Optimizing p-n Heterojunction Interfaces: Enhanced Room-Temperature Ammonia Sensing of Polyaniline/α-Fe₂O₃ Nanocomposites

The rapid advancement of technological systems has led to vast vehicular emissions, improper waste management policies, and domestic daily activities, all of which are considered significant sources of environmental pollutants [1]. Monitoring these contaminants is critical to mitigating ecological degradation and safeguarding public health. Gas sensors have emerged as pivotal tools in both industrial and research settings, owing to their ability to detect airborne toxic compounds that pose substantial risks to human health [2]. Prolonged exposure to hazardous gases like carbon monoxide (CO), nitrogen oxides (NO, NO₂), In addition to volatile organic compounds (VOC) can lead to respiratory disorders (e.g., asthma and hypoxia), skin irritation, neurological symptoms (dizziness, drowsiness), gastrointestinal distress (nausea, vomiting), besides the chronic conditions, like carcinogenic effects and pulmonary dysfunction [3]. Regulatory frameworks were established in which the threshold limit values (TLVs) for gaseous pollutants were specified as the maximum permissible atmospheric concentrations, that is typically expressed as parts per million (ppm), considered safe for human exposure. Figure 1 shows the dose-dependent health effects of toxic gas inhalation, underscoring the urgent need for real-time air quality detection systems.

Figure 1. The health impacts of toxic gases on humans

These gases are emitted from two main sources: natural emissions (e.g., volcanic activities) and artificial activities (e.g., industrial processes). Gas sensors convert chemical interactions with target toxic gases into measurable electrical signals, enabling real-time monitoring [4]. One of the important industrial gases is ammonia, which is considered a very dangerous gas because of its toxicity even at very low concentrations.

1.1 Ammonia toxicity

Ammonia (NH₃) is a toxic, colorless gas that poses significant health risks and requires monitoring in medical, industrial, and residential settings. This inorganic compound plays a crucial role in atmospheric, industrial, and biological systems. It is widely utilized across numerous sectors, such as fertilizer manufacturing. detergent products, petroleum refining products, electricity power generation, the rubber industry, food processing, pharmaceuticals, and the automobile sector [5]. However, these industrial applications carry a substantial risk of NH₃ being liberated into the air, leading to serious pollution issues. Even at low concentrations, ammonia can cause skin irritation, ocular discomfort, and irritation of the upper respiratory tract. In addition, this gas may cause dizziness, fatigue, and nausea. At high concentrations, ammonia can lead to harmful consequences, such as cardiac arrest and damage to the reproductive system. The permissible exposure limit for this gas is about 35 ppm for 15 minutes exposure period and 25 ppm as an average over an 8-hour workday, beyond which it poses a clear risk to human health according to the U.S. Occupational Safety and Health Administration (OSHA) [6].

In summary, the ability to achieve high sensitivity detection of NH₃ gas, instant response, accuracy, and reliable performance at room temperature has become essential for healthcare diagnostics, and for industrial safety, in addition to environmental monitoring [7].

Sensors are effective techniques to detect toxic gases. Various types are used with different mechanisms based on the nature of the gas itself, besides the characteristics of every gas. So, this field has witnessed rapid developments to match the fast developments in industrial sectors and to control the toxic gas emissions [8].

The last developments in the sensor industry have increasingly focused on the conductive polymer (CP) such as polyaniline (PANI), which has gained increasing attention in recent times because of the simple and easy synthesis method, adjustable electrical conductivity, and it can work at room temperature. Despite these merits, pure PANI-based sensors show some limitations, such as mild sensitivity, moderate response and recovery duration, in addition to their low selectivity. On the other hand, metal oxides exhibit high thermal stability and high sensitivity, but these oxides need elevated operation conditions, which may on ammonia detection process [9].

To take the utmost benefit of the properties of both, Hybrid composites including PANI and metal oxide have been classified as a promising strategy. In this hybrid composite system, the PANI enhances the electrical conductivity and the mechanical flexibility, while the metal oxide with high surface and strong adsorption sites for ammonia molecules will strengthen the synergistic interaction, leading to improved sensor performance and other sensing properties [10]. The polyaniline (PANI) integrated with metal oxide nanoparticles (e.g., α-Fe₂O₃) has shown exceptional and promising sensing materials. These hybrids capitalize on inherent redox characteristics, tunable electrical conductivity, and the ambient operational stability [11].

Despite all these merits and developments in conductive polymer (CP)/metal oxide nanoparticles, there are challenges, such as fast response, low concentration detection, and high selectivity of ammonia gas. Furthermore, the systematic investigations that focus on the conductive polymer (CP)/metal oxide nanoparticles are still limited and require a deeper understanding to reach an effective design of a high-performance ammonia sensor.

1.2 Gas sensing principles

Chemical sensors operate by detecting the interactions between target gas and sensing materials, which induce the measurable variations in the physical properties, such as electrical resistance, current magnitude, optical absorbance, or mass variation [12]. In conductive polymer (CP) based sensors, key performance metrics, including sensitivity and response kinetics, are enhanced by optimizing the charge transfer dynamics at the gas/polymer interface [13]. These sensors are classified based on their transduction mechanisms, which are controlled by physicochemical interactions between the CP matrix and the gas. These interactions are either chemical redox reactions or physical adsorption that are powerfully affected by the doping state of the polymer, which modulates its electronic structure and charge transport behavior [14]. The doping process works on inserting the charge carriers into the conductive polymer (CP) framework, thereby tuning its conductivity and changing its affinity for specific gas molecules. Upon gas exposure, chemo-electrical transduction happens by gas molecule adsorption or interfacial charge transfer, leading to the generation of detectable physical or electrochemical signals proportional to gas concentration. So, the operational efficiency of sensors is directly governed by the doping level. Besides, sensor activity is controlled by the active sites and the energy barrier for charge exchange during the sensing process [15].

The sensing mechanism mainly depends on the direct interactions of charge transfer of the polymer surface with gas molecules. The inorganic compounds, such as metals or metal oxides, show catalytic activity upon exposure to the target gas. When these oxides are embedded into a polymeric matrix to form nano-hybrids, these nanocomposites demonstrate enhanced gas sensing performance because of synergistic effects [16]. The nanohybrid structure offers plentiful active sites besides a high surface area, enhancing the gas adsorption efficiency. Also, the polymer matrix restrains the inorganic nanoparticles' agglomeration, while the inorganic compounds prevent polymer degradation, thereby ensuring long-term structural stability [17].

Upon exposure to ambient conditions, atmospheric oxygen adsorbs onto the metal oxide surface, acquiring electrons from its conduction band to form ionized oxygen species (e.g., O₂²⁻, O²⁻). This process generates a charge-depletion layer at the oxide surface, which increases the electrical resistance. When the sensor interacts with the target gases, the adsorbed oxygen reacts with the analyte molecules and releases electrons into the conduction band. This electron exchange reduces the width of the depletion layer, thereby modulating the sensor’s conductance [18, 19].

Heterojunctions formed between n-type metal oxides and p-type polymers play a critical role in conductivity modulation. Due to the lower Fermi energy level of metal oxides relative to polymers, electrons transfer from the inorganic phase to the polymeric phase until Fermi-level alignment is achieved. This redistribution reduces the electron density at the oxide surface and interface, elevates the potential barrier, and expands the depletion layer, thereby amplifying the resistance changes [20].

Sensor response dynamics depend on the semiconductor type: n-type materials exhibit increased conductance upon exposure to electron-donating (reducing) gases and decreased conductance when exposed to electron-withdrawing (oxidizing) analytes [21]. Conversely, p-type semiconductors display the opposite behavior, where oxidizing gases enhance conductivity by increasing the hole concentration, while reducing gases diminish it. A comprehensive understanding of these charge-transfer mechanisms and heterostructure interactions is essential for optimizing sensor selectivity and sensitivity in practical applications [22].

The sensing reactions predominantly occur on the sensor layer surface. These reactions take place on the sensor surface, the thin polymer films, as shown in Figure 2. The sensor's sensitivity is mainly depended on the size of the semiconductor particles, which is one of the primary requirements for enhancing the sensor performance. The sensor sensitivity can be defined as the relative change in the resistance of the sensitive thin film, expressed as a percentage per ppm of the applied gas concentration, as shown in Eq. (1):

$S=\frac{\left( {{R}_{g}}-{{R}_{a}} \right)}{{{R}_{a}}}\times 100%$                                 (1)

where, R_a is the electrical resistance of the sensor in air, while R_g are and the presence of gas.

Figure 2. The sensing reaction on the polymer surface

1.3 Response and recovery times

A gas sensor's response time (τ_res) is the time required for the sensor to reach 90% of its maximum or minimum conductance value upon exposure to a reducing or oxidizing gas. Conversely, the recovery time (τ_rec) is the interval needed for the sensor to return to within 10% of the original baseline once the reducing or oxidizing gas flow is stopped. Figure 3 illustrates this measurement by plotting conductance versus time based on the sensor data [23].

Figure 3. Response and recovery times

Table 1 shows the responses of the sensing elements towards oxidizing and reducing gases and their behaviour according to the type of semiconductor elements either n- or p-type.

Table 1. The response of the sensing elements

This work focuses on investigating the structure- property relationship of PANI doped with different loadings of Α-Fe₂O₃ nanoparticles (1 wt.%, 3 wt.%, and 5 wt.%). The sensor will be synthesized, and then it will be characterized using different tests to investigate the structural, morphological, and elemental composition in addition to its performance as an NH₃ sensor will assess reaching the optimal ratio.

Table 6. Sensing properties for NH₃ gas of thin samples

Table 6 shows the NH₃ sensing properties of α-Fe₂O₃/PANI samples. The highest sensitivity was observed for SH3, the composite (PANI + 3% α-Fe₂O₃), with a value of 86.96%, a response time of 48 s, and a recovery time of 48 s, which is attributed to the nanocrystalline phase. The chemically prepared PANI nanofiber films exhibit high surface roughness, providing a large surface area that enhances NH₃ adsorption. The high sensitivity, fast response, and short recovery times can be explained by the catalytic role of the α-Α-Fe₂O₃ nanoparticles, which facilitate the decomposition of gas molecules into free radicals and enhance their interaction with oxygen and functional groups on the polymer surface. So, the behavior of 3% α-Fe₂O₃/PANI (SH3 sample) is optimal as an NH₃ gas sensor, as explained based on the following points:

1. Structurally (XRD test): The composite with a 3% nanoparticle ratio exhibits the best performance, as the well-ordered Α-Fe₂O₃ provides stable conduction pathways within the amorphous PANI structure. This ratio promotes moderate crystal growth (19.99 nm), contributing to enhanced gas sensing performance. While the crystal growth was weak at 1% doping ratio (15.95 nm), the 5% nanoparticle composite shows more agglomeration and defects (19.95 nm), which likely block or cover active sites in the polymer, resulting in reduced gas sensing performance.
2. Chemically (FTIR): The results confirmed the successful synthesis of metal oxide NPs/PANI composites. The strong interfacial interaction between PANI and α-Α-Fe₂O₃ is evident from the high vibrational intensity of the Fe–O bond and shifts in PANI bonds such as C=N and C–N. The optimal interaction was observed at 3%, which demonstrated a balanced structural integration, favorable for enhancing charge transfer and improving NH₃ sensing performance.
3. Morphologically (SEM test): The image of SH3 shows the most uniformity compared to other doping ratios. In addition, the images confirmed the high surface interconnected paths that enhanced the charge transfer and provided numerous active sites. Compared to other doping ratios, which either provide few active sites with disconnected paths as in low doping ratio (1%) or show the particles agglomeration leading to blocking these sites and reducing and preventing charge transfer as in high doping ratio (3%).
4. Optically (Energy gap): The main reason for the best NH₃ sensing activity is the optimal combination of film thickness (230.1 nm), dopant ratio, and the band gap value 1.96 eV). At this ratio, the p-n heterojunction depletion layer occupies most of the film, leading to the strong NH₃-induced modulation of the interfacial barrier, and this did not happen either in low or high doping ratios. The effective response of this doping ratio can mainly be attributed to the balanced film structure rather than the lowest Eg alone.
5. Electrically (Hall effect test): 3% α-Α-Fe₂O₃ ratio show the highest conductivity (7.99 × 10^8- S/cm) compared to other doping ratios due to its well-dispersed nature within the PANI matrix, and strong interactions between these particles and the nitrogen atoms of PANI lead to the formation of new defect states, known as localized energy levels. These levels enhance conductivity and improve the gas sensing performance.
6. Ultimately, these combined advantages translate into the highest sensitivity, fastest response/recovery (48 s for both times response and recovery).

3.8 Discussion on the charge carrier type

The nanocomposites' electronic characterization shows the difference between surface sensing behavior and bulk transport related to sensor type, whether n-type or p-type. Whereas, the Hall effect results confirmed negative Hall coefficient values of all samples, including pure PANI (R < 0), indicating an n-type semiconductor. The NH₃ sensing behavior of the same samples showed a characteristic p-type response (the resistance increased with increasing exposure time of reducing gas). This phenomenon  can be justified  by the roles of Hematite as the core and with PANI as the shell as follows:

1. PANI as an n-type semiconductor: Although PANI is usually classified as p-type, in this work, it appeared as n-type according to Hall effect measurements. This can first be attributed to complex conduction mechanism. In a disordered polymer like PANI, the sign of the Hall coefficient does not match with the bulk charge carrier because of the hopping transport mechanism, besides the existence of bipolaronic and polaronic states. The second reason is that the Hall coefficient changes to reversed charge because of high levels of structural disorder or due to a specific arrangement of PANI chains, leading to n-type-like electronic transport characteristics in the bulk despite the p-type chemical nature of PANI in sensing performance. This behavior is matched with the results that were confirmed in Rajyalakshmi et al. [24]
2. Relating to nanocomposites, α-Fe₂O₃ acts as an n-type compound because of its oxygen vacancies. Whereas, PANI-based emeraldine salts are p-type. Upon contact, an n-p heterojunction is created at interference points. The metal oxide (α-Fe₂O₃) electrons diffuse into PANI layers, and at the same time, PANI holes diffuse into the metal oxide, forming a depletion at contact points. So, in such hybrid systems, α-Fe₂O₃ electrons move toward p-type PANI to equalize the Fermi level for the depletion region. While the bulk n-type core behaves as n-type in magnetic field measurements, the electrical percolating bath for NH₃ sensing is controlled by p-type surface and heterojunction barriers matched with the results of Bandgar et al. [25].

Relating to NH₃ sensing, it is noticed that samples act as p-type semiconductors with increasing resistance values, which is attributed to two mechanisms. The first one is PANI deprotonation, in which ammonia gas acts as a strong base to react with the protonated imine in the PANI backbone and consumes the positive holes, leading to a rise in resistance. The second reason for resistance increasing is because of depletion layer expansion when NH₃ reduces the PANI positive hole concentration, leading to the depletion layer expansion at  α-Fe₂O₃/PANI interference points. This expansion will narrow the conductive channels between successive chains and grains, leading to resistance increasings and this is correlated with the interpretation of Li et al. [26] in their work.