1 Introduction

Humidity is a crucial factor in several fields including ecology, industry, medicine, and indoor/outdoor environmental monitoring. It has a vital role in supporting living organisms as deviations from the optimal humidity range of 30 to 60% can severely affect the quality of life. For instance, high humidity levels can make it harder to breathe, causing asthma patients to cough and wheeze [1]. High levels of humidity can also create a favourable environment for dust mites and mould to thrive in homes, which can be detrimental to human health. Additionally, humidity plays a significant role in agriculture productivity. Because of this, humidity sensors are crucial for many indoor and outdoor applications, such as those in smart buildings, healthcare, agriculture, semiconductor manufacturing, and food packaging. A wide variety of humidity sensors have been developed via a variety of methods, including capacitive, resistive, optical fibre, thermal, and surface/bulk acoustic waves, to satisfy the demand. These sensors are constructed using a variety of nanomaterials, including metal oxides, 2D nanomaterials, and their composites, such as SnO2/RGO, graphene oxide/black phosphorus (GO/BP) [2], MoS2/Ag, and C3N4/Ag nanoparticles [3]. Indeed, the synthesis and fabrication processes of some of these material-based sensors can be complicated and time-consuming, making them less suitable for flexible and disposable applications. The invention of flexible and cost-effective sensors is the result of developments in nanomaterials and printing technologies [4]. For example, graphene-based materials are extensively studied due to their remarkable electrical, mechanical, and thermal properties, as well as their potential for large-scale production and printing. Furthermore, paper has become a popular substrate for humidity sensors because of its low cost, biodegradability, and ease of processing. Combining graphene with other conductive polymers and carbon black has also yielded promising results in terms of sensing performance and cost-effectiveness. Undoubtedly, the development of humidity sensors that can be seamlessly integrated onto flexible substrates using simple and low-cost fabrication techniques is crucial for emerging applications in wearable electronics and environmental monitoring [5,6,7]. A promising approach to achieving this goal is the direct printing of sensitive materials onto flexible substrates, which can yield sensors with high sensitivity, low cost, and disposability. Moreover, this approach can help to mitigate concerns related to electronic waste by reducing the amount of material used in sensor production and facilitating their easy disposal. In the end, the creation of adaptable and affordable humidity sensors has the potential to greatly raise the standard of living and encourage environmental sustainability across a range of sectors and applications [8].

Resistive sensors are also sensitive to a wide range of humidity levels, making them suitable for various applications. Graphene, as a sensing material, has several advantages, including high sensitivity, fast response, and good stability, making it a popular choice for humidity sensing [9]. The screen-printing process utilised in this study is a quick and economical procedure that allows the sensing material to be deposited on a variety of substrates, including flexible and disposable materials like paper. The sensor is lightweight and convenient to discard owing to the use of paper as a substrate, making it an excellent choice for environmental monitoring applications [9, 10]. Also, screen printing provides ease of operation and rapid processing; however, it is bound by the effects of viscous inks, high material waste, and the requirement for masks, to name a few. Inkjet printing, on the other hand, is well suited for mass fabrication that takes advantage of computer-aided design software for pattern modifications.

The G-P–C ink is an affordable substitute for existing metal-based ink formulations since it is both biocompatible and conductive. It is suitable for making disposable sensors because of this property. Screen printing is an ideal technology for mass-producing sensors, allowing for the control of design and scale and the use of different combinations of ink and functional materials with various substrates [11,12,13]. Paper is a suitable substrate for the presented sensor due to its low cost, disposability, recyclability, lightweight, and easy availability [14]. The objective of this study is to solve the difficulties related to developing a biocompatible, screen-printed, flexible, low-cost resistive humidity sensor on a paper substrate. The sensor can be used for a variety of tasks, including tracking soil humidity and space humidity. Examined and discussed is how different substrates and the number of printed layers affect the sensor’s ability to measure humidity.

2 Experimental

2.1 Fabrication of G-P–C black conductive ink

A solvent consisting of ethanol (10 mL), ethylene glycol (25 mL), glycerol (25 mL), and deionised water (40 mL) was prepared by mixing and stirring for 10 min. Carbon black was combined with graphene/polypyrrole binary conductive powder at several weight ratios (from 2 to 35 wt.%) to form G-P–C ternary conductive powder. Following that, the solvent was combined with the G-P-C black ternary conductive powder (600 mg) and CMC (450 mg), which were both sonicated at 950 W for 45 min before being centrifuged at 7000 rpm for 20 min. G-P-C black conductive ink was recognised as the liquid supernatant that resulted (as shown in Fig. 1b) [15].

Fig. 1
figure 1

a Graphene/polypyrrole powder and b graphene/polypyrrole/carbon black conductive ink

Full size image

2.2 Sensor fabrication using screen printing

The humidity sensors were created using graphene-polypyrrole-carbon (G-P-C) and the screen-printing technique on matt paper (240 gsm) substrates. The number of screen-printed layers was optimised for the paper substrate, resulting in three types of sensors: single-layer (1), double-layer (2), and triple-layer (3) [16]. Each layer was dried before the next one was applied, and the sensors were heat-treated at 60 °C for 1 h. The sensors were then connected to 2-wire resistance monitoring connections and a dielectric was placed on contact points according to wiring specifications.

3 Results and discussion

3.1 Material characterization

Figures 2 and 3 depict the process and resulting humidity sensor using a sensing layer composed of graphene-polypyrrole-carbon (G-P-C). The process for creating the nanocomposite ink is detailed in Fig. 4. The surface morphology of the sensing layer was examined using scanning electron microscopy (SEM), and Fig. 5 displays SEM pictures at two different scales (5 m and 500 nm). These photos show that the graphene sheets are decorated with polypyrrole and carbon particles and that a G-P-C network is randomly distributed throughout the sensing layer. Figure 6 shows a diffractogram of the printed layer with a peak at 26 matching the (002) plane, which is characteristic of carbon, as well as SEM images of the sensing layer at 100 m and 50 m (XRD pattern, Fig. 6) [17,18,19,20].

Fig. 2
figure 2

General schematic for fabrication of screen-printing electrode

Full size image
Fig. 3
figure 3

Screen-printed G-P-C humidity sensor

Full size image
Fig. 4
figure 4

Overall process for nanocomposite

Full size image
Fig. 5
figure 5

SEM images of cross-sections of conductive ink lines with 10 passes: a–f 2, 9, 16, 23, 30, and 37 wt.% ink development

Full size image
Fig. 6
figure 6

XRD pattern of G-P-C nanocomposite

Full size image

Hydrophilicity is a measure of a material’s ability to attract and absorb water molecules from its surroundings. In the case of humidity sensors, the sensing layer’s hydrophilicity is crucial because it allows the sensor to absorb moisture from the environment and change its electrical conductivity, which is the basis for the sensor’s operation [21]. If the sensing layer is too hydrophilic, it may become saturated with water and unable to absorb more moisture, resulting in slow recovery or response of the sensor. On the other hand, if the sensing layer is not hydrophilic enough, it may not be able to absorb enough moisture to produce a measurable change in electrical conductivity, resulting in a less sensitive sensor [22]. Therefore, it is important to optimise the hydrophilicity of the sensing layer to achieve the best possible sensor performance. A moderate level of hydrophilicity allows for rapid response and recovery of the sensor while still being able to absorb enough moisture to produce a measurable change in electrical conductivity [23].

The cross-sectional SEM image analysis shown in Fig. 5 can also be used to measure the thickness of the single, double, and triple G-C layers. For the single, double, and triple layers specifically, the layer thicknesses are approximately 10 m, 16 m, and 22 m, respectively [24, 25]. Table 1 displays the thickness of the ink layer after 10 passes, showing that the thickness increase per pass was higher than the results reported in the state of the art. This may be attributed to human errors in the manual writing process, as well as the high solid content of the G/P/CB conductive ink used in this study.

Table 1 Ink layer thickness of G/P/CB conductive ink
Full size table

3.2 Sensing mechanism

The resistance of the sensor rises as the relative humidity level rises. At lower %RH values, a physiosorbed layer of water molecules forms on the surface of the sensing material, which results in a change in resistance in the specified range. Figure 7 depicts a potential method to explain the resistance variations that have been found across various humidity levels [26].

Fig. 7
figure 7

Possible mechanism of change in resistance observed at different humidity levels

Full size image

Sensors with single (sensor 1), double (sensor 2), and triple (sensor 3) G-P-C layers printed on paper substrates were studied within a humidity range of around 25%RH to 95%RH in order to examine the effect of the number of printed layers on humidity sensing [27]. The configuration of the screen-printed G-P-C layers was blamed for the variations in resistance values between the sensors. All three sensors showed a rise in resistance with increasing relative humidity percentage (%RH), demonstrating sensitivity to changes in humidity levels.

The comparison of the sensors’ performance utilising the change in resistance (R) with respect to the change in %RH (%RH) is shown in Fig. 8. Throughout a humidity range of around 25%RH to 95%RH, the values of R/%RH for sensors 1, 2, and 3 are 12/%RH, 10/%RH, and 2.5/%RH, respectively. It can be shown that sensor 1, which is based on a single G-P-C layer, works better than sensors 2 and 3 within the measured humidity range and exhibits the largest resistance change [28]. The single G-P-C layer–based sensor 1 was found to be the best humidity sensor after comparing the performance of humidity sensing for all three sensors. Further, the percent response (% R) of the humidity sensor is determined at each intermediate humidity level. The % R-value is calculated as follows:

$$\frac{\Delta R}{\mathrm{RB}}\times 100$$
(1)

where ΔR = RH − RB; RH is resistance at a specific humidity level; and RB is the baseline resistance of the sensor analysed at 25%RH.

Fig. 8
figure 8

Comparison of sensor performance

Full size image

During the course of more than 4 weeks, the constructed sensor’s performance is assessed by routinely checking its stability. As shown in Fig. 9, the sensor’s response is initially measured at 12%RH and is then contrasted with its responses at 1 week (12.30%), 2 weeks (11%), and 3 weeks (11.50%). Based on the comparison analysis, it can be said that the sensor’s performance is still largely steady 3 weeks later, with only a slight drop in its sensing capacity. Additionally, by repeating the fabrication process, the reproducibility of the sensor is examined, and the responses of the replicated sensors (RS1, RS2, and RS3) are contrasted with those of the original sensor. The outcomes show that, with only small differences, the responses of the replicated sensors are consistent with those of the original sensor. This shows that the created sensor may be suitable for various measurements.

Fig. 9
figure 9

Stability analysis plot

Full size image

3.3 Bending analysis

Assessing the durability of the sensor involves examining its capacity to flex or bend, which is an essential characteristic. As shown in Fig. 10a–c, the fabricated humidity sensor is subjected to bending tests.

Fig. 10
figure 10

a–c Bending analysis on fabricated sensor

Full size image

The sensor’s operation is evaluated under distinct bending circumstances, with the sensor being bent at varying radii, as depicted in Fig. 10d–f. For the analysed bending radii, a small variation in baseline resistance is seen (less than 5% increase in resistance), which may be due to conductivity changes brought on by the G-P-C layer’s expansion. The sensor is tested under comparable humidity conditions both before and during bending at the defined radius in order to evaluate the impact of bending on its ability to detect humidity. The sensor’s humidity sensing performance is evaluated prior to and after undergoing 100 bending cycles, and it is determined to be almost identical. As a result, these findings indicate that the proposed humidity sensor is resilient enough to be used in wearable and flexible electronic devices.

4 Applications

4.1 Soil moisture monitoring

The development and growth of plants and crops are greatly influenced by the amount of moisture or humidity in the soil. Plant death or root diseases might be the result of insufficient or excessive soil moisture. The right amount of moisture is necessary for healthy plant development and increased crop output. Therefore, it is crucial to constantly check the humidity and soil moisture levels. The developed sensor is examined for measuring the moisture content of compost soil. The sensor is initially placed in a typical environment, and its resistance is measured. The sensor is then positioned on a sample of compost soil, and variations in its resistance are watched as depicted in Fig. 11. The changes in humidity were recorded against changes in resistance over a consecutive time period [29, 30].

Fig. 11
figure 11

Sensing response of the fabricated sensor

Full size image

5 Conclusion

A low-cost, flexible humidity sensor with high detection performance (12%RH) between 35 and 95%RH was presented in the study. The sensor was very adaptable, repeatable, stable, and reproducible, which made it suited for a variety of uses. It might be integrated to create a multipurpose sensory patch and be used to track environmental stress and temperature changes. Unfortunately, the substrate of paper restricts its use to dry areas. The identical electrode layout might be printed on disposable polymeric sheets to get around this restriction. The flexible sensor’s current high cost restricts the integration process as a whole. These issues might be solved by integrating the sensors with more energy-efficient technologies, including adding energy harvesters and energy storage systems to make room for autonomous sensory modules.