Carbon Nanohorns and Their Nanohybrid / Nanocomposites as Sensing Layers for Chemoresistive RH Sensors – a Review

1. Introduction

Relative humidity (RH), a key environmental parameter that significantly impacts all life forms, is defined as the ratio of the amount of water vapor in the air relative to the amount of water vapor the air can hold at a specific temperature [1]. The RH level affects human ability to regulate body temperature, influences the growth of harmful microorganisms, and can impact the quality of indoor and outdoor environments. Exposure to high RH levels can lead to discomfort, trigger respiratory issues such as bronchoconstriction, and increase the risk of heat-related illnesses. On the contrary, low RH can promote dryness of the skin, a dry throat, irritated eyes, and constricted respiratory passages [2,3,4]. Monitoring RH is crucial in various commercial, industrial, and residential applications, some of which are depicted in Figure 1 [5,6,7,8].

Given the above, numerous principles and technologies have been developed over time for performing RH monitoring. The need for accurate and reliable RH measurements has driven innovation in sensor technology, resulting in improvements in range response, linearity, response time, drift, cost-effectiveness, accuracy, and other key metrics [9]. Among the types of sensors used so far in RH monitoring, such as capacitive [10], thermal conductivity [11], magnetoelastic [12], electrochemical [13], surface acoustic wave [14], bulk acoustic wave [15], and optical [16], the resistive sensors are an attractive option [17]. Simple design, small size, low cost, robustness, quick response times, and wide measurement ranges are just some of the advantages that resistive sensors offer in RH monitoring.

The nature of the sensing layer, a key element of a resistive RH sensor, plays a crucial role in manufacturing a resistive device with optimal performance. Consequently, a wide range of materials have been tested as sensing layer within the design of resistive RH sensors: conducting polymers, such as polypyrrole [18], dielectric polymers, such as polyimide [19], semiconductors, such as cadmium sulfide [20], ZnO [21], SnO2 [22], polyelectrolytes [23], perovskites [24] and so forth.

Carbonic materials were also widely reported as sensing films in the manufacture of RH sensors. Over the last few decades, these materials have gained popularity due to their outstanding properties: high chemical and thermal stability, versatile covalent functionalization to optimize the surface for proper interaction with water molecules, fast charge transfer, a large surface area, low cost, and facile and scalable synthesis [25]. Several carbon-based materials and their nanohybrids/nanocomposites used as sensing elements in the design of RH resistive sensors are listed in Table 1.

In recent years, carbon nanohorns (CNHs) and their nanocomposites or nanohybrids have also gained significant attention as potential materials for gas sensing applications due to their outstanding properties and potential for enhancing gas detection performance [50,51].

This review article focuses on the latest advancements and new perspectives of CNHs (pristine and functionalized), including nanocomposites and nanohybrids, as sensing materials for RH monitoring using resistive sensors. The review is organized into six main parts. In the first part, the main approaches to synthesizing CNHs are described. Significant attention will be devoted to the functionalization of these nanocarbonic structures to optimize their sensing properties towards water molecules. The second part primarily focuses on the most relevant physicochemical and electronic properties of CNHs, which make them valuable sensing materials for RH monitoring. The third part briefly presents the design of the RH resistive sensors, which employ CNHs-based materials as sensing films. The fourth part includes the synthesis and performance of several CNHs-based sensors for RH monitoring within the design of resistive sensors. Thus, pristine and functionalized CNHs, nanocomposites with several hydrophilic polymers, and nanohybrids with several semiconducting metal oxides are analyzed and compared in terms of sensitivity, response time, and recovery time. The fifth part of this review presents several sensing mechanisms involved in RH detection, as well as an analysis of how the properties of each component in the nanocomposite/nanohybrid and the mutual interaction between them influence the discussed RH sensing mechanisms. Finally, in the sixth part of this review, possible opportunities and future research directions are presented. Ultimately, this review seeks to address the question: Is the production of commercial sensors based on CNHs feasible?

2. Structure and Synthesis of Pristine CNHs and Their Derivatives Used in Resistive RH Monitoring

2.1. Structure of CNHs

CNHs (Figure 2) are conical carbon nanocages with cone angles of about 20°, constructed from a sp2 carbon sheet of average length from 40 to 50 nm and a diameter range of 2-5 nm [52]. In contrast to carbon nanotubes (CNTs), which are essentially rolled-up graphene sheets, CNHs have a more complex structure that includes a mix of pentagons, hexagons, and heptagons. This structural diversity contributes to their unique chemical properties and potential applications [53].

CNHs tend to aggregate into three distinct structural arrangements: "dahlia-like", "bud-like", and "seed-like". These structures differ in their overall morphology and the arrangement of the individual horn-shaped nanohorns within the aggregate: "dahlia-like resembles a dahlia flower with many petals, "bud-like" resembles a flower bud, and "seed-like" is a more compact arrangement. For many years, this structural characteristic was considered a significant drawback in the functionalization of individual CNHs. However, this limitation has recently been overcome by using a novel technique, recently reported, to separate these supramolecular architectures into individual CNHs [54].

2.2. Synthesis of Pristine CNHs

In the last decades, several approaches to synthesizing pristine CNHs have been developed. All these methods involve the injection of energy to vaporize and rearrange a graphite target, followed by rapid quenching, typically in an inert gas atmosphere. The morphology, size, and purity of CNHs are modulated through variations in various operational parameters, including temperature, current, voltage, and pressure, among others [55].

One of the most commonly used CNHs synthesis routes is that of arc discharge, which involves passing a high current between two graphite electrodes in atmospheric air. The purity of synthesized CNHs is higher than 90% [56]. The arc discharge approach comprises three steps: carbon vaporization, recondensation (where the vaporized carbon atoms recondense as they cool, forming CNHs and other carbon nanomaterials), and purification. The surrounding atmosphere (e.g., air, Ar, CO₂, or CO), gas flow rate, and arc current play a crucial role in determining the type, size, and morphology of the synthesized CNHs [57]. H. Wang et al. [58] reported a cost-effective synthesis of a mixture of 'dahlia-like' and 'bud-like" CNHs based on arc discharge between two graphite rods submerged in liquid nitrogen.

The synthesis of CNHs by CO₂ laser ablation of graphite in the absence of any catalyst was also performed in the last few decades [59]. Synthesis via Joule heating [60] and through inductively coupled plasma [61] represent two other feasible methods for the mass production of CNHs. It is essential to note that the synthesis of CNHs is conducted in the absence of a catalyst, which is a significant advantage compared to, for example, the synthesis of CNTs [54].

2.3. Synthesis of Functionalized CNHs for Resistive RH Monitoring

Due to their predominantly nonpolar carbon structure, pristine CNHs are generally hydrophobic. Therefore, to increase their affinity toward water molecules, their surface must be functionalized to make them more hydrophilic. Oxidation of CNHs with concentrated nitric acid, H₂O₂/hv, or H₂O₂/H₂SO₄ can introduce carboxyl groups onto the surface of the nanocarbonic materials, enhancing their hydrophilicity and dispersibility in polar solvents such as water, isopropanol, and ethanol [62]. Treatments like plasma exposure or oxidation can introduce polar functional groups (like carboxyl or hydroxyl groups) onto the CNHs surface.

The synthesis of oxidized CNHs (CNHox) is carried out using oxygen plasma treatment and water plasma treatment, as depicted in Figure 3. Both types of hydrophilization allow the functionalization of the CNHs by grafting carboxyl, hydroxyl, carbonyl, and epoxy groups. The degree of hydrophilization of the CNHs necessary to achieve superior RH sensing performance (high sensitivity, low response time, low hysteresis, etc.) can be modulated by adjusting the plasma power and exposure time [63].

Fluorinated carbon nanohorns (CNHs-F, having the structure presented in Figure 4) can also be synthesized and represent a viable option for resistive RH monitoring [64]. The synthesis of CNHs-F is performed by plasma treatment of CNHs in F₂ and N₂ (volume mixture 1:10) at a pressure of 0.5 bar in a nickel reactor at RT. The injection time is 4 minutes, and the exposure time ranges from 2 to 8 minutes.

Another functionalized carbon nanohorn–based material, oxyfluorinated carbon nanohorns (CNHox-F, with the structure inserted in Figure 5), can be synthesized in a two-step plasma procedure [65]. The synthesis of CNHox-F begins by treating CNHs in a volumetric mixture of Ar-O₂ (3:1) in a quartz tube at a pressure of 3 Torr and room temperature (RT). The injection time is 5 minutes, and the exposure time ranges from 2 to 4 minutes. The fluorination of CNHox is carried out by treating ox-CNHs in a F₂ and N₂ plasma (volumetric mixture 1:10) at a pressure of 0.5 bar in a nickel reactor at RT. The injection time is 4 minutes, and the exposure time ranges from 2 to 4 minutes.

The use of CNHox-F as a sensitive layer has several significant advantages. Firstly, the presence of oxygenated functions, generated by treating simple nanocarbon materials with Ar-O₂ plasma, ensures the degree of hydrophilicity necessary for interaction with water. Secondly, due to their increased electronegativity, fluorine atoms enhance the polarity of the nanocarbon material's surface, creating temporary dipoles that facilitate interaction with water molecules.

3. Properties of CNHs

Due to their unique nanostructure, CNHs exhibit outstanding properties as follows:

• High thermal conductivity - The thermal conductivity of CNHs is about 6250 W/m K, larger than that of other nanocarbonic materials, such as CNTs [66,67]
• High surface area - CNHs have a large specific surface area, which is a key feature for applications such as catalysis and adsorption [68]
• Excellent porosity - The partial oxidation of CNHs gives direct access to internal pores via the generation of nanowindows onto the skeleton of CNHs. Holes can be easily created in pristine CNHs by heat treatment under oxidative and/or acidic conditions [69]
• High adsorption capacity - [70]
• Thermal stability - CNHs generally exhibit good thermal stability, particularly in inert atmospheres. In air, the oxidation of SWNHs starts above 300°C and is completed at 720°C. CNHs can remain stable in a vacuum up to 1800° C [71]
• High purity - CNHs can be synthesized with high purity, often exceeding 95%, and without the need for metal catalysts [72]
• Chemical stability - CNHs generally exhibit good corrosion resistance, particularly when compared to some other nanocarbonic materials [73];
• Low toxicity - Several experiments conducted in recent years show that CNHs have low toxicity [74]
• Catalytic properties - CNHs can act both as catalysts and catalyst supports for metal nanoparticles [75]
• Good electrical conductivity - CNHs generally have lower electrical conductivity compared to CNTs; however, the conductivity of both materials can be influenced by several parameters, such as purity, structure, and type of synthesis. The electrical percolation threshold of carbon nanohorns and their derivatives in several hydrophilic polymers is a key parameter in the evaluation of resistive sensing capabilities of nanocomposites based on CNHs for different gases and RH [76,77]
• Facile covalent and noncovalent functionalization - [78]

We anticipate that all these features will recommend CNHs as an appropriate substitute for CNTs in the near future.

4. Structure of CNHs-Based Resistive RH Sensors

CNHs-based resistive RH sensors typically include a substrate, a sensing layer, and two metal electrodes. The most used sensing structure (presented in Figure 6) is manufactured on a Si substrate (470 µm thick), covered by a SiO2 layer (1 µm thick). The metal stripes of interdigitated transducers (IDT) electrodes typically consist of Cr (10 nm thickness) and Au (100 nm thickness). The width of the electrodes is about 200 µm, with a 6 mm separation between them. The digits of the electrodes have a width and spacing of 10 µm.

Alternatively, a flexible substrate made from polyimide (3.95 × 3.95 mm²), with gold interdigitated electrodes (Au IDTs), as depicted in Figure 7, can also be used for a CNHs-based resistive RH sensor [76].

At the same time, in a simple experimental approach, Selvam et al. [80] have demonstrated the use of cellulose sheet as a flexible substrate. The increase in resistance with the RH level was measured between two nickel electrodes, placed 5 mm apart.

5. RH Resistive Sensors Based on CNHS and Their Nanocomposites/Nanohybrids

The idea of using CNHs and their derivative as a sensing layer within the design of a resistive RH sensor was recently introduced [81]. Oxidized carbon nanohorns (CNHox) – carboxymethylcellulose and CNHox-agarose are the first two carbon nanohorn–based nanocomposites proposed to be used as a sensitive film for resistive monitoring of RH.

5.1 Oxidized CNHs as Sensing Layers in RH Resistive Sensors

Serban et al. used for the first time a sensitive layer based exclusively on a derivative of CNHs, namely CNHox [79,82]. An essential feature of these sensors was the use of a CNHs concentration higher than the percolation threshold in the polymeric matrix, thereby providing lower resistance values. The resistive response of the RH sensor used was investigated by applying a current between the two Cr and Au electrodes deposited on a Si/SiO2 substrate and measuring the resistance when varying the RH from 0% up to 90%, both in humid nitrogen environment (Figure 8) and in humid air (Figure 9). The resistance of the CNHox-based sensing film increased when the RH increased. The sensor response was compared to that of a commercial sensor (Sensirion® RH sensor).

The manufactured CNHox-based sensor exhibited good linearity in both humid air (R² = 0.9844, Figure 10) and humid nitrogen (R² = 0.9729, Figure 11).

The sensitivity was approximately 2 times lower in humid nitrogen compared to humid air (9.1 mΩ/RH unit compared to 21 mΩ/RH unit). The response time of the CNHox-based RH sensor in humid nitrogen was 8 seconds, while in humid air, it was 3 seconds.

5.2. Nanocomposite-Based CNHs as Sensing Layers in RH Resistive Sensors

5.2.1. Pristine CNHs–Hydroxyethylcellulose as Sensing Layer in RH Resistive Layers

By combining the appropriate electrical and mechanical properties of pristine CNHs and hydrophilic properties of hydroxyethylcellulose, Selvam et al [80] synthesized a nanocomposite with different loading concentrations of CNHs (5- 50 wt%). The resistance of the nanocomposite was shown to increase with the RH. The response time of the CNHs/cellulose-based sheet was 4 s, while the recovery time was 13 s.

5.2.2 CNHox–PVP as Sensing Layer in RH Resistive Layers

To further improve the sensitivity to water molecules, Serban et al. combined a hydrophilic type of CNH, namely CNHox, with a hydrophilic polymer, namely PVP (the structure is depicted in Figure 12), at 1/1 and 2/1 w/w ratios [83], significantly above the percolation threshold of CNHox in PVP. The synthesis of the nanocomposite was shown to be very simple. Initially, both the nanocarbonic material and the PVP were dispersed in deionized water. The CNHox-PVP-deionized water dispersion was further deposited by the drop-casting technique on the IDT structure previously deposited on the Si/SiO₂ substrate to generate the sensing film, in which CNHox and PVP were in a 1/2 w/w/ ratio. The sensing layer was then heated at 80^oC for one hour in a vacuum. The sensing capabilities of the manufactured RH detector employing the novel sensing film were explored in a humid nitrogen atmosphere for different RH values and compared with the response of a Sensirion commercial RH sensor (Figure 13). The sensor exhibits a quasi-similar response to that of a commercially available capacitive RH sensor. Experimental data reveal a linear relationship between R and RH for RH < 40%, and a second-order polynomial function variation for RH > 40% in a humid nitrogen atmosphere. The response time of the proposed sensor structure was in the 5.5–5.9 s range.

5.2.3. CNHox - poly (ethylene glycol)-blockpoly(propylene glycol)-block-poly (ethylene glycol) (PEG-PPG-PEG) as Sensing Layer in RH Resistive Layers

A matrix nanocomposite based on CNHox and PEG-PPG-PEG (the structure of which is depicted in Figure 14) was reported as a sensing layer in the resistive monitoring of RH [83]. In the synthesis process, CNHox and PEG-PPG-PEG (1/6 w/w ratio) were dispersed in deionized water, subjected to magnetic stirring for three hours at RT, and spin-coated on a Si/SiO₂ substrate. The RH detection experiments were conducted by applying a current between the two electrodes: Cr with a 10 nm thickness and Au with a 100 nm thickness. The electrode width was approximately 200 mm, with 6 mm of separation between them, and measuring the voltage difference when varying RH from 0% to 98% (a constant current of 0.1 A was passed through the sensing device). As presented in Figure 15, for RH < 60%, the voltage had a relatively low increase with RH, while for RH > 70%, the sensing device exhibited a stronger RH sensitivity. For the entire RH domain, the electrical resistance of the sensing film increases with RH.

5.2.4. GO-CNHox–PVP as Sensing Layer in RH Resistive Layers

A matrix nanocomposite based on GO (the structure is depicted in Figure 16), CNHox, and PVP at mass ratios of 1/1/1, 1/2/1, and 1/3/1 w/w/w was recently reported as the sensing layer within the design of RH resistive sensors [84,85]. PVP is a well-known binder, while GO disperses CNHox, increasing the specific surface area of ​​the RH-sensitive layer. The synthesis of the ternary nanocomposite was conducted in an ultrasonic bath using isopropanol. At the end of the synthesis process, annealing of the sensing film was achieved using a two–step procedure: a thermal treatment at 353 K and 2 mbar for 20 hours, followed by a thermal treatment at 383 K and 2 mbar for 90 hours.

Scanning electron micrographs (SEM) of the GO-CNHox-PVP RH sensing film deposited onto the Si/SiO₂ substrate reveal a homogeneous surface, as depicted in Figure 17. Multiple mutual interactions between CNHox, GO, and PVP (shown in Figure 18), such as hydrogen bonds, hydrophobic interactions, and π-π interactions, create a supramolecular aggregate that is the key element of the RH monitoring process.

The linearity of the RH response of the manufactured resistive sensors (sensor 111, sensor 121 and sensor 131 stand for GO-CNHox-PVP at the corresponding 1:1:1, 1:2:1 and 1:3:1 w/w/w mass ratios, respectively), tested in humid nitrogen (for the whole RH range) was shown to be excellent, as depicted in Figure 19.

The response and recovery times of sensors 111, 121, and 131 are shown in Figure 20. Response time was measured when increasing RH from 40% to 50%, while recovery time was measured when varying RH from 100% to 0% RH. Sensor 111 showed response times in the range of 40–90 seconds, while sensors 121 and 131 had response times between 50–100 seconds and 50–110 seconds, respectively.

The manufactured sensors 111, 121, and 131 had a shorter recovery time compared to the commercial RH sensor when the relative humidity values were decreased from 100% RH to 0% (62 vs 121, 73 vs 111, 73 vs 114).

Upon a simple examination of Table 2, we observe that a sensing film based on CNHox with a hydrophilic polymer, such as PVP, has superior performance in terms of sensitivity compared to the sensitive layer based on CNHox. Moreover, GO addition has a beneficial effect regarding the sensitivity toward water molecules (Table 2).

5.2.5. Pristine CNHs-PVP as Sensing Layer in RH Resistive Layers

To optimize the hydrophobic-hydrophilic ratio of CNHs-based sensing layers used in resistive RH sensors, Serban et al. developed a sensing layer based on CNH–PVP at a 9:1 w/w ratio [86,87]. Considering the low solubility of CNHs in water and their hydrophobicity, the nanocomposite was synthesized in dimethylformamide (DMF) using an ultrasonic bath. The CNHs-PVP-DMF dispersion was deposited on a polyimide substrate employing gold electrodes. The CNH-PVP sensing film was annealed for two hours at 100 °C under a vacuum.

The RH detection experimental measurements of the sensor using the CNHs-PVP nanocomposite as the sensing film were conducted by applying a current (0.5- 1 A) and measuring the voltage difference between the electrodes over the entire RH range (0% to 100%). The response of a CNHs–PVP–based manufactured RH sensor was compared to that of a commercial sensor. The resistance of the CNH–PVP–based sensing layer increased when varying RH from 0% to 70%. Once the 70% RH value was reached, the resistance began to decrease with increasing RH. For RH larger than 90%, the resistance started to rise again, as depicted in Figure 21. The combination of different sensing mechanisms (decreasing the number of holes in nanocarbonic materials, proton conduction, and swelling of PVP), as well as their relative prevalence, determines the type of response.

5.3. Organic-Inorganic Nanohybrids Comprising CNHs Used as Sensing Layers in RH Resistive Sensors

Due to the complementary and/or synergetic effects between inorganic and organic components, nanohybrid materials exhibit outstanding properties, such as good mechanical properties, biodegradability, tuned electrical properties, flexibility, enhanced surface area, porosity, and catalytic properties [88,89,90]. Therefore, nanohybrids have gained increased interest as a sensing layer in gas detection, with enhanced sensitivity, selectivity, and stability [91]. At the same time, recently, several CNHox-based nanohybrids were reported as sensing layers in resistive RH monitoring, as follows.

5.3.1. Organic-Inorganic CNHox/KCl/PVP Nanohybrids Used as Sensing Layers in RH Resistive Sensors

Several nanohybrids based on CNHox/KCl/PVP, synthesized at mass ratios of 7/1/2, 6.5/1.5/2, and 6/2/2 (w/w/w), were used as sensing films in the design of the resistive RH sensor [92,93,94]. The associated RH sensors were abbreviated as K1, K2, and K3, respectively. The sensing structure comprised a Si substrate, a SiO₂ layer, and an interdigitated transducer (IDT) based on Cr/Au electrodes. All RH detection measurements, presented in Figure 22, Figure 23 and Figure 24, were conducted by applying a current between the IDTs and measuring the voltage difference as the RH was varied from 0% to 100%. The resistance versus RH behavior of the manufactured sensors, based on CNHox/KCl/PVP as sensing layer, was compared to that of a commercial capacitive RH sensor, used as reference.

The manufactured CNHox/KCl/PVP–based sensors, which showed room temperature RH detection behavior comparable to that of the commercial capacitive RH sensor (Fig. 22-24). The devices are characterized by rapid response time (Figure 25), good sensitivity, and excellent linearity. For RH < 70%, the commercial sensor has a response time of 60 s ± 10 s, while for RH > 70%, the response time increases to approximately 90 s ± 10 s.

5.3.2. Organic-Inorganic CNHox/TiO2/PVP Nanohybrids Used as Sensing Layers in RH Resistive Sensors

Serban et al reported the use of CNHox/TiO₂/PVP nanohybrid as a sensing layer for RH resistive monitoring [95,96]. Three types of CNHox/TiO₂/PVP nanohybrids were synthesized, at 1/1/1 (corresponding to the manufactured sensor T1), 2/1/1 (corresponding to the manufactured sensor T2), and 3/1/1/ (corresponding to the manufactured sensor T3) mass ratios (w/w/w). The synthesis of the nanohybrids was conducted in ethanol using an ultrasonic bath. The mutual interaction between CNHox, titania, and PVP was confirmed using Raman spectroscopy. The Raman spectra for the CNHox/TiO₂/PVP at a 3:1:1 w/w/w mass ratio, deposited on the substrate, recorded at four points of the nanohybrid, are presented in Figure 26.

Three active Raman bands (D, G, and 2D) were measured at wavenumbers of 1316.9, 1589, and 2623.3 cm-1, which confirm the presence of CNHox [79,83]. Distinct anatase TiO2 bands, such as Eg1 mode at 150 cm-1, B1g at 398.7 cm-1, A1g at 513,7 cm-1, and Eg3 at 634.6 cm-1, were also recorded [97,98,99]. The corresponding peaks of PVP are likely hidden, most probably due to being masked by CNHox. The shift in Raman peak positions between individual TiO2 and CNHox and those of the same materials as components of the nanohybrid is a consequence of noncovalent interactions between them, such as hydrogen bonds and electrostatic interactions.

The resistance of the CNHox–TiO2–PVP–based sensing film increases when RH increases from 0% to 80% RH (Figure 27, Figure 28 and Figure 29). For RH larger than 80%, subtle differences are recorded. Thus, the resistance of T1 sensor moderately decreases with increasing RH, while the resistance of manufactured sensors T2 and T3 rises with RH. As shown in Figure 27, Figure 28 and Figure 29, the performance of the manufactured CNHox-TiO₂–PVP-based RH sensors is comparable to that of a commercial RH sensor used as a reference.

5.3.3. Organic-Inorganic CNHox/ZnO/PVP Nanohybrids Used as Sensing Layers in RH Resistive Sensors

Serban et al [100] deposited a ternary nanohybrid based on CNHox, ZnO, and PVP at a 5/2/1 mass ratio on a Si/SiO₂ substrate using the drop casting method. The morphology and composition of the sensing film were evaluated and confirmed through SEM and Raman spectroscopy. Experimental measurements showed that the resistance of the sensitive film increased with RH, varying from 0% to 100%. Increased RH sensitivity was recorded for RH > 60%. The response time of the manufactured CNHox-ZnO-PVP-based RH sensor was shown to be comparable to that of a commercially available capacitive RH sensor.

5.3.4. Organic-Inorganic CNHox/SnO2/ZnO/PVP Nanohybrids Used as Sensing Layers in RH Resistive Sensors

A thin film based on a quaternary nanohybrid comprising CNHox/SnO₂/ZnO/PVP was tested as a sensing layer in the resistive monitoring of RH [101]. Two CNHox/SnO₂/ZnO/PVP–based sensing layers were synthesized and deposited, at 1.5/1/1/1 (abbreviated as "sensor 1.5") and 3/1/1/1 (abbreviated as "sensor 3") mass ratio. The RH sensing device consists of a Si/SiO₂ substrate and interdigitated transducers (IDT) electrodes ( Cr/Au). As depicted in Figure 30, for both CNHox/SnO₂/ZnO/PVP nanohybrid-based sensing layers, the resistance increases when RH increases from 0% to 100%. Their experimental performance was compared to that of a commercial RH sensor, used as a reference.

The response time of the CNHox/SnO2/ZnO/PVP–based manufactured RH sensors ranged from 35 to 100 seconds for both devices, as presented in Figure 31. The highest values of response time were recorded at RH > 70%, most likely due to a decrease in the number of active sites.

Figure 32 shows the recovery pattern for both quaternary nanohybrid-based RH sensing layers measured when the RH from the testing box dropped from 100% to 0%. The calculated recovery times varied from 65 s to 100 s, values similar to those exhibited by the commercial sensor, which was employed as a reference.

5.3.5. Organic-Inorganic CNHox/GO/SnO2/PVP Nanohybrid Used as Sensing Layers in RH Resistive Sensors

Thin films based on a quaternary nanohybrid based on CNHox/GO/SnO2/PVP were also demonstrated as sensing layers in RH resistive sensors [102]. Two CNHox/GO/SnO2/PVP-based sensing layers were synthesized, at 1/1/1/1 (abbreviated as "sensor 1") and 0.75/0.75/1/1 (abbreviated as "sensor 0.75") mass ratios. The RH sensing device consisted of a Si/SiO2 substrate and interdigitated, Cr-Au transducers (IDT) electrodes. The composition and the morphology of the CNHox/GO/SnO2/PVP-based sensing films were explored and confirmed through X-ray diffraction (XRD), Scanning Electron Microscopy (SEM), and RAMAN spectroscopy. As depicted in Figure 33, for both CNHox/GO/SnO2/PVP nanohybrid-based RH sensing layers employed, the resistance increased when varying RH from 0% to 100%. A notable characteristic of these two manufactured CNHox/GO/SnO2/PVP–based sensors is their low power consumption, which is below 2 mW. Their experimental performance was compared to that of a commercial capacitive RH sensor, used as a reference.

The linearity of the manufactured CNHox/GO/SnO2/PVP–based sensors was excellent, as demonstrated by the transfer function shown in Figure 34.

6. Sensing Mechanisms for RH Resistive Monitoring Using CNHs and Their Nanocomposites/Nanohybrids

Any hypothesis regarding the RH resistive monitoring sensing mechanisms using CNHs (pristine and their derivatives) or nanocomposites/nanohybrids based on these nanocarbonic materials begins with the fact that CNHs are p-type semiconducting materials [103,104]. At the same time, it is to be expected that chemisorbed water molecules on the CNHs surface operate as electron donors [105]. As the electron density increases, the positive charge concentration in CNHs decreases and, thus, the p-type CNHs-based films become more resistive. This scenario accounts for most of the reported experimental results concerning resistive RH sensors employing CNHs-based sensing layers [76,77,83,84,85,86,87,92,93,94,95,96].

The interaction between water and CNHs (as p-type semiconducting materials) can also be discussed from the perspective of the Hard-Soft Acid-Base (HSAB) theory. This theory, proposed by Ralph Pearson in 1963, operates with Lewis acids and bases: a molecule capable of donating electron pairs acts as a base, while a molecule that acts as an electron acceptor is classified as an acid. Lewis acids and bases can be classified into three types: hard, soft, and borderline [106,107,108,109]. According to the HSAB concept, strong bases have a greater affinity for interacting with strong acids, while soft bases preferentially interact with strong acids. In contrast, borderline bases tend to interact with borderline acids. Given the above definitions and rules, due to the electron pairs in the oxygen atom, H₂O is a typical example of a hard base and, consequently, has an affinity to interact with the positive charge carriers (hard acids) in the CNHs. In recent years, the HSAB theory has become a valuable tool for selecting sensitive materials for gas sensing, as well as for explaining specific reaction mechanisms [110,111,112].

At the same time, hydrogen bonds, as well as the electron-withdrawing effect of the carboxyl group (present in CNHox), have an impact on the hole concentration in the nanocarbonic films and can modulate the RH sensitivity, ultimately affecting the RH sensor response [113].

The second plausible RH sensing mechanism considered for the interpretation of the RH sensing experimental results presented in the previous chapter is the swelling of the hydrophilic polymer employed in the nanohybrids sensing layer in contact with water [76,77,84,85,86,87,92,93,94,95,96]. PVP and PEG-PPG-PEG are dielectric polymers with hydrophilic properties, which swell upon interaction with moisture. The swelling of polymers leads to the displacement of the nanocarbonic particles, increasing the distance between the CNHs and decreasing the electrically percolating pathways, as depicted in Figure 35. Consequently, the sensing film resistance increases upon exposure to a higher level of RH because more water molecules move into the bulk of the nanocarbonic film.

PVP swelling is more pronounced at higher levels of RH. The swelling of PEG-PPG-PEG is somewhat different than that of PVP and requires a supplementary explanation. PEG-PPG-PEG is less hydrophilic than PVP, and more water molecules are needed to initiate swelling. For RH < 40%, the resistance of PEG–PPG–PEG–based films exhibits a relatively low increase with RH, while for RH > 65% the sensing layer resistance increases sharply with RH (switch-type behavior).

A third plausible RH sensing mechanism that needs to be discussed is based on the fact that, given the type of employed analyte (water), protonic conduction must be considered as a potential contributor to RH sensing. This type of sensing mechanism refers to the dissociation of water (a major contributor) and/or the ionization of carboxylic acids (from CNHox). The adsorbed water molecules on the surface of CNHox may dissociate into H⁺ and OH⁻. The protons generated by water dissociation and the ionization of carboxylic groups decrease the overall electrical resistance of the sensing layer [114]. However, since the experimental data presented in the previous section show an increase in resistance with RH, one can conclude that the contribution of this third RH sensing mechanism, based on proton tunneling, is relatively small compared to the impact of the first two sensing mechanisms discussed above.

Beyond the three RH sensing mechanisms, the individual contributions of other components in the CNHs-based sensing films, as well as the mutual interactions between them, could also play an essential role in the overall RH sensing mechanism.

7. Why Are CNHs Used Less Frequently than CNTs and Graphene Derivatives for Resistive RH Sensing? Possible Opportunities and Future Research Directions

CNHs with hydrophilic properties (CNHox or holey CHNs) seem to be viable solutions as a sensing layer within resistive RH sensors design due to specific outstanding properties:

• The absence of metallic particles as impurities- Synthesis of the CNHs is conducted in the absence of metal catalyst [54].
• High Surface Area- CNHox possesses a large specific surface area (1,300–1,400 m²/g BET), providing more sites for the adsorption of water, which increases its RH sensitivity [79].
• Tunable Surface Chemistry– The versatile hydrophilization of CNHs (through oxidation in solution or plasma treatment) allows for the optimization of the response of the sensor to different RH levels [54].
• Good Electrical Conductivity - CNHox retains good electrical conductivity even after hydrophilization, which is a key feature for resistive RH sensors. The interaction of water molecules with the CNHox surface can modify the electrical resistance of the sensing film, allowing for accurate RH monitoring [63].
• RT Operation- CNHox-based sensors can operate at RT, which is an advantage for manufacturing ultralow-power resistive sensors [93,94,95].
• Excellent Linearity and Stability - CNHox-based RH sensors have demonstrated excellent linearity in their response across a wide range of RH levels, and they have also shown good stability over time [83].
• Rapid Response and Recovery Times - The large surface area and tunable surface chemistry of CNHox contribute to obtaining fast response and recovery time [84].
• Compatibility with other materials- CNHox can be easily incorporated into nanocomposites or nanohybrids with different materials, like polymers (e.g. PVP, PEG-PPG-PEG), carbonic materials (GO), metal oxides (e.g., TiO₂, ZnO, SnO₂), further enhancing their sensing properties and enabling the development of RH sensors [51]
• Potential for Flexible and Wearable Sensors- The ability to create thin films and dispersions makes CNHox suitable for integration into flexible and wearable sensors, expanding the possible applications of RH sensors [54].

Despite the above properties, CNHs are used less than other nanocarbonic materials, such as CNTs, GO, and reduced GO, in RH monitoring. The statement is valid not only for RH sensors but also for gas sensors in general. The question is: Why are CNHs not as popular as other carbon allotropes?

A possible answer to that question is that, during their synthesis, CNHs tend to aggregate into spherical clusters, making it difficult to disperse them individually and create a homogeneous sensing layer. For many years, this was the major limitation in the chemistry of CNHs. Recently, this drawback was overcome by employing a new technique for separating the clusters into distinct nanohorns [54]. Another possible answer is that the low symmetry of CNHs reduces the number of predictive simulations. For these reasons, CNHs are less understood and used than other nanocarbonic materials [54].

Thus, several refined computational techniques, such as molecular dynamics (MD) and Monte Carlo (MC) simulations, and density functional theory (DFT) calculations, could offer more opportunities related to achieving an optimal level of hydrophilicity or proper functionalization.

We believe that carbon nanohorns hold significant future potential and will be utilized in an increasing number of applications, particularly in gas sensing, biomedicine, and energy storage.

8. Conclusions

The paper introduces and analyzes recent progress on resistive RT RH sensors based on CNHs, operating above the percolation threshold in the dielectric matrix. Several methods to synthesize CNHs and strategies for their hydrophilization, as well as the most important physical, chemical, and electronic properties of CNHs, were discussed in the first two sections of this review. The design of RH resistive sensors using CNHs-based materials as sensing films, as well as the synthesis and performance of several CNHs-based sensing materials in resistive RH monitoring, within the context of resistive sensor design, are presented in the following two parts of this review. Pristine and functionalized CNHs, nanocomposites including CHNs and different hydrophilic polymers (PVP, PEG-PPG-PEG) or other nanocarbonic materials (GO), nanohybrids comprising CNHs and several metal oxide semiconductors (SnO₂, TiO₂, ZnO) or hygroscopic inorganic salt (KCl) were presented and compared in terms of linearity, sensitivity, response time, and recovery time.

The fifth part of the review presented several sensing mechanisms suitable for explaining RH detection when using CNHs-based sensing layers in resistive structures. In the sixth part of the review, the authors listed the key properties that recommend CNHs as sensing layers in resistive RH detection. At the same time, the paper discusses why, for the time being, compared to graphene, GO, reduced GO, SWCNTs, and MWCNTs, CNHs are less frequently used in resistive RH detection.

Given the above rationale, one can conclude that CNHs hold huge future potential and will eventually be used in an increasing number of applications, especially in gas sensing, biomedicine (drug delivery), and energy storage (supercapacitors, batteries).