O₄ heterostructure nanocages exhibited a low detection limit of 1 ppm for acetone, which shows good applicability in the food industry. The improved gas detection performance was later determined to be due to the distinctive structure and heterojunction created at the ZnO and ZnFe₂O₄ interfaces. In a different investigation, Zhang et al. [92] examined the gas sensing capabilities of ZnFe₂O₄ both before and after CuO surface modification. The results showed that the CuO@ZnFe₂O₄ heterostructures had a greater gas response and higher xylene selectivity. In contrast, Li et al. [96] created ZnFe₂O₄/α-Fe₂O₄ porous microrods with improved TEA sensing performance using a quick and reliable sacrificial template method. Three types of n-n junctions, including heterojunctions of ZnFe₂O₄-Fe₂O₃ and homojunctions of ZnFe₂O₄-ZnFe₂O₄ and Fe₂O₃-Fe₂O₃, were anticipated to occur in the nanocomposite when ZnFe₂O₄ nanoparticles were placed in Fe₂O₄. While only the Fe₂O₃-Fe₂O₃ homojunction is present in pure-Fe₂O₃, the enhanced TEA sensitivity is believed to be mainly due to the newly created ZnFe₂O₄-Fe₂O₃ heterojunction. Figure 5 displays the TEM of the microrods and the corresponding SAED displaying the diffraction rings, which were indexed to rhombohedral a-Fe₂O₄ and cubic ZnFe₂O₄. The HR-TEM image clearly confirms a heterojunction between ZnFe₂O₄ and a-Fe₂O₄, as displayed in Figure 8. Table 3 lists the literature on ZnFe₂O₄ heterostructures and their responses to several gases.

Figure 7. Schematic energy-level diagrams of hollow ZnO/ZnFe₂O₄ microspheres with a down/up-shift of actual Fermi-level modulated by thermally dependent ionization reactions: (a) room temperature, (b) low temperature, (c) middle temperature, and (d) high temperature. Reproduced with permission from [94], Copyright 2017, Elsevier.

Figure 7. Schematic energy-level diagrams of hollow ZnO/ZnFe₂O₄ microspheres with a down/up-shift of actual Fermi-level modulated by thermally dependent ionization reactions: (a) room temperature, (b) low temperature, (c) middle temperature, and (d) high temperature. Reproduced with permission from [94], Copyright 2017, Elsevier.

Figure 8. (a,b) TEM images; (c) SAED pattern; and (d) HRTEM image of the prepared ZnFe₂O₄/a-Fe₂O₃ PMRs. Reproduced with permission from [96], Copyright 2018, Elsevier.

Figure 8. (a,b) TEM images; (c) SAED pattern; and (d) HRTEM image of the prepared ZnFe₂O₄/a-Fe₂O₃ PMRs. Reproduced with permission from [96], Copyright 2018, Elsevier.

Table 3. Literature of different ZnFe₂O₄ heterostructures towards different target gases.

ZnFe2O4 Heterostructure	Morphology	Interface/ Heterojunction	Target Gas	Concentration (ppm)	Reproducibilty (Cycles)/Stability (Days)	Ra/Rg	Ref.
ZnO/ZnFe2O4	Lychee-like core, shell, and hollow microsphere	n-n junction	Acetone	100	-/20	36.6	[97]
CuO@ZnFe2O4	Yolk shell microspheres	p-n junction	Xylene	100	5/60	24	[92]
ZnFe2O4/ZnO	Flower-like microstructures	n-n junction	Acetone	50	6/60	8.3	[98]
ZnO/ZnFe2O4	Microsphere	n-p-n junction	Benzene	1	-/-	1.1	[94]
ZnO/ZnFe2O4	Tetrapod-like	n-n junction	Ethanol	100	-/30	13.97	[73]
ZnFe2O4/ZnO	Nanosheets assembled into microspheres	n-n junction	Trimethylamine	100	10/30	31.5	[99]
ZnO/ZnFe2O4	Hierarchical kiwifruit-like	n-n junction	Triethylamine	100	6/30	40.15	[98]
ZnFe2O4/α-Fe2O3	Porous microrods	n-n junction	Triethylamine	100	4/30	42.4	[96]
Fe2O3/ZnFe2O4	Porous spindles	n-n junction	Triethylamine	20	3/-	60.24	[100]
CaFe2O4/ZnFe2O4	Walnut	p-n junction	Isoprene	30	9/30	19.5	[101]
ZnFe2O4/(Fe-ZnO)	Nanocomposite	n-n junction	Acetone	100	3/-	30.8	[102]

Table 3. Literature of different ZnFe₂O₄ heterostructures towards different target gases.

ZnFe2O4 Heterostructure	Morphology	Interface/ Heterojunction	Target Gas	Concentration (ppm)	Reproducibilty (Cycles)/Stability (Days)	Ra/Rg	Ref.
ZnO/ZnFe2O4	Lychee-like core, shell, and hollow microsphere	n-n junction	Acetone	100	-/20	36.6	[97]
CuO@ZnFe2O4	Yolk shell microspheres	p-n junction	Xylene	100	5/60	24	[92]
ZnFe2O4/ZnO	Flower-like microstructures	n-n junction	Acetone	50	6/60	8.3	[98]
ZnO/ZnFe2O4	Microsphere	n-p-n junction	Benzene	1	-/-	1.1	[94]
ZnO/ZnFe2O4	Tetrapod-like	n-n junction	Ethanol	100	-/30	13.97	[73]
ZnFe2O4/ZnO	Nanosheets assembled into microspheres	n-n junction	Trimethylamine	100	10/30	31.5	[99]
ZnO/ZnFe2O4	Hierarchical kiwifruit-like	n-n junction	Triethylamine	100	6/30	40.15	[98]
ZnFe2O4/α-Fe2O3	Porous microrods	n-n junction	Triethylamine	100	4/30	42.4	[96]
Fe2O3/ZnFe2O4	Porous spindles	n-n junction	Triethylamine	20	3/-	60.24	[100]
CaFe2O4/ZnFe2O4	Walnut	p-n junction	Isoprene	30	9/30	19.5	[101]
ZnFe2O4/(Fe-ZnO)	Nanocomposite	n-n junction	Acetone	100	3/-	30.8	[102]

5. Applications of ZnFe₂O₄ Gas Sensor

Gas sensing is of great importance in various fields such as environmental safety, food safety and fermentation control in the food industry, safety from toxic and explosive gases, detection of combustible gases in households, and patient diagnostics. Several sensing methods have been adopted for various sensing applications in different fields. Numerous findings reported in the literature on ZnFe₂O₄ based sensors do not describe their direct application in a specific field. Instead, reports are specific about the target gas, and sensing measurements are conducted in practical experiments that simulate real-life applications. Several reports have demonstrated the potential of ZnFe₂O₄ as an acetone detector [25,70,73,76,87,91,102,103]. In general, acetone is a type of important organic synthesis raw material that is obtained and widely used in agricultural chemicals, plastics, rubber, medicine, coatings, and commercial production, such as spray paint [104]. Although acetone is necessary in industrial production, it is flammable and volatile and can easily ignite when exposed to light or high temperatures, posing a significant risk to commercial production. Furthermore, acetone has a strong impulse with prolonged exposure that causes chronic illness with a paralysis effect produced in the central nervous system, triggering nausea, weakness, dizziness, pharyngitis, bronchitis, and even stupor. As a result, to ensure industrial safety and human health, acetone vapors in commercial production must be detected quickly and accurately. This only requires high sensitivity, a quick response time, and a long-term, stable acetone gas sensor. Furthermore, the breath of a diabetic person releases acetone, suggesting high levels of ketones in their blood. Therefore, acetone can be detected in human breath for the diagnosis of diabetes at concentrations above 1.8 ppm. ZnFe₂O₄ based sensors have lower acetone detection limits in ppb, making them good candidates for medical diagnosis [74].

In addition, in the food sector, highly sensitive ZnFe₂O₄ can be used to monitor and adjust the ethanol content in fruit ripening storage chambers online or with an alcoholmeter [75]. Previous reports have also demonstrated the utilization of ZnFe₂O₄ sensors in the chemical and production industries for the detection of neurotoxic gases such as benzene, toluene, xylene, styrene, and trichloroethylene [105]. Benzene, toluene, and xylene gases, which are also collectively known as BTX, are frequently used in industries and require sensitive gas sensors to detect them. A full discussion of the toxicity levels of these gases was reported by Mary Ellen Fleming-Jones and Robert E. Smith [105].

ZnFe₂O₄ is also sensitive to triethylamine (TEA), which is considered an organocatalyst and solvent with a wide range of applications. It is often employed in industrial manufacturing as a synthetic dye and preservative due to its relative safety, commercial availability, and low price [106]. Due to its good physical and chemical characteristics, it is also frequently utilized in large amounts in chemical investigations. However, even at modest levels of exposure, TEA can cause a variety of other health problems when the concentration is too high. It can also cause pulmonary swelling and poisoning when inhaled [107]. TEA is also explosive, and sensitive ZnFe₂O₄ can be used for its detection.

Table 4. Different applications of ZnFe₂O₄ based gas sensors.

Gas	Source of Emission/Product	Detection Test/Application	SMO Used	Detection Range (ppm)/Environmental Setting	Ref.
Acetone and Ethanol	Household and industrial products, laboratories, and chemical industries	Safety purposes	Mg0.5Zn0.5 Fe2O4	20–200/dry air RH, RT	[108]
Propanol	Alcoholic beverages	Quality and classification of wines	ZnFe2O4 NPs	0.5–40/10–80%, 120 °C	[64]
Acetone	Human breath, fish products	Diabetes diagnostics and spoilage detection	ZnFe2O4 nanoparticles	0.5–40/0–60%, 120 °C	[109]
n-butanol	petroleum refineries and insect repellents,	Human health and safety	Sb-doped 3D ZnFe2O4 MPs	0.049–200/25–90 RH, 250 °C	[88]
Toluene	lacquers, medicine, pesticides, leather manufacturing, and explosives	Environmental and human safety, Human health	urchin-like hollow core-shell ZnO/ZnFe2O4 ZnFe2O4	0.2–100/10–98% RH, 275 °C 0.2–100/10–98% RH, 250 °C	[78]
Toluene	Chemical Industry	Environmental safety and human safety	Monodisperse ZnFe2O4 nanospheres	1–100/Dry air. 300 °C	[71]
Acetone	Human breath	Acetone-threat or diabetes-breathalyzer tests	Double-shelled ZnFe2O4 microsphere	0.13–200/Dry air, 206 °C	[74]
Acetone	Human breath, industry	Diagnosis of diabetes, industrial processes, and health control	ZnO/ZnFe2O4/Au 0.125%Graphene-ZnFe2O4 ZnFe2O4 hollow sphere and Ag-ZnFe2O4	0.3–200/33–95% RH, 225 °C 1–1000/Dry air, 275 °C 0.8–500/25–100% Rh, 175 °C	[27,91,110]
Ethanol	Human breath; alcohol beverages	Medical or clinical applications; brewing process control.	ZnFe2O4 thin film	1–50/Dry air, 390 °C	[50]
Triethylamine (TEA)	Meat, wastewater, dead fish, and marine products	Meat spoilage, environmental monitoring, and human health protection	MOF-ZnO/ZnFe2O4	2–100/Dry air, 170 °C	[111]
TEA	Industrial production	Industrial monitoring	kiwifruit-like ZnO/ZnFe2O4 ZnO/ZnFe2O4	1–200/dry air, 200 °C 5–1000/dry air, 80 and 240 °C	[112,113]
Propanol	Industrial plants	Leak detection	ZnFe2O4/ZnO polymer nanocomposite	500–5000/Dry air, N/A	[114]

lists some of the recent developments in ZnFe₂O₄ sensors targeting different applications.

6. Future Perspective

In the food industry, several molecules can contribute to the aroma and odor of food; thus, the selectivity problem becomes more imperative for sensor application. While standard techniques such as GC/MS can isolate and quantify the molecular information of VOC mixtures, concentration extraction methods are required to detect VOCs below the ppm range, and complex instrumentation makes this impractical for daily use. On the other hand, ZnFe₂O₄ based sensors suffer several limitations, as outlined in Section 3. Poor selectivity is by far the biggest limitation hindering the wide practical application of SMO-based sensors. This is because SMO-based sensors lack an adequate ability to discriminate between one gas and a mixture of gases. As discussed in previous sections, several attempts have been made to address this selectivity issue. However, while these strategies have been found to increase the response values towards certain gases, the total discrimination of the interfering gas remains a major challenge. Artificial olfactory systems have been proven to be effective methods for resolving the selectivity problem in SMO-based sensors [115]. These systems use cross-sensitive olfactory receptor arrays, olfactory codes, and the brain’s recognition system to imitate the biological olfactory organ’s ability to discriminate [115]. Similar to natural smart sensor systems, artificial smart sensor systems can be built that include output vectors, pattern recognition algorithms, and gas sensor arrays.

Smart sensors can wirelessly capture sensory data from the Internet of Things (IoT) and transmit it to the cloud, enabling machine learning (ML) algorithms to be used for pattern recognition and prediction techniques to address the selectivity problem [116]. Algorithms applicable to SMO sensors focus on three parts: (a) preprocessing of data; (b) dimensionality and feature extraction; and (c) recognition systems to identify and classify different gas types. The first part looks at correcting the base line of the sensor response caused by drift. Most SMOs suffer from drift, whereby the resistance in the air changes after exposure to the target gas. This may be caused by an inability to desorb all target gas molecules when the sensor is no longer exposed to them or the diffusion of oxygen vacancies that causes a change in conductivity in the space charge layer [117]. The second part is to identify meaningful characteristics to describe the sensor output specific to each gas, and then to classify or predict the last part. Figure 9 shows the systematic procedures adopted in the gas sensor array. These algorithms have not been used in the case of ZnFe₂O₄ gas sensors. In this case, nanostructured ZnFe₂O₄ sensors can be used in a gas sensor array to improve performance in a specific application.

7. Conclusions

In summary, ZnFe₂O₄ nanostructures are good candidates for application in different industrial sectors due to their ability to easily sense various gases that are generally found in different industries. Applications such as food quality control and ZnFe₂O₄ need to be further developed for good stability, excellent selectivity, and sensitivity under specific ambient conditions. Several strategies to achieve enhanced performance can be summarized as follows:

• The development of novel synthesis procedures with excellent control of morphology can create access to more active sites.
• Different noble metals (i.e., Ag, Au, and Pd) can improve the sensing properties by chemically or electronically sensitizing ZnFe₂O₄.
• Introduction of structural defects by controlling the concertation of Fe²⁺ in the octahedral site and narrowing the band-gap of ZnFe₂O₄ to enhance sensing properties.
• Heterostructure design to form hybrid nanocomposites with either p-type or n-type materials, which will result in a new depletion layer that enhances the sensing response.
• Integration of several ZnFe₂O₄ nanostructures with different morphologies to produce a sensor array with enhanced sensing capabilities for specific applications in the food industry.