Highly responsive carbon dioxide sensing by graphene/Al₂O₃ quantum dots composites at low operable temperature

Abstract

Novel chemiresistive gas sensors based on graphene/Al₂O₃ quantum dots composites were fabricated and examined for carbon dioxide sensing. Composite samples with different wt% of graphene (20–80 wt%) mixing in constant 1 g Al₂O₃ were prepared and characterized by X-ray diffraction, transmission electron microscopy along with selected area electron diffraction, ultraviolet–visible spectroscopy, fluorescence spectroscopy and thermo gravimetric-differential thermal analysis. The experimental results showed that the graphene/Al₂O₃-based chemiresistor exhibited much higher sensing response and it enhanced linearly with addition of graphene. The gas sensing mechanism was discussed on the basis of defect chemistry through fluorescence measurements. 80 wt% graphene/Al₂O₃ composite exhibited good sensing response (10.84) at room temperature, low operating temperature (398 K), fast response time (14 s) and recovery time (22 s) along with good stability.

1 Introduction

In recent years, global warming and climate change has been received extensive attention. Ocean absorbing carbon dioxide (CO₂) from the atmosphere results in ocean acidification and subsequently cause to climate change [1]. Likewise, CO₂ is potent greenhouse gas released to the atmosphere by human activities through combustion of fossil fuels to generate electricity, transportation and industrialization. In addition, CO₂ leads to diseases like chronic asthma and bronchitis [2]. Therefore, development of low-cost CO₂ sensors for a continuous and effective monitoring is required [3].

Graphene has attracted much interest due to its outstanding electrical [4] and mechanical properties [5]. Graphene has excellent gas sensing properties due to its high surface to volume ratio [6]. Schedin et al. [7] have developed field-effect transistors to detect the absorption of a single gas molecule. Graphene doped with metal oxides can considerably modify their electronic properties and put forward more surprising results [8].

Metal oxides like SnO₂ [9], ZnO [10], TiO₂ [11] and In₂O₃ [12] are the most efficient CO₂ gas sensing materials. Normally, metal oxide gas sensors are operated at high temperature. This significantly increases power consumption with an increase in operation cost. Al₂O₃ is presently one of the most practical oxide ceramics, as it has been used in many fields of engineering such as heat-resistant materials, coatings, advanced ceramics and cutting materials [13, 14]. Konttinen et al. [15] have reported another interesting characteristic of Al₂O₃, which is moisture resistance. They have shown that Al₂O₃ based sensors are not affected by humidity strappingly. The Al₂O₃ is one of the most important ceramic materials with a range of applications such as chemical sensors [9] and fuel cells [16]. Fan et al. [17] have reported Al₂O₃/graphene nanocomposite and explore its electrical properties. Jiang et al. [18] fabricated and tested Al₂O₃/graphene nanocomposite sensors for ethanol sensing. Venkatesan et al. [19] have reported the stacked graphene-Al₂O₃ nanopore sensors for sensitive detection of DNA and DNA–protein complexes.

Recently, experimental investigations demonstrated that sensing properties are significantly affected by defects concentration on sensing surface [20, 21]. Thus, the analysis of sensing properties done through defects chemistry by fluorescence measurements and it is extremely suitable approach. Nemade et al. [22] have demonstrated CO₂ sensing by few layered graphene. Yoon et al. [23] have synthesized graphene by stamping method and tested CO₂ sensing properties. Similarly, Nemade et al. [24] have reported graphene/Y₂O₃ quantum dots composite based CO₂ gas sensor. In present investigation, we have found some interesting results over the pristine graphene based sensors.

The present work is planned to investigate CO₂ gas sensing characteristics of graphene/Al₂O₃ quantum dots (QDs) composites. There are few reports related to graphene/Al₂O₃ nanocomposite and it is for the first time, we have explored graphene/Al₂O₃ QDs composites as a CO₂ sensing material. The CO₂ sensing performance of materials is studied at room temperature as well as at different temperatures. The present work is devoted to the detection of low CO₂ concentration. Detection of lower CO₂ concentration is crucial, to monitor environmental pollutant released due to human activities. Some attractive accomplishments are reported; such as chemiresistor have high sensing response, low operating temperature, fast response and good stability.

2 Experimental details

Graphene was obtained by previously reported method [22]. Aluminium oxide (Al₂O₃) QDs was synthesized by chemical route using aluminium nitrate (Al(NO₃)₂) and hexamethylenetetramine (HMT) of an analytical grade. The 1 M Al(NO₃)₂ was added into 1 M HMT in aqueous medium. The solution was thoroughly mixed by magnetic stirrer for 2 h at room temperature. Subsequently, the product was kept for a centrifuge operating at 3,000 rpm for 30 min. This centrifuged precipitate was collected through cellulose nitrate filter paper. The filtrate was dried at room temperature for over night in vacuum chamber and then sintered at 773 K for 3 h.

The graphene/Al₂O₃ QDs composites were prepared by varying the wt% of graphene in constant 1 g Al₂O₃ QDs using acetone media. The solution was thoroughly mixed by magnetic stirrer for 30 min at room temperature. After this procedure, the solution was kept for over night for evaporation of acetone. In this manner, prepared composite was sintered at 373 K for 1 h for complete evaporation of acetone. It was considered as direct and efficient approach for preparation of graphene/metal oxide composites [25–27]. Four samples were prepared by altering the wt% of graphene from 20 to 80 wt%.

The structural purity of prepared materials was evaluated by X-ray diffraction (XRD) analysis with CuK_α radiation (λ = 1.5406 nm) (Rigaku Miniflex). The high resolution-transmission electron microscopy (HR-TEM) along with selected area electron diffraction (SAED) was performed to analyse the morphology (Philips Tecnai F-30107). Optical characterizations were done using ultraviolet–visible (UV–VIS) spectrophotometer (Perkin Elmer) and fluorescence Spectrophotometer (Hitachi, F-7000). The thermo gravimetric-differential thermal analysis (TG–DTA) was obtained in a Shimadzu DTG-60 h thermal analyser. The TG–DTA measurement was carried out from room temperature to 775 K in nitrogen atmosphere.

The synthesized samples were used as sensing layer. The chemiresistors of as-prepared samples were fabricated by screen-printing technique on glass substrate of size 25 mm × 25 mm and drying under controlled temperature rate up to 343 K. The silver electrodes were deposited on adjacent sides of the film for the measurement of the electrical resistance. CO₂ sensing properties of materials were studied in a homemade gas sensor assembly. Humidity and temperature within the chamber were precisely controlled. The sensing response was studied by using air as background gas having H₂O lower than 2 ppm. The known volume of the CO₂ was inserted into the gas chamber to maintain required concentration inside the chamber. Each reading of chemiresistor have acquired after period of 10 s. Resistance of the chemiresistor was measured by using a voltage divider method [28]. The sensing response of chemiresistor was defined as [27]:

$$ S = \frac{\Delta R}{{R_{a} }} = \frac{{\left| {R_{g} - \left. {R_{a} } \right|} \right.}}{{R_{a} }} $$

(1)

where R _a and R _g is the resistance of chemiresistor in air and gas, respectively.

3 Results and discussion

X-ray diffraction (XRD) patterns of graphene, Al₂O₃ QDs and (20–80 wt%) graphene/Al₂O₃ QDs composites are shown in Fig. 1. Diffraction peaks in the patterns Fig. 1(a), and 1(b) are in agreement with standard XRD peaks, which confirm the formation of graphene (JCPDS No. 01-0646) and Al₂O₃ (JCPDS 10-0173), respectively. We have not observed presence of any peaks due to impurities. Reflecting planes observed in graphene and Al₂O₃ QDs appear at the same positions in composites, curves shown in Fig. 1(c)–1(f), indicating good coordination between the components of the composites; i.e. graphene and Al₂O₃ QDs. The crystallite size obtained using Debye–Scherrer formula. Average crystallite size of Al₂O₃ QDs is found to be 5.5 nm and for composites, it is in the range 3.23–4.41 nm. The smallest crystallite size is found to be 3.23 nm for 80 wt% graphene/Al₂O₃ QDs composite.

Fig. 1

XRD patterns for (a) graphene, (b) Al₂O₃ QDs along with (c) 20 wt%, (d) 40 wt%, (e) 60 wt% and (f) 80 wt% of graphene/Al₂O₃ QDs composite

Figure 2(a) depicts TEM image of the synthesized Al₂O₃ QDs. The small amount of agglomeration is present in the sample. Average crystallite size estimated from the TEM analysis agrees with that obtained from XRD analysis. Diffraction spots shown in SAED image presented in Fig. 2(b), agreeing with the results obtained from XRD analysis. The several diffracting rings (012), (104), (110), (113) are identified in the diffraction patterns.

Fig. 2

(a) TEM and (b) SAED image for Al₂O₃ QDs

Figure 3(a) and 3(b) shows TEM and SAED images of 80 wt% graphene/Al₂O₃ composites, respectively. Broad rings in Fig. 3 are indicative of the good coordination between graphene and Al₂O₃ QDs. From magnified image shown in Fig. 3(c), it is observed that Al₂O₃ QDs are attached to surface and edges of graphene. Hybridised structure of Al₂O₃ QDs and graphene has been directly visualised in HR-TEM, shown in Fig. 3(d). Well-defined lattice fringes of the Al₂O₃ QDs are clearly observable.

Fig. 3

(a) TEM (b) SAED image (c) Magnified image of the red section in panel a. (d) HR-TEM image for hybridised structure of 80 wt% graphene/Al₂O₃ QDs composite

UV–VIS spectra of graphene and Al₂O₃ QDs are shown in Fig. 4. The spectrum of graphene shows intense absorption at 268 nm, which is attributed to the C–C bonds in graphene [29]. UV–VIS spectroscopy gives definite idea about the quantum confinment, which is an intrinsic characteristics of quantum dots [30]. UV–VIS spectrum of Al₂O₃ QDs shows an intense absorption at 355 nm. The average crystallite size of Al₂O₃ estimated from XRD and TEM analysis is 5.5 nm and absorption in UV region confirms that the synthesized particles are quantum dots [31].

Fig. 4

UV-VIS spectra for graphene and Al₂O₃ QDs

The emission spectra of 20–80 wt% graphene/Al₂O₃ composites, have been recorded under 254 nm irradiation in the range 315–700 nm, are shown in Fig. 5. Defect density has been directly estimated using the intensities ratio of the ultraviolet (I _UV ) to visible deep levels (I _DL ) [20]. I _UV, I _DL and I _UV /I _DL values are listed in Table 1. Oxygen vacancies are the most probable point defects, which increase the probability of adsorption of oxygen on sensing surface [21].

Fig. 5

Fluorescence spectra for 20–80 wt% graphene/Al₂O₃ QDs composites

Table 1 (I _UV /I _DL ) ratio for 20–80 wt% graphene/Al₂O₃ QDs composites

Figure 6 shows TG–DTA profile of 80 wt% graphen/Al₂O₃ QDs composites. The significant mass loss is observed below 373 K and DTA curve shows an endothermic peak at 346 K due to the evaporation of absorbed water. The total mass loss from room temperature to 473 K is about 10.20 %. The DTA curve show another peak at 541 K which corresponds to the evaporation of the constitution water in Al₂O₃ QDs. The exothermic peak appears at 705 K, which may be associated with the phase change of α-Al₂O₃ → θ-Al₂O₃ in composite [32]. TG–DTA study shows that the material is stable in temperature range from 350 to 500 K. Beyond 500 K, the mass starts to decrease smoothly, suggesting the pyrolysis of the carbon component of the composite.

Fig. 6

TG and DTA curves for 80 wt% of graphene/Al₂O₃ QDs composite

Figure 7 shows the comparative gas sensing response of chemiresistors measured with CO₂ and liquid petroleum gas (LPG) for 40 ppm at 398 K. All chemiresistors possess more than 50 % response for CO₂ and this value is 96 % for 80 wt% chemiresistor. This result shows that chemiresistors are more selective towards CO₂ as compared to LPG.

Fig. 7

Comparative gas sensing responses of chemiresistors towards CO₂ and LPG for 40 ppm at 398 K

The gas sensing response of chemiresistors as a function of CO₂ concentrations at room temperature (298 K) is shown in Fig. 8. Chemiresistors exhibit an increase in response as a function of CO₂ concentration up to 200 ppm. This indicates that chemiresistors demonstrate a good dependence on the gas concentrations. As discussed in introduction section, humidity does not hinder the performance of Al₂O₃ based gas sensors. Moreover, the response of (20–80 wt%) graphene/Al₂O₃ QDs composites chemiresistors is almost linear. From Fig. 8, it is clearly evident that composition with 80 wt% graphene/Al₂O₃ has the maximum sensing response. To evaluate reproducibility of the results, each measurement has been repeated 5 times and insignificant deviation is observed in the results. This may be due the smaller crystallite size, which provides a larger surface area for gas–solid interaction [33]. Figure 8 also clearly shows an enhancement in sensing response obtained with the increase in wt% of graphene. The XRD studies reveal that there is no significant difference between the particle sizes of composites. Therefore, it is necessary to concentrate on defect density that affects sensing response. The defect density on the sensing surface is determined by I _UV /I _DL ratio using fluorescence mesurments. From Table 1, it is directly observed that density of defects (I _UV /I _DL ) increases with an increase in wt% of graphene. Figure 9 shows variation of (I _UV /I _DL ) ratio and CO₂ sensing response as a function of wt% of graphene.

Fig. 8

Variation of response of chemiresistors with the concentration of CO₂ at room temperature

Fig. 9

Variation of I _UV /I _DL ratio and CO₂ sensing response with wt% of graphene

It is observed that excellent correlation exists between I _UV /I _DL ratio and gas sensing response. It also shows that defects density linearly increase with wt% of graphene. Density of defects may increase due to damage to graphene surface during addtion of graphene into Al₂O₃, or by interaction between Al₂O₃ and graphene which may create vacancies or dangling bonds. Incraesing the amount of graphene added into fixed amount of Al₂O₃, i.e. increasing number of graphene sheets becomes defective. Thus, defective sites of the graphene increase, where an adsorption energy increases. Hence, gas sensing response is influenced by defects concentration and is linelarly proportinal to the defect density. This result is consistent well with the results reported early by [21, 34]. The gas sensing response is related to defects through the oxygen vacancies, which can act as adsorption sites for atmospheric oxygen during gas sensing. More is the number of defects, more oxygen ions are adsorbed. This is one of the possible reasons for increase in sensing response with an increase in defects density.

Oxygen adsorbed on the surface of sensing material is mainly involved in sensing process of CO₂. The reaction for adsorbed oxygen ions are as follows [35].

$$ {\text{O}}_{2} ({\text{gas}}) \to {\text{O}}_{2} ({\text{ads}}) $$

(2)

$$ {\text{O}}_{2} ({\text{ads}}) + {\text{e}}^{ - } \to {\text{O}}_{2}^{ - } ( {\text{ads)}} $$

(3)

$$ {\text{O}}_{2}^{ - } ( {\text{ads}}) + {\text{e}}^{ - } \to 2{\text{O}}^{ - } {\text{ (ads}}) $$

(4)

CO₂ gas sensing mechanism is based on the reaction between the surface of chemiresistors and adsorbed oxygen ions. CO₂ has strong electron injecting tendency. Upon exposure to CO₂ (oxidising gas) environment, CO₂ molecules is adsorbed on bridging oxygen atoms with the formation of a surface carbonates and increase the resistance of chemiresistor indicating that sensing materials have n-type characteristics [36, 37].

The response of chemiresistors (graphene, Al₂O₃ QDs and 20–80 wt% graphene/Al₂O₃ QDs composites) as a function of operating temperature towards 50 ppm CO₂ is shown in Fig. 10. Response of chemiresistors to CO₂ varies with amount of wt% graphene and temperature. It is observed that 80 wt% graphene/Al₂O₃ composites exhibit the maximum value of sensing response at 398 K. This shows that the effect of graphene is significant. From Fig. 10, it is observed that all composites possess same operating temperature at relatively low temperature than pure graphene and Al₂O₃. Lower operating temperature would result in low power consumption. This is the main accomplishment of present work. Here, operating temperature 398 K fits in thermal stability domain of sensing material, which is comprehensively supported by TG–DTA. However, sensing response smoothly declines above 398 K. This may be due to desorption of oxygen ions from chemiresistor surface. When thermal vibration becomes sufficient, adsorbed oxygen gets desorbed [34]. Also at high temperatures, chemiresistors experience a continuous drift in resistance value [38].

Fig. 10

Sensing response of graphene, Al₂O₃ QDs and 20–80 wt% graphene/Al₂O₃ chemiresistors as a function of operating temperature

In order to check the stability of 80 wt% graphene/Al₂O₃ chemiresistor, its response to 100 ppm CO₂ has been measured for 30 days, at an interval of 5 days. Stability response result is shown in Fig. 11. Chemiresistor has almost constant sensing response indicating the excellent stability. Combining results obtained for all above-mentioned sensing properties, 80 wt% graphene/Al₂O₃ QDs composite is a potential sensing material to be used for efficient sensing of CO₂.

Fig. 11

Stability characteristics of 80 wt% graphene/Al₂O₃ chemiresistor to 100 ppm CO₂. Inset shows Transient response of 80 wt% graphene/Al₂O₃ chemiresistor to 100 ppm CO₂

Transient response of the 80 wt% graphene/Al₂O₃ chemiresistor at room temperature to 100 ppm concentration of CO₂ is displayed in the inset of Fig. 11. The time required for resistance to rise from its initial value to 90 % of highest value is known as response time of chemiresistor. The time required for decrease in resistance value to 90 % of highest value known is as recovery time. It is found that the fast response time for chemiresistor to 100 ppm CO₂ is 14 s, and recovery time is 22 s. The difference in response and recovery time may be due to faster CO₂ diffusion [38], while slow recovery due to formation of surface carbonates.

4 Conclusions

In summery, chemiresistors fabricated by using graphene/Al₂O₃ composite through screen-printing method are successfully demonstrated sensitive towards CO₂. Structural and morphological studies the composites indicate good coordination between graphene and Al₂O₃, which results in an increase in sensing response with the wt% of graphene. Also, the sensing response of chemiresistors shows good dependence on defects density on sensing surface. 80 wt% graphene/Al₂O₃ QDs composite chemiresistor can be used to practical sensor at room temperature and relatively low temperature (398 K).