Highly Rapid and Sensitive Formaldehyde Detection at Room Temperature Using a ZIF-8/MWCNT Nanocomposite.

Formaldehyde is a volatile organic compound (VOC) with extensive applications, volatility, and toxicity, which have made it an important risk to human health even at low concentrations. Therefore, rapid detection of formaldehyde vapors in the environment is a necessity. Herein, we introduce a resistive gas sensor based on zeolitic imidazolate framework-8/multiwalled carbon nanotube (ZIF-8/MWCNT) for detection of formaldehyde vapors at room temperature. In this sensor, a low amount of MWCNTs was used in order to improve the electrical conductivity of the porous nanoparticles of ZIF-8. The sensor was fabricated by deposition of a thin layer of the nanocomposite onto interdigitated electrodes, and its sensing ability was investigated on exposure to formaldehyde vapors. The obtained sensor showed sensitive and fast responses to different concentrations of formaldehyde, and the sensor response to formaldehyde was higher than toward some other VOCs, including methanol, ethanol, acetone, and acetonitrile. Furthermore, because of the hydrophobic nature of ZIF-8, the effect of relative humidity on the gas-sensing performance was insignificant, which proves that this sensor is suitable for use under humid conditions.

Introduction

Formaldehyde or n class="Chemical">methanal (CH2O) is a colorless, strong-smelling volatile organic compound (VOC), which is an important precursor for many chemicals. Formaldehyde is one of the major causes of sick building syndrome, a situation in which occupants of a building may encounter health problems including eye irritation, cough, headache, nausea, and respiratory issues. These problems are mostly created by the sources inside the building, including building materials, some wooden materials, adhesives, resins, insulation materials, and carpeting and cleaning agents.[1,2] Outdoor sources of formaldehyde include exhaust gases of motor vehicles, flue gases from heating facilities, and incinerators. There are also some natural sources of formaldehyde emission, such as photochemical oxidation and incomplete combustion of hydrocarbons.[3] In addition, formaldehyde is known as a human carcinogen on long exposures, and because of its extensive applications, volatility, and toxicity, it is an important risk to human health. Therefore, quick detection of formaldehyde is very important and necessitates the use of gas sensors.

Gas sensors are widely used in the industries for detection of toxic or explosive gases, air quality monitoring, medical diagnostics, and food quality control.[4] Among different gas sensors, those that work at low temperatures are more desirable because they consume lower energy, have a simpler system, and so are not expensive. In addition, in the case of detection of explosive gases, high operating temperature sensors may bring the risk of ignition or explosion. Moreover working at high temperatures may reduce the lifetime of sensors.[5]

Resistive sensors are one of the most desirable sensors, mostly because of a noncomplicated and low-cost system that can be assembled in a small and portable setup.[5] In these sensors, gas adsorption on an adsorbent material causes changes in the electrical resistance of the material. For this purpose, the adsorbent needs to be a semiconducting material for being able to make electrical signals.[6]

One of the widely used sensing materials in resistive gas sensors are metal-oxide sensors which are known for their high sensitivity; however, they have low selectivity and low surface area, and they work at high temperatures because of their purely inorganic structure.[4,7] Therefore, porous sensing materials with more surface area and lower operating temperature are more desirable; this includes n class="Chemical">metal organic frameworks (MOFs). MOFs are constructed from metal ions or clusters coordinated to organic ligands and possess remarkable properties such as a high specific surface area, a hybrid organic–inorganic structure, and tunable porosities,[8,9] which make them ideal for gas sensing.

Zeolitic imidazolate framework-8 (n class="Chemical">ZIF-8) is a kind of MOF in which divalent zinc cations are linked through 2-methylimidazole (Hmim) linkers and is reported to show high crystallinity, nanoporous structure, large surface area, high chemical and thermal stabilities, and hydrophobicity.[10,11] ZIF-8 is considered as a high-performance sensing material in many types of gas sensors, such as optical sensors,[12−15] quartz crystal microbalance sensors,[16] microcantilever devices,[17] and other sensors for detection of different gases and VOCs. However, till date there are few reports of ZIF-8 as a sensing material in resistive gas sensors because of the poor electrical conductivity of this material. In most of the reports, ZIF-8 was used in the ZnO@ZIF-8 core–shell structure in which ZIF-8 improved the selectivity of the semiconducting ZnO gas sensor; however, these sensors work at high temperatures. For example, Tian et al. used this structure to detect formaldehyde at 300 °C,[18] and Drobek et al. did the same work for sensing hydrogen with improved selectivity.[19] Also, Matatagui et al. have used ZIF-8 in a composite with ZIF-67 in a resistive sensor to improve the response of ZIF-67 sensor to hydrogen and toluene at the temperature of 180 °C.[20]

Furthermore, in the case of formaldehyde, some works have been done which report detection of different concentrations of n class="Chemical">formaldehyde by resistive sensors. For example, Jia et al. fabricated a formaldehyde sensor by ZnSnO3, which could detect formaldehyde at a concentration range of 5–100 ppm; however, the working temperature for this sensor was higher than 200 °C.[21] In addition, Chimowa et al. tested encapsulated zinc, zinc oxide, and iodine within double-walled carbon nanotubes (CNTs) for detection of various mass concentrations of formaldehyde ranging from 1.8 to 4%.[22]

In this research, we selected ZIF-8 as an acceptable sensing material for sensing n class="Chemical">formaldehyde based on the literature. However, it should be used in a composite form with some conductive or semiconductive material. For this purpose, in the present work, ZIF-8 was combined with functionalized multiwalled CNTs (MWCNTs) to compensate for its low conductivity and to be able to work as a sensing material for a resistive gas sensor at room temperature.

Materials and Methods

Materials

All of the chemicals including Hmim (n class="Chemical">C4H6N2), zinc nitrate hexahydrate [Zn(NO3)2·6H2O], methanol (99.9%), formaldehyde (CH2O, 37% aqueous solution), ethanol (99.9%), acetone (99.9%), and acetonitrile (99.9%) were purchased from Merck Company and used without further purifications. MWCNT (8–15 nm outer diameter, 50 μm length and purity > 95%) was purchased from Neutrino Company (Iran). A custom-made polymer substrate with an interdigitated electrode (IDE) was used as the substrate for fabrication of the sensing layer and consisted of 38 silver fingers with 150 μm distance between each pair. Ag paste was purchased from Hezareh 3 Company (Iran) and used to connect the wires to sensor electrodes.

Characterization Instruments

X-ray diffraction (XRD) patterns of samples were collected by a Bruker D8 Advance X-ray diffractometer with Cu Kα radiation (λ = 0.15418 nm) in the 2θ range of 5–80°. The morphology was investigated by scanning electron microscopy (SEM) images which were taken by a TESCAN VEGA3 microscope. Fourier transform infrared (FTIR) spectra were collected using a PerkinElmer Spectrum RXI FTIR spectrometer to investigate chemical bonds. N2 adsorption/desorption isotherms of samples were measured by a Micromeritics n class="Gene">ASAP 2020 instrument at 77 K. The specific surface area of samples was calculated by the Brunauer–Emmett–Teller (BET) and Langmuir methods, and the t-plot pore volume was calculated for evaluating the micropore volume.

Synthesis of ZIF-8 Nanoparticles

ZIF-8 was synthesized according to a reported procedure with some modifications.[23] n class="Chemical">Zinc nitrate hexahydrate (0.5 g) was dissolved in 5 mL of deionized water and 20 mL of methanol, and 1.5 g of Hmim was dissolved in 25 mL of methanol. The solutions were mixed together rapidly, and the obtained milky suspension was stirred at room temperature for 16 h, until the suspension got white color. The final solution was centrifuged to separate the precipitated ZIF-8 solids, washed several times with deionized water and methanol, and kept in methanol for 1 day to bring unreacted precursors out. Finally, the precipitates were centrifuged again and dried at 60 °C overnight to yield white fine powders of ZIF-8.

Carboxyl Functionalization of MWCNTs

Commercial CNTs are supplied in the form of heavily entangled bundles, and their dispersion in the solvent is not satisfying. Furthermore, the interactions between CNTs and the composite matrix are poor;[24] so CNTs need to be functionalized before compositing. For this purpose, these particles were carboxyl-functionalized for better distribution in the composite. MWCNTs (40 mg) were added to 100 mL of n class="Chemical">H2SO4/HNO3 (1:3 volume ratio) solution and were stirred at 120 °C under reflux for 12 h. The obtained MWCNTs were washed with water several times and dried at 60 °C.[25]

Preparation of ZIF-8/MWCNT Nanocomposite

For preparing the ZIF-8/n class="Chemical">MWCNT nanocomposite, “ex situ synthetic approach” was selected which is a typical method for the synthesis of MOF/carbon composites.[26] According to this method, MWCNTs were directly mixed with presynthesized ZIF-8 and ethanol as the solvent. Different combinations of ZIF-8 and MWCNTs were tested, and finally, the suspension with 2.5% wt of MWCNTs was found to be the best one. For this purpose, 1 mg of MWCNTs and 39 mg of ZIF-8 were separately dispersed in 10 mL of ethanol in an ultrasonic bath for 30 min and then mixed together and stirred for 24 h. The final suspension was sonicated and heated slightly to lose the solvent without missing the homogeneity, until the solution volume reached 2 mL.

Preparation of Sensing Layer

The prepared composite was deposited on top of the IDEs with the drop-casting method. For making a uniform film, the homogeneity of the prepared composite suspension is important, so it should be sonicated well before deposition. After drop-casting onto the IDE substrate, it was kept in the oven for 1 h at 60 °C. In this step, high evaporation rates should be avoided to prevent crack formations on the final film. Finally, two pieces of n class="Chemical">copper wire were connected to the electrodes by silver paste, and they were kept in 100 °C for 1 h until the paste was dried completely (Figure ).

Figure 1

Silver IDEs (a) before the composite deposition and (b) after the composite deposition.

Silver n class="Species">IDEs (a) before the composite deposition and (b) after the composite deposition.

The observed resistance of this layer was about 120 kΩ at room temperature (25–27 °C) which was acceptable for the purpose of this report.

Gas-Sensing Procedure

A home-made sensor apparatus was used for testing of the fabricated sensor which included a well-sealed Teflon/Pyrex chamber (3.7 L volume) (Figure ). The fabricated sensor was located in the testing chamber, and while the chamber lid was closed, specified amounts of liquid VOCs were imported through Agilent GC syringes (1 or 10 μL) onto top of a heater. For each VOC, the temperature of the heater was set at its boiling point, so the VOCs were evaporated immediately after reaching the heater surface. Resistance of the sensing layer was measured by a digital multimeter (Hioki DT4252). In order to make a homogeneous concentration of vapors in the chamber, a fan was installed inside the chamber. Also, the chamber was equipped with a humidity sensor, which showed the relative humidity (RH) inside the container during the sensing process. All of the experiments were performed at room temperature, and the temperature was monitored by a thermometer placed into the chamber.

Figure 2

Gas sensor testing apparatus.

The main analyte in this report was formaldehyde, and other VOCs including n class="Chemical">methanol, ethanol, acetone, and acetonitrile were analyzed for comparison. After injection of different amounts of these VOCs, resistance of the sensing material increased instantly. The resistance changes were recorded during the time until it reached an equilibrium. Then, the gas molecules were released by introducing fresh air into the chamber. Every test was repeated three times for investigating the repeatability of the sensor. Furthermore, formaldehyde sensing experiment was performed under different humidity conditions to examine the effect of humidity on the sensor performance. For other experiments, the RH of the chamber was 18–20%, and the air temperature was 25–27 °C.

Results and Discussion

Characterization of Prepared Materials

The structures of synthesized ZIF-8, functionalized n class="Chemical">MWCNTs, and ZIF-8/MWCNT composite were characterized by XRD analysis (Figure ). The pattern of ZIF-8 is in perfect agreement with the literature.[27,28] The sample has pure phase of ZIF-8, and predominant peaks at 2θ = 7.44, 10.43, 12.79, 14.81, 16.50, 18.13, 22.19, 24.62, 26.79, 29.77, 30.71, 31.59, and 32.47° are attributed to the (110), (200), (211), (220), (310), (222), (411), (332), (422), (431), (440), (433), (442), and (532) planes, respectively.[27] In addition, the sharpness and narrowness of the peaks prove the high crystallinity of the structure. The comparison between ZIF-8 in pure form and composite form indicates that a combination of MWCNTs with ZIF-8 had no significant effect on the XRD pattern of this material. As it is evident in the inset of Figure , MWCNT has only one weak peak around 2θ = 25.75°, which is ignorable compared to strong peaks of ZIF-8. Also, the concentration of MWCNTs in the composite form is too low to make change in the XRD pattern of ZIF-8.

Figure 3

XRD patterns of ZIF-8 and ZIF-8/MWCNT nanocomposite. The inset graph shows the XRD pattern of functionalized MWCNTs.

XRD patterns of ZIF-8 and n class="Chemical">ZIF-8/MWCNT nanocomposite. The inset graph shows the XRD pattern of functionalized MWCNTs.

The morphologies of ZIF-8 and n class="Chemical">ZIF-8/MWCNT are investigated by SEM images. Figure a shows the uniform crystals of ZIF-8 with sharp facets, which have a particle size in the range of 80–120 nm. The crystals have a rhombic dodecahedral morphology that is a typical shape for ZIF-8, which has 12 exposed (110) faces. This claim is proved by the XRD pattern of ZIF-8 in which the sharpest peak is related to the mentioned structure.[27]Figure b shows the MWCNT/ZIF-8 composite in which the existence of MWCNT strands among ZIF-8 crystals can be observed. These strands link the crystals to each other and make paths for electron transfer, which gives the semiconductivity feature to the composite. MWCNTs are hardly seen in the mixture, first because the amount of these materials is lower than ZIF-8 crystals. Another reason is that carboxyl-functionalized MWCNTs had interaction with Zn2+ of ZIF-8[23] and are covered with these crystals, so MWCNTs are not so clear in the SEM image.

Figure 4

SEM images of (a) ZIF-8 and (b) ZIF-8/MWCNT nanocomposite.

For investigating the chemical bond vibrations of ZIF-8 and n class="Chemical">ZIF-8/MWCNT, FTIR spectra of the sample were studied in the range of 4000–400 cm–1 (Figure ). ZIF-8 spectra is in good agreement with the literature.[29] The bands at 3133 and 2928 cm–1 are attributed to the aromatic and aliphatic C–H stretches of the imidazole ring, respectively. The peak at 1587 cm–1 can be associated with the C=N stretch mode. The intense bands at 1350–1500 cm–1 can be referred to the entire ring stretching and the bands in the region of 900–1350 cm–1 are for the in-plane bending of the ring, whereas those between 600 and 800 cm–1 are associated with out-of-plane bending. The band at 419 cm–1 exhibits the Zn–N stretch mode and shows that Zn cations are connected to nitrogen atoms of Hmim successfully.[29]

Figure 5

FTIR spectra of ZIF-8, COOH-functionalized MWCNTs, and ZIF-8/MWCNT nanocomposite.

FTIR spectra of ZIF-8, n class="Chemical">COOH-functionalized MWCNTs, and ZIF-8/MWCNT nanocomposite.

FTIR spectra of acid-functionalized MWCNTs shows a broad peak at 3421 cm–1, which is associated with the O–H stretch of carboxylic groups. The peak at 1730 cm–1 is attributed to the C=O stretch mode of carboxylic groups and indicates that n class="Chemical">MWCNTs are successfully functionalized. Also, the C–O stretch mode appeared at 1283 cm–1, and the peak at 1638 cm–1 indicates aromatic C=C groups.[30,31]

In the composite form, all of the ZIF-8 peaks can be seen and no significant change is observed because of the low concentration of n class="Chemical">MWCNTs in the composite. However, the peak of 1738 cm–1 in the composite form can be associated with the C=O stretch mode of the carboxylic group of MWCNTs. Also, the broad peak of O–H stretch in 3448 cm–1, which is intensified in the composite in comparison to pure ZIF-8, is another sign of the presence of MWCNTs in the composite.

The porous structure of evacuated ZIF-8 and n class="Chemical">ZIF-8/MWCNT composite was evaluated by N2 adsorption and desorption in the range of p/p0 = 0.01–0.99. According to the IUPAC classifications,[32] a combination of type I and type IV isotherm behaviors was observed for both samples with a small hysteresis loop at relatively high pressures (Figure ), which reveals the dominance of microporous nature with minor mesopores. At the same pressure, the composite exhibits lower N2 adsorbed quantity than pure ZIF-8 as a consequence of some probable aggregations in the composite and blocking of pore openings of ZIF-8 crystals because of MWCNT presence.

Figure 6

Nitrogen adsorption/desorption of (a) ZIF-8 and (b) ZIF-8/MWCNT nanocomposite at 77 K.

The Langmuir surface area, BET surface area, total pore volume, t-plot micropore volume, and average pore width for ZIF-8 and n class="Chemical">ZIF-8/MWCNT composite are calculated by using the obtained data and are summarized in Table . For synthesized ZIF-8, as shown in Table , the BET surface area is 1483.13 m2/g which is higher than the report with a similar synthesis method.[23] It might be due to remaining unreacted precursors in cavities which common washing or heating is unable to bring out; so in this work, an extra washing step was added by keeping the particles in methanol for 1 day. The BET surface area of 1412.76 m2/g was obtained for the ZIF-8/MWCNTs nanocomposite, which is a little lower than for the pure ZIF-8. This decrease in the internal surface area is expectable because of MWCNT presence. However, this decrease is not so significant and has a negligible effect on the performance of the ZIF-8/MWCNT nanocomposite as a highly porous sensing material. Also, according to the results shown in Table , the average pore sizes of ZIF-8 and ZIF-8/MWCNT are about 2.4 and 2.3 nm respectively, which confirm the isotherm types mentioned above.

Table 1

Textile Properties of ZIF-8 and ZIF-8/MWCNT

	Langmuir surface area (m2/g)	BET surface area (m2/g)	total pore volume at p/p0 = 0.97 (cm3/g)	t-plot micropore volume (cm3/g)	average pore width (4V/A by BET) (Å)
ZIF-8	1961.86	1483.14	0.91	0.64	24.51
ZIF-8/MWCNT	1861.62	1412.76	0.82	0.61	23.25

Sensor Performance Parameters

To verify the performance of the sensor, the response of the sensor was measured on exposure to different concentrations of formaldehyde (5–100 ppm). Injection of analyte and adsorption of gas molecules onto the surface and pores of the nanocomposite caused an increase in the electrical resistance of the sensing material. Figure shows the response–recovery cyclic diagram of the sensor on exposure to different n class="Chemical">formaldehyde concentrations at room temperature. The separated diagram for each concentration is shown in Figure S1. As is evident, the sensor shows high response to formaldehyde. In addition, this sensor was able to detect a trace amount of formaldehyde as low as 5 ppm. This considerable response even at low concentrations can be related to the high surface area and the microporous structure of the ZIF-8 material.

Figure 7

Response–recovery cycles of the sensor on exposure to 5–100 ppm of formaldehyde (at 25–27 °C and 18–20% RH).

Response–recovery cycles of the sensor on exposure to 5–100 ppm of formaldehyde (at 25–27 °C and 18–20% n class="Chemical">RH).

Because formaldehyde is a reductive gas and can cause an increase in resistivity (or reduction of conductivity) of the sensor, it can be concluded that the obtained composite is a p-type semiconductor.[5] n class="Chemical">Formaldehyde decreases the conductivity by donating electrons to the valence band of the sensing material, which decreases the number of holes and increases the separation between the Fermi level and the valence band. This increases the resistance of the material, which is illustrated in Figure .

Figure 8

Schematic of p-type semiconductor on exposure to reductant gas.

In contrast to n-type ones, as the temperature increases, the resistance variation of p-type semiconductors toward the reducing gases decreases. Therefore, p-type semiconductors have relatively lower operating temperatures than n-type ones.[7]

In practical sensors, the value of resistance needs to return back to the initial value after evacuation of the analyte from the chamber.[7] As can be seen in Figure , in all concentrations, the resistance has returned to the initial value after removing the formaldehyde from the chamber. It is a good confirmation for the reusability (or reversibility) of the sensor. Furthermore, for every concentration, sensor measurements were repeated three times, and as it can be seen in Figure , similar results were obtained which proves the repeatability of the fabricated sensor.

A comparison between the performances of the sensor for formaldehyde and some other VOCs is shown in Figure and the real-time response of the sensor to n class="Chemical">ethanol, methanol, water, acetone, and acetonitrile is shown in Figure S2. For this comparison, the response of the sensor was calculated usingwhere R is considered as the electrical resistance of the film on exposure to the analyte and R0 is the electrical resistance of film on exposure to air.

Figure 9

Sensor response to 100 ppm of different analytes.

All tests were performed at 100 ppm concentration of analytes at room temperature with 18–20% RH. According to the data presented in Figure , for different analytes at the similar condition, the highest response was achieved for the n class="Chemical">formaldehyde. Response of the sensor to 100 ppm of water vapor was also measured, and no significant sensor response was observed. It means that the proposed sensing material has shown good stability against moisture. This observation can be related to the hydrophobic nature of the ZIF-8 structure.[33]

As mentioned before, ZIF-8 lacks conductivity, and its electrical resistance is much more than can be measured by common multimeters. So, pure n class="Chemical">ZIF-8 cannot be used as a sensing material of the resistive sensor and should be used with some conductive materials to be able to exhibit electrical signals. So, a composite of ZIF-8 with MWCNTs was used in this work. As we know, MWCNTs are used as gas-sensing materials too, and their presence helps the sensing process. However, since the surface area of ZIF-8 is much more than that of MWCNTs, the sensitivity of the ZIF-8/MWCNT composite is obviously higher than that of pure MWCNTs. Furthermore, in the absence of the analyte, the MWCNT sensor was unable to reach the initial resistance (Figure ).

Figure 10

Response of ZIF-8/MWCNT sensor and pure MWCNT sensor to different concentrations of formaldehyde at room temperature.

Response of ZIF-8/n class="Chemical">MWCNT sensor and pure MWCNT sensor to different concentrations of formaldehyde at room temperature.

Real-time resistive response of the sensor after exposure to different formaldehyde concentrations is shown in Figure . After introducing n class="Chemical">formaldehyde into the chamber, it took some time until the resistance reached a plateau. This time interval is known as the response time, and it depends on the concentration of the analyte. Response and recovery times of the sensor give us an idea about how fast the sensor works. In sensor reports, for evaluation of the response time, t90 is measured, which is defined as the time needed for the sensor to reach from the initial equilibrium state to 90% of the final equilibrium state after analyte entrance. Based on the achieved data, t90 for the analyte concentration of 5 and 100 ppm was 0.2 and 1.28 min, respectively, which shows the fast response of the sensor. Also, the recovery time for the obtained sensor is within seconds, which demonstrates the fast performance of the sensor.

Figure 11

Real-time resistive response of the sensor in exposure to 5–100 ppm of formaldehyde.

Since RH usually affects the sensor performance, it is necessary to consider its probable effect on sensor evaluations. Therefore, the sensor performance was compared under different n class="Chemical">RH conditions. For this purpose, first, a specified amount of water was injected into the evacuated chamber to create a specific amount of RH, and the system was allowed to reach the stable condition. Then, 50 ppm formaldehyde was introduced into the chamber, and resistance variations were recorded. Figure shows the resistance variations during time in the presence of 20, 40, and 60% RH. It is evident from this figure that at humid environments, the response of the sensor has decreased a little and the presence of water vapors in the environment has increased the response time. Nevertheless, even under these conditions, high and fast response could be observed. It is notable that after evacuation of the chamber, for reaching RH of 20%, water vapor with a concentration of more than 6000 ppm was introduced into the chamber, which is much higher than formaldehyde concentration in this condition (50 ppm).

Figure 12

Effect of RH on sensor response for detection of 50 ppm formaldehyde.

Effect of RH on sensor response for detection of 50 ppm n class="Chemical">formaldehyde.

Stability is an important parameter which tells how long a sensor can work without any remarkable changes in performance. For testing the sensor stability, the fabricated sensor was tested for a couple of months, and its response to formaldehyde vapors at three different concentrations was compared. Figure shows that no significant change was seen in the sensor performance during this period.

Figure 13

Response of the sensor on exposure to 25, 50, and 75 ppm formaldehyde during the time.

For evaluation of sensor performance, the calibration curve is used, which depicts how the output signal of an analytical device changes with the different concentrations of a specific analyte and the derived calibration function is generally used for determining the unknown concentrations of the analyte.[6]

The calibration curve of formaldehyde is exhibited in Figure and shows an almost linear relationship between the response of the sensor and the concentration of formaldehyde vapor with a correlation coefficient (R2) of 0.9985.

Figure 14

Calibration curve of the sensor in the concentration range of 5–100 ppm formaldehyde.

The minimum concentration of vapor at which the gas sensor still supplies reliable signals is defined as the limit of detection (LOD). LOD is the lowest amount of an analyte that a sensor can distinguish from the absence of the analyte at a known confident level. A lower LOD shows higher sensitivity.[6] The signal value of LOD (yLOD) is calculated by:where yb is the mean value of blank signal (the intercept of calibration curve) and Sb is the standard deviation of the blank measurements and is calculated by:[34]where y is the experimental signal of every concentration, y̅ is the theoretical value calculated by the equation of calibration curve for every given concentration, and n is the number of different concentrations used to plot the calibration curve. Then, LOD in terms of concentration is calculated from yLOD. According to the achieved calibration function for formaldehyde and the above equations, LOD for n class="Chemical">formaldehyde detection by the ZIF-8/MWCNT sensor was calculated as low as 4.83 ppm. It proves that in this range of formaldehyde concentration (above 5 ppm), the sensor gives reliable responses.

Conclusions

Nanocomposite of ZIF-8 nanoparticles and functionalized-n class="Chemical">MWCNTs was synthesized at room temperature and used as the sensing material onto IDEs to fabricate a resistive gas sensor. The achieved ZIF-8/MWCNT nanocomposite was used for detection of formaldehyde vapors at room temperature (25–27 °C) and showed fast and high responses because of the high surface area of the nanocomposite (1412.76 m2/g). LOD for formaldehyde was calculated as low as 4.83 ppm, and the response of this sensor to formaldehyde was found higher than some other vapors, including methanol, ethanol, acetone, acetonitrile, and water with the same concentration. The effect of RH on the gas response was studied as well and the sensor showed good performance under different RH conditions. In addition, contrary to metal-oxide sensors that are used under high temperatures, this sensor is used at room temperature. To the best of our knowledge, this is the first report on ZIF-8 as the main sensing material of resistive gas sensor working at room temperature for the detection of formaldehyde.