1 Introduction

In recent year, the rapid development of science and technology has promoted the development of the country and society, but it has brought about environmental problems, especially the problem of gas pollution. Gas pollution is one of the important environmental pollution problems, producing serious threats to human health and security [1]. Mostly, formaldehyde (HCHO), carbon oxides (COx), nitrogen oxides (NOx), and other toxic and harmful substances cause air pollution [2,3,4,5]. Among them, HCHO is widespread generated from chemical factories, packing house, and explosives [6,7,8,9]. And HCHO, as a toxic gas, can harm people's health and even cause death. Therefore, in order to protect human health, highly selective and rapid detection of gas sensors is needed to detect HCHO at low concentrations in homes and industries.

Metal oxide semiconductors (MOSs) have the advantages of fast response speed and good selectivity. Therefore, the MOSs have attracted much attention in the detection of harmful gases. Among the MOSs, the band gap of copper oxide (CuO) is about 1.2 eV, which is a typical p-type metal oxide semiconductor [10]. And CuO has many excellent features, such as easy availability, low cost, easy production, and so on. Therefore, CuO is widely used in gas sensing. In particular, one-dimensional (1D) CuO nanoribbons have been widely studied due to its large specific surface area. For example, Gou et al. reported that CuO nanoribbons had good gas-sensing properties for HCHO [11]. However, CuO has the disadvantages of high operation temperature [12]. In especial, the operation temperature of pure CuO is 200–450 °C. Meanwhile, high operation temperature is not convenient for the application of pure CuO in daily life, which will make its work unstable and reduce its service life.

As a promising room-temperature gas-sensing material, reduced graphene oxide (rGO) has unique two-dimensional atomic thickness structure and excellent physical properties, such as large specific surface area and high electron mobility at room temperature, which meets the basic requirements of room-temperature gas detection for sufficient RT conductivity and high surface reactivity. For examples, Zeng et al. reported that rGO exhibited good gas sensibility to 20 ppm NO2 at the room temperature [13]. Next, the rGO can improve the stability of nanomaterials [14, 15]. In order to improve the gas-sensing performance, the rGO/MOSs nanocomposites were formed by metal oxide semiconductor composite on reduced graphene oxide nanosheets [16,17,18]. With further research, the gas-sensing property of the rGO/MOSs nanomaterials was improved at low temperature. For instance, Chao et al. found that the rGO/In2O3 nanocomposites had better gas-sensing performance for NO2 at 50 °C [19]. Also, Yin et al. report that the Sn3O4/rGO nanocomposites detect HCHO gas-sensing performance at 150 °C [20]. However, studies have scarcely been investigated on the determination of HCHO at low temperature by CuO nanoribbons/rGO.

Herein, we report the design and preparation of novel CuO nanoribbons/rGO nanocomposites, in which CuO nanoribbons were uniformly dispersed on the surface of rGO nanosheets via a facile one-step water-bath heating path. Meanwhile, the gas-sensing properties of the CuO nanoribbons/rGO nanocomposites were studied. Through the analysis of gas-sensing data, the CuO nanoribbons/rGO nanocomposites have good gas sensitivity to HCHO. The CuO nanoribbons/rGO nanocomposites have good selectivity and fast response recovery time to HCHO because of rich active sites, high surface area, and excellent carrier transport performance. Therefore, the CuO nanoribbons/rGO nanocomposites synthesized in this experiment have a wide application prospect in HCHO detection.

2 Experimental

Graphene oxide (GO) was synthesized from natural graphite powder by a modified Hummers method [21,22,23,24]. In this experiment, the water-bath heating path for the preparation of new CuO nanoribbons/rGO nanocomposites was designed, as shown in Fig. 1. In a typical process, 100 mL of 0.01 M CuCl2 solution was added into a 250 mL beaker, then 10 mL of 2.0 M NaOH solution was added to form a uniformly dispersed solution by magnetic stirring. Then, 10 mg of 1wt% GO was added to the above solution to disperse it in the solution. The 10 mL 0.6 M sodium citrate is added to reduce the Cu(OH)2 to CuO and the GO to rGO, respectively. The resulting solution was reacted in a 55 °C water-bath heater for 5.5 h to obtain the brown solution. Then, the as-obtained solution was cooled to room temperature and precipitated. The precipitates were washed three times with deionized water and anhydrous ethanol, respectively. Subsequently, the as-obtained precipitate was freeze dried overnight at − 50 °C. Finally, CuO nanoribbons/rGO nanocomposites were prepared by annealing the precursor in argon atmosphere at a heating rate of 3 °C min−1 at 400 °C for 2 h.

Fig. 1
figure 1

The schematic of growth mechanism of CuO/rGO nanocomposites

Full size image

The gas-sensing measurements were carried out using the WS-30A system (Winsen Technology Co. Ltd, Zhengzhou, China) via a static process, which was previously reported by our research team elsewhere [25]. The sensor can measure the response in air or target gas by monitoring the voltage on the reference resistor. The sensor response is defined as follows:

$$ {\text{Response}}~ = ~(R_{{{\text{gas}}}} - R_{{{\text{air}}}} )/R_{{{\text{air}}}} \times 100\%, $$
(1)

where Rair and Rgas are the resistance of sensor in air and in the presence of the test gas, respectively [26,27,28,29].

3 Results and discussions

3.1 Compositional and morphological analysis

The crystal structure and phase composition of the CuO nanoribbons/rGO nanocomposites were studied by X-ray diffraction. Figure 2 shows the XRD patterns of pure CuO nanoribbons and CuO nanoribbons/rGO nanocomposites. All the peaks of CuO nanoribbons are corresponding to the cubic phase of CuO (JCPDS card No. 65-2309). It is found that the diffraction peak of the as-obtained CuO nanoribbons is very sharp and there is no other impurity peak. Therefore, it is proved that the as-obtained CuO nanoribbons have good crystallinity and high crystal purity [30,31,32,33]. For the XRD pattern of CuO/rGO nanocomposites, the characteristic broad diffraction peak corresponding to rGO is not observed. The possible reason may be that the rGO content in the composite is relatively low, and the diffraction peaks of rGO are greatly suppressed by that of CuO because of their much lower intensity.

Fig. 2
figure 2

XRD patterns of the as-obtained CuO and CuO/rGO nanocomposites

Full size image

As shown in Fig. 3, the Raman spectra of CuO nanoribbons/rGO nanocomposites display four distinctive peaks at 1361.33 cm−1 in d-band, 1599.2 cm−1 in G-band, and 2656.01 cm−1 in 2D band [34]. The difference peak of 2D band was amplified from 2200 to 3300 cm−1, as shown in the insertion diagram, which indicated that the prepared rGO had good morphology [35]. D/G-band strength ratio is often used to evaluate the ordering degree of graphite in carbon materials. The D/G-band strength ratio of CuO nanoribbons/rGO nanocomposites is 0.95, indicating that graphene has SP2 hybrid structure [36]. Meanwhile, the peaks at 346 and 631 cm−1 appeared in the CuO/rGO nanocomposites are the characteristic modes of CuO [37]. The as-prepared sample is relatively darker due to absorption of light by the rGO, and hence, the Raman peaks associated to the stretching vibrations of short Cu–O bonds are weak.

Fig. 3
figure 3

Raman spectroscopy of CuO/rGO nanocomposites

Full size image

The morphology and elemental composition of CuO nanoribbons/rGO nanocomposites were characterized by field emission scanning electron microscopy (FESEM) and energy dispersive spectrometer (EDS). FESEM images of CuO nanoribbons/rGO nanocomposites are shown in Fig. 4a–d. It can be found that CuO samples have ribbon-like nanostructures, which are uniformly distributed on the rGO substrates. The CuO nanoribbons/rGO nanocomposites samples were further characterized using EDS patterns to determine the elemental composition, as shown in Fig. 4e–h. The EDS result reveals the co-existence of C, Cu, and O. Thus, based on the results given by XRD, Raman, SEM, and EDS analysis, it can be undoubtedly concluded that CuO/rGO nanocomposites, with the feature of CuO nanoribbons anchoring uniformly on the surface of rGO sheets.

Fig. 4
figure 4

SEM images of CuO/rGO (a–d), EDS images of CuO/rGO nanocomposites (e–h)

Full size image

The structure and morphology of CuO nanoribbons/rGO nanocomposites were further analyzed by transmission electron microscopy (TEM). From the TEM images shown in Fig. 5a, b, it can be confirmed that ultrathin CuO nanoribbons are dispersed on rGO sheet. Meanwhile, Fig. 5c clearly shows the width size of CuO nanoribbons and the folded structure of rGO. It can be seen from the TEM image that the CuO nanoribbons with a width of about 100–200 nm are in close contact with the surface of rGO. In Fig. 5d, the HRTEM diagram of lattice fringes of CuO nanoribbons/rGO nanocomposites is shown, and the crystal plane spacing is about 0.25 nm. They can be attributed to the (002) crystal face of cubic CuO phase [38,39,40]. Through the above TEM analysis, CuO nanoribbons have no obvious agglomeration owing to the presence of rGO substrate and have the porous structure, which will improve the gas-sensing performance.

Fig. 5
figure 5

a, b TEM image of the CuO/rGO nanocomposites; c, d high-resolution TEM image of CuO/rGO nanocomposites

Full size image

Subsequently, the surface areas and porous structures of CuO nanoribbons/rGO nanocomposites were characterized by N2 adsorption and desorption measurements. As shown in Fig. 6, the adsorption isotherm of the sample is recognized as type IV based on Brunauer–Demin–Deming–Teller classification, indicating the existence of abundant mesoporous structures [41]. The pore-size distribution curve shows that the corresponding pore sizes are centered at 6.4 nm. In addition, the large specific surface area of CuO nanoribbons/rGO nanocomposites is calculated to be 31.32 m2/g. Therefore, due to its large specific surface area and mesoporous structure, the adsorption capacity of CuO nanoribbons/rGO nanocomposites is greatly improved, which also helps enhance the gas-sensing response.

Fig. 6
figure 6

XPS spectra of CuO/rGO nanocomposites: a wide XPS spectrum of CuO/rGO nanocomposites; b C 1 s spectrum; c Cu 2p spectrum; d O 1 s spectrum

Full size image

The chemical state of the composite was further analyzed by X-ray photoelectron spectroscopy (XPS). The characteristic peaks of C, O, and Cu elements are clearly found in the CuO nanoribbons/rGO nanocomposites in Fig. 7a. Furthermore, Fig. 7b shows that two characteristic peaks with band energy at 962.9 eV and 944.3 eV correspond to Cu 2P3/2 and Cu 2p1/2 peaks, respectively [42]. In addition, Fig. 7c shows the C 1 s spectrum of three peaks, which can explain the C–C/C=C (284.9 eV) and C=O (285.5 eV) groups [43]. The dominant O 1 s peak at 538.8 eV belongs to the lattice oxygen of CuO nanoribbons (Fig. 7d). Due to the existence of two characteristic peaks of CuO in the sample, it indicates that CuO nanoribbons/rGO nanocomposites have been successfully synthesized [44].

Fig. 7
figure 7

N2 adsorption–desorption isotherm curve and pore-size distribution plots (inset) of CuO/rGO nanocomposites

Full size image

3.2 Gas-sensing properties of the CuO/rGO nanocomposites

In order to determine the optimal operating temperature of CuO nanoribbons/rGO nanocomposites, the sensor response to 100 ppb HCHO was firstly examined in the temperature range of 35 to 280 °C. From the results of Fig. 8a, response of CuO nanoribbons/rGO nanocomposites first increases and then decreases with the increase of working temperature. The gas response of the sensor reaches the maximum value of 49% at 80 °C. Due to the remarkable electrical conductivity of CuO nanoribbons/rGO nanocomposites, the optimal working temperature is reduced. As shown in Fig. 8b, the sensor shows a favorable response to HCHO even at a very low concentration of 1 ppb, and the response increases with the increase of HCHO concentration. When the working temperature is 80 °C, the corresponding responses of CuO nanoribbons/rGO nanocomposites sensor are 22%, 26%, 30%, 36%, and 49% at different HCHO concentrations of 1, 5, 10, 50, and 100 ppb, respectively. In addition, the enlarged response–recovery curve to 1 ppb HCHO is inseted in Fig. 8b. It can be seen that the response of the sensor to 1 ppb HCHO is very fast, and the response time and recovery time are 1 s, respectively. Figure 8c shows that the sensor was tested three times in 100 ppb HCHO at the optimal operating temperature of 80 °C. The results show that the response of the sensor has no obvious change with good reproducible, indicating that the sensor has good stability in practical application [45]. Figure 8d shows that the response of the sensor to 100 ppb VOCs (such as NO2, acetone, ethanol, toluene, methanol, and HCHO) was measured at the optimal operating temperature of 80 °C. It is revealed that the response of the sensor to NO2, acetone, ethanol, toluene, methanol, and HCHO is 2.5%, 2.3%, 1.5%, 3%, 3.5%, 3%, and 49%, respectively. The results show that the sensor has good selectivity for HCHO gas.

Fig. 8
figure 8

a Optimum working temperature of CuO/rGO nanocomposites to 100 ppb HCHO. b Response and recovery curves of sensors were exposed to 1–100 ppb of HCHO at 80 °C and the response and recovery time to 1 ppb of HCHO at 80 °C (inset). c Response and recovery curves of sensors to the three cycle of 100 ppb HCHO at 80 °C. d Response of sensors ground on CuO/rGO nanocomposites to 100 ppb various gases at 80 °C

Full size image

Table 1 shows the comparison of gas-sensing properties of CuO nanoribbons/rGO nanocomposites and other nanomaterials for formaldehyde. It can be clearly observed that the CuO nanoribbons/rGO nanocomposite sensor has excellent sensing properties compared with other nanomaterial sensors [46,47,48,49]. It can be inferred that the CuO nanoribbons/rGO nanocomposite sensor has good gas-sensing performance for acetone and has potential application prospects in many fields.

Table 1 Comparison of HCHO-sensing performances between CuO/rGO nanocomposites and previously reported sensing materials
Full size table

3.3 Gas-sensing mechanism of the CuO/rGO nanocomposites

CuO nanoribbons/rGO nanocomposites also show p-type characteristics, because both rGO and CuO materials have p-type semiconductor characteristics when detecting HCHO gas [50, 51]. The electron transfer process on the surface of the sensor is shown in Fig. 9, which can better show the gas-sensing mechanism of CuO nanoribbons/rGO nanocomposites for HCHO detection. Through the flow equation, it can be found that when the sensor is exposed to air, oxygen molecules are adsorbed on the sensor surface, and oxygen molecules attract free electrons from the conduction band of CuO nanoribbons/rGO nanocomposites, leading to form of O− and O2− species, thus, increasing the resistance of the sensor [52].

$$ {\text{O}}_{{2\left( {gas} \right)}} + {\text{e}}^{ - } ~ \to {\text{O}}_{{2\left( {ads} \right)}}^{ - }. $$
(2)

In addition, from the following equation, when the CuO nanoribbons/rGO nanocomposites sensor is exposed to HCHO-reducing gas, the o-species on the sensor surface reacts with HCHO molecules, releasing electrons on the sensor surface. The trapped electrons are released into the conduction band and coupled with holes in the valance band. Thus, the concentration of holes decreases and the hole accumulation layer becomes thinner, leading to an overall decrease in the conductivity [53].

$$ {\text{HCHO}}_{{\left( {gas} \right)}} \to ~{\text{HCHO}}_{{\left( {ads} \right)}} $$
(3)
$$ {\text{HCHO + O}}_{{2\left( {ads} \right)}}^{ - } \to {\text{CO}}_{2} + {\text{H}}_{2} {\text{O + e}}^{ - } $$
(4)
$$ {\text{h}}^{ \bullet } + {\text{e}}^{ - } \to {\text{null}} $$
(5)

First, the above reactions show that CuO and rGO constitute P–P heterojunction. A large number of adsorbed oxygen ions are transformed through the surface of CuO nanoribbons/rGO nanocomposites, indicating that the balance of carrier concentration is broken. Second, two-dimensional nano-rGO has fast carrier mobility. So the electrons on the surface of rGO can be rapidly released into the conduction band of CuO to balance the carrier concentration. Subsequently, the adsorption rate of oxygen species and HCHO molecules can be increased due to the fast electron migration, which makes the detection of HCHO concentration in low concentration gas even to reach 1 ppb. Above all, the agglomeration of CuO nanoribbons was avoided by the matrix effect of rGO. Therefore, the CuO nanoribbon/rGO nanocomposites have a specific surface area up to 31.32 m2g−1 and mesoporous structure. These properties make that CuO nanoribbons/rGO nanocomposites have high adsorption and diffusion capacity for HCHO molecules, which makes the response and recovery time of the sensor faster [54, 55]. In brief, the excellent gas-sensing properties of CuO nanoribbons/rGO nanocomposites are attributed to their fast electron mobility and mesoporous structure [56, 57].

Fig. 9
figure 9

The schematic of gas sensitivity mechanism of CuO/rGO nanocomposites

Full size image

4 Conclusions

In conclusion, we have developed a simple experimental scheme for the synthesis of CuO nanoribbons/rGO nanocomposites for the HCHO detection. Through a series of characterization methods, the diameter of CuO nanoribbons is about 100–200 nm with the main pore diameter that is about 6.4 nm and the BET surface is analyzed to be 31.32 m2/g. The results show that the sensor based on CuO nanoribbons/rGO nanocomposites has high response (about 22%), fast response, and recovery time (1 s and 1 s, respectively) and excellent selectivity for 1 ppb HCHO at 80 °C. The CuO nanoribbons/rGO nanocomposites with mesoporous structures provide more active sites and channels for the gas diffusion, promote the reaction between test gas molecules and oxygen ions, resulting in the improvement of the gas-sensing properties to HCHO gas.