Monitoring the Concentration of Carbon Dioxide by Measuring the Change in the

Resistance of TiO2 Nanowires Doped with Potassium

Chris Jaejun Lee

Korea International School

Abstracts

The usage of the gas sensor has been increasing very rapidly in industry and in daily life for

various potential applications. In recent years, metal oxide semiconductors have become the

primary choice for designing highly sensitive, stable, and low-cost real-life applications-

based gas sensors due to their inherent physical and chemical properties. In this study, the K-

doped TiO2 nanowires showed the possibility of use as a sensing material due to their

consistent response to the presence of CO 2 gas. The reproducibility of sensor assembly has

been proved by testing seven different sensors fabricated using the same procedure. In

addition, treatment with H2 plasma improved the sensitivity of the CO 2 gas sensor made of K-

doped TiO2 nanowires. The sensing mechanism was attributed to electron adsorption to the

TiO2 as a result of CO2 reduction on the surface of K-doped sites. It reduces the hole

accumulation layer due to electron–hole recombination process, so the surface resistance of

K-doped TiO2 increases depending on the amount of CO2 adsorption. This development

suggests opportunities for the fabrication of simple, inexpensive CO 2 microsensors with low

power consumption.

1. Introduction
 Gas sensors have been developed to monitor the presence of carbon dioxide (CO 2) in

the environment because of its significant role in global warming. Recently the partial

pressure of CO2 gas generally ranges from 410 to 420 ppm ((https://gml.noaa.gov/ccgg/,

Earth system research laboratories, Global monitoring laboratory)), it is important to monitor

the concentration of CO2 in the atmosphere in order to understand its impact on climatic

change. In addition to its effect on climate change, CO 2 gas also has a significant effect on the

corrosion rate of the industrial equipment used in chemical processing and the combustion

efficiencies in automotive applications. Because of the prevalence of CO 2 and its numerous

negative effects, small, inexpensive, reliable solid-state CO 2 gas sensors with low power

consumption have been in high demand for several years ((K. Sahner, A. Schulz, J. Kita, R.

Merkle, J. Maier and R. Moos. CO2 Selective Potentiometric Sensor in Thick-film

Technology. Sensors. 8, 4774-4785. (2008).))’((C. Wang, L. Yin, L. Zhang, D. Xiang and R.

Gao. Metal Oxide Gas Sensors: Sensitivity and Influencing Factors. Sensor. 10, 2088-2106.

(2010).)).

Carbon dioxide is a gaseous chemical compound that is composed of two oxygen

atoms that are covalently bonded to a single carbon atom. The central carbon atom is

connected to two oxygen atoms, and there are no lone pairs of electrons around the central

atom. Therefore, carbon dioxide has a linear geometry, so the angle of the O−C−O bond is

180°, and it is a non-polar molecule.

Despite various approaches for monitoring the concentration of CO 2, only a limited

number of CO2 sensing materials exist due to the stable chemical properties of CO 2 gas ((J.

Maier. Electrochemical sensor principles for redox–active and acid-base–active gases.

Sensors and Actuators B. 65, 199-203. (2000).)). Currently, the most common type of CO 2

sensor is a potentiometric device that consists of either an Na + or an Li+ electrolyte and alkali

metal carbonate sensing electrodes. The potentiometric CO 2 sensors are promising because
they show satisfactory EMF signals over a wide range of gas concentrations at high

temperatures ((J-S. Lee, J-H. Lee, S-H. Hong. NASICON-based amperometric CO2 sensor

using Na2CO3–BaCO3 auxiliary phase. Sensors and Actuators B. 96, 663–668. (2003).))’((I.

Lee, S.A. Akbar. Potentiometric carbon dioxide sensor based on thin Li 3PO4 electrolyte and

Li2CO3 sensing electrode. Ionics. 20, 563–569. (2014).)). Most existing CO2 sensors are large,

bulky devices that involve complicated fabrication processes. Because of its numerous

negative effects, small, inexpensive, reliable solid-state CO 2 gas sensors with low power

consumption have been in high demand for several years ((K. Sahner, A. Schulz, J. Kita, R.

Merkle, J. Maier and R. Moos. CO2 Selective Potentiometric Sensor in Thick-film

Technology. Sensors. 8, 4774-4785. (2008).))’((C. Wang, L. Yin, L. Zhang, D. Xiang and R.

Gao. Metal Oxide Gas Sensors: Sensitivity and Influencing Factors. Sensor. 10, 2088-2106.

(2010).)).

To overcome the disadvantages of potentiometric CO 2 sensors, measurable interaction

has been studied. It is related to a surface reaction caused by adsorption–desorption, which

leads to a change in the electrical conductance (or resistance) of metal oxides as a result of

the charge-carrier transfer ((G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions. Metal oxide

semi-conductor gas sensors in environmental monitoring. Sensors. 10, 5469–5502. (2010).)).

The surface reaction of metal oxide sensors is driven by many factors, such as

semiconducting nature (n-type or p-type), target gas (reducing or oxidizing gas), size, and

morphology.

To achieve a miniaturized CO2 sensor, we have been aggressively exploring new CO 2

sensing materials by measuring electrical conductance in response to exposure to CO 2 gas.

Thus, we report the possibility of using nanocrystalline TiO 2 doped with potassium as a

resistive CO2 sensing material, which we have demonstrated successfully for the detection of

CO2 gas. In this work, we investigated the feasibility of using K-doped TiO 2 nanowires for
the detection of CO2 gas, and the effect of the H2 plasma treatment of the nanowires on their

detection sensitivity.

2. Experimental methods

2.1. Fabrication of K-doped nanowires as sensor elements

We fabricated TiO2 nanowires by the moisture-assisted, direct oxidation of titanium in

the presence of KOH. Thin films of Ti were prepared by using a DC magnetron sputtering

system. Before deposition, we used a turbo molecular pump to reduce the pressure in the

chamber down to less than 1  10-7 Torr. A water-cooled Ti target, which had a diameter of 3

inches and a purity of 99.99%, was mounted at a distance of 13 cm from the substrate holder.

A pure, thin film of Ti was deposited at a constant operational pressure, i.e., 5 mTorr, in Ar

(99.999%) flowing at 45 sccm by heating the substrate to 400 C. A constant DC power of

120 W was maintained. The deposition time of Ti metal thin film was 20 minutes, and the

thickness of the Ti thin film was estimated to be 200 nm.

For the fabrication of the TiO2 nanowires, one drop (2 mL, area of ~1 cm 2) of a 1 wt%

aqueous solution of KOH (Aldrich, 99+%) was placed at the center of the Ti deposited on the

substrate. Then, the substrate was inserted into a quartz tube and placed in a horizontal tube

furnace (Lindberg, TF55035A) for the heat treatment. First, the tube was purged with 5% H 2

balanced with Ar gas flowing at a rate of 1000 mL/min for 10 min, and then the tube was

heated to 150 °C at a ramping rate of 5 °C/min. After 30 min at 150 °C, the temperature of

the furnace was increased at the rate of 30 °C /min to 650 °C, at which it was maintained for

4 hours in the humid Ar gas that was flowing at the rate of 50 mL/min. The humid gas was

prepared by bubbling the Ar gas through DI water at room temperature. The humidity was

monitored by a humidity sensor (HMP234, Vaisala), and the relative humidity of the gas was

maintained at approximately 60%. After heat treatment, the sample was cooled to room
temperature rapidly by air quenching to minimize the oxidation of the sample while it was

cooling.

2.2. Microstructure and XRD analysis

After the growth of the nanowires, the samples were washed with deionized water for

24 hr until the pH reached about 7 with stirring to remove any excess KOH. Then, the

samples were dried at 60 °C for 24 hr. X-ray diffraction (XRD) patterns of the products were

recorded using a Scintag PAD-V diffractometer with Cu Kα radiation at 45 kV and 20 mA in

2θ, ranging from 20o to 60o. Field emission scanning electron microscopy (FESEM, Model

XL-30, Philips) images were obtained.

2.3. Measurement of sensing characteristics

Sensing tests were performed by attaching Au electrodes and Au lead wires on top of

the TiO2 nanowires on the Al2O3 substrate, as shown in Figure 1. The Au electrodes were

cured at 500 C for 1 h. The sensor was held in a tube furnace and heated from 500 C at

different H2 levels from 2% to 10%, and then CO 2 gas was introduced into the furnace. The

response of the sensor was observed in the range of 500 to 2500 ppm of CO 2, and the change

in the resistance was measured at each CO2 concentration.

2.4. Condition of H2 plasma treatment

To improve their sensitivity, the K-doped TiO 2 nanowires were treated with H 2

plasma at 350 °C. H2 Plasma treatments were performed with ICP source frequency 2 MHz

and power 30 W. The RF frequency applied to the chuck was 13.56 MHz, with the chuck

power 200W. H2 gas flow rate was 50 sccm. The source power controls the ion density in the

plasma, while the chuck power determines the energy of ions in plasma.

3. Results

3.1. Characteristics of K-doped TiO2 nanowires

 The SEM images of the nanowires obtained by the oxidation of the thin film of Ti

with KOH are shown in Fig. 1.

Fig. 1. SEM images of nanowires fabricated on thin Ti film with KOH

To understand the mechanism for nanowires formation, heat-treatment was

interrupted after 5 min and then surface morphology was investigated as shown in Fig. 1(a).

Small islands were formed, and the separate islands grew into short nanowires. If the

potassium was not continuous on the surface of Ti thin film and consisted of particles smaller

than 5 nm, the K particle could have acted as a catalyst to adsorb oxygen. Therefore, the

adsorption of oxygen is more predominant on K particles than others. Due to the difference of

oxygen supply, the growth rate of TiO 2 on the K particles might be drastically faster than that

of the sides, thus causing them to survive and maintain their morphology. After 30 min heat

treatment, as seen in Fig. 1(b), the Ti thin film shows well-formed nanowires that their

lengths were up to 5 μm with 5nm or less in diameter. The density of the nanowires was so

high and uniform on the entire area of the substrate that many of the nanowires were made

into large bundles. It is interesting to recognize that Some nanowires exhibit a rectangular
cross-section while others consist of connected polyhedral, as shown in Fig. 1(c) based on

ultra-high resolution SEM image.

There was no growth of the nanowires unless the KOH was used, and most of the Ti

film was oxidized easily after the heat treatment. Therefore, the amount of oxygen or the

oxidation sites on the thin film of Ti should be limited. KOH played an important role in the

formation of the TiO2 nanowires ((T. Zaremba, D. Witkowska. Effect of potassium titanate

whiskers on the performance of vacuum molded glass/epoxy composites. Materials Science-

Poland. 28(1), 25-41. (2010).)).

For the growth of TiO2 nanowires in a limited oxygen atmosphere, Ti always has to

be available at the droplets by the diffusion of Ti. In order to check this assumption, the

diffusion length of a Ti cation can be used. The chemical diffusivity (D) of Ti cation

interstitials in a single crystal rutile TiO2 in the temperature ranges of 900 ~ 1400 K was

reported by Radecka et al. ((M Radecka, P Sobaś, M Rekas. Ambipolar Diffusion in TiO2.

Solid State Ionics. 119(1–4), 55-60. (1999).)), and they presented the equation of the

diffusivity in the following form:

D = (0.001 ± 0.4) exp {(-48.7 ± 8 kJ/mol)/RT} cm2/s (1)

From equation (1), the diffusion length of titanium interstitials at 800 °C for 30 min is

estimated to be 5 μm by using √Dt. According to this calculation, there is enough Ti to grow

nanowires on the droplets.

In related previous work, potassium iodide was used as a catalyst for the growth of

tungsten oxide nanoribbons, and it was proposed that the nanoribbons grew by a vapor–

liquid–solid (VLS) mechanism from a liquid K-W-O alloy ((K. Q. Hong, W. C. Yiu, H. S.

Wu, J. Gao M. H. Xie. A simple method for growing high quantity tungsten-oxide

nanoribbons under moist condition. Nanotechnology. 16, 1608-1611. (2005).)). The results of
this earlier work may indicate that the likely mechanism for the growth of the K-doped TiO 2

nanowires from the K-Ti-O alloy droplets was similar to that of the tungsten oxide

nanoribbons from a K-W-O alloy. In general, the VLS mechanism has a eutectic reaction for

the liquid phase at the temperature of the heat treatment. However, there were no caps of the

catalyst at the end of the nanowires, which is a well-known characteristic of the VLS

mechanism ((V. Schmidt, J. V. Wittemann, S. Senz, U. Gosele. Silicon Nanowires: A Review

on Aspects of their Growth and their Electrical Properties. Advanced Materials. 21, 2681–

2702. (2009).))’((V.G. Dubrovskii, G.E. Cirlin, N.V. Sibirev, F. Jabeen, J.C. Harmand, P.

Werner. New Mode of Vapor−Liquid−Solid Nanowire Growth. Nano Letters. 11(3), 1247–

1253. (2011).)). Thus, there also is the possibility of growth by the VSS (vapor-solid-solid)

mechanism. On the surface of thin film of Ti, potassium can adsorb oxygen easily at the

limited oxygen atmosphere, so the oxidation of the thin film of Ti is more pronounced on

potassium. One-dimensional TiO2 can grow because of the role of the potassium ((H. Kim, H.

Na, J. Yang, D. Kim. Growth, structural, Raman, and photoluminescence properties of rutile

TiO2 nanowires synthesized by the simple thermal treatment. Journal of Alloys and

Compounds. 504, 217-223. (2010).)).

Fig. 2. (a) XRD pattern of the nanowires and (b) representative EDX spectrum

The XRD pattern indicated that the rutile phase was stable (JCPDS card: No 21-

1276). The dominant peaks at 26.9°, 53.7°, and 55.8° can be assigned to rutile-TiO 2 (110),

(211), and (220), respectively ((A. Wisitsoraat, A. Tuantranont, E. Comini, G. Sberveglieri,

W. Wlodarski. Characterization of n-type and p-type semiconductor gas sensors based on

NiOx doped TiO2 thin films. Thin Solid Films. 517, 2775–2780. (2009).)), as shown in Fig.

2(a). A strong (110) peak indicates preferential growth along c-axis parallel to the plane. The

XRD pattern also shows narrow and sharp peaks indicating good crystalline characteristics of

the samples. Fig. 2(b) shows the data of Energy Dispersive X-ray spectroscopy (EDX) was

performed in the SEM mode, and an EDX examination of the nanowires indicated that they

consisted of titanium, oxygen, and potassium.

3.2. Sensing behaviors of K-doped TiO2 nanowires

Fig.3. Resistance changes of TiO2 nanowires sensor from CO2 concentration at 450 C and
the different levels of H2 background gas

Fig. 3 shows the resistance changes to the CO 2 concentration from 500 ppm to 2500

ppm at 500 °C on as-grown, K-doped TiO2 nanowires. The sensing test temperature was 500

C, and the level of the H 2 background gas varied from 2% to 10%. Since the as-grown K-

doped TiO2 nanowires were p-type semiconductors and CO2 was categorized as a reducing

species, the resistance of the nanowires increases when they are exposed to CO 2 ((C.S.

Huang, B.R. Huang, C.H. Hsiao, C.Y. Yeh, C.C. Huang, Y.H. Jang. Effects of the catalyst

pretreatment on CO2 sensors made by carbon nanotubes. Diamond & Related Materials. 17,

624–627. (2008).)). The baseline of the sensors drifted slightly in air, and this observation is

being investigated further. Sensitivity (S) was defined as the ratio of [(R CO2−Ra) / Ra]×100,

where RCO2 is the resistance of the nanowires when they are exposed to specific

concentrations of CO2, and Ra was measured as the resistance in air.

Fig.4. (a) The sensitivity changes of TiO2 nanowire sensors at different levels of H2
background gas and (b) the sensitivity changes of the sensors at post H2 plasma treatment
with different times

Fig. 4(a) shows the sensitivity changes of the K-TiO 2 nanowire sensor at sensing

temperatures for different levels of H2 background gas. The sensitivity is defined as the

resistance of the sensor at a given concentration of the target gas (saturated R values in Figure

3) normalized by the resistance in the absence of the target gas (R0). Although the sensitivity

at a CO2 concentration of 500 ppm did not depend significantly on the levels of background

H2 gas, the sensitivity was improved by increasing the level of H 2 background gas at high
concentrations of CO2. According to this result, hydrogen on the surface of TiO 2 nanowires

played an important role in the absorption of CO2.

Fig. 4(b) shows the sensitivity at post H 2 plasma treatment with different times. The

sensitivity of the as-grown nanowires was 6.6% at a CO 2 concentration of 500 ppm. After

treatment with the H2 plasma for 5 min, the sensitivity increased from 6.6% to 9.5% at the

CO2 concentration of 500 ppm. The improvement of the sensitivity was not dependent on the

time of the plasma treatment. After the H 2 plasma treatments for times ranging from 5 to 20

min, the improvements were similar. This improvement might be due to physical and

chemical effects from ion bombardment, surface charges, and etching by the H 2 plasma ((A.

Prim, E. Pellicer, E. Rossinyol, F. Peiró, A. Cornet, J.R. Morante. A novel mesoporous CaO-

loaded In2O3 material for CO2 sensing. Advanced Functional Materials. 17, 2957–2963.

(2007).)). Hydrogen ion bombardment might modify the surfaces of the nanowires and create

dangling bonds that can react easily with CO 2 and produce hydrocarbonates on the nanowires

((Qiang Wang, Zhanhu Guo, Jong Shik Chung. Formation and structural characterization of

potassium titanates and the potassium ion exchange property. Materials Research Bulletin.

44(10), 1973-1977. (2009).)). The H2 plasma treatment on TiO2 is related to the number of

oxygen vacancies and Ti3+. The H2 plasma treatment might create electron-hole separation in

TiO2 and then enhances the surface-active states of TiO 2 by changing its electronic structure

from TiO2 (Ti4+) to Ti2O3 (Ti3+). This behavior makes a positive contribution to absorbing O 2

ions on TiO2 exposed to air, the sensitivity of the sensor can be enhanced by increasing the

hole accumulation layer on the surface of TiO 2. ((Shang-Chou Chang, Tsung-Han Li, Huang-

Tian Chan. Hydrogen Plasma Annealed Titanium Dioxide Oxide/Aluminum-doped Zinc

Oxide Films Applied in Low Emissivity Glass. Int. J. Electrochem. Sci., 16 1-8. (2021).))

Table 1. Sensing performance of p-type metal oxide towards CO2 gas

 Operatin
Response/ Long-Term
Synthesi g Gas/ Respon
Material Recovery Stability/Reproduc Ref.
s Method Temperat ppm se
Time (s) ibility
ure (°C)

((R. Wimmer-
Teubenbacher, F.
Sosada-Ludwikowska,
B.Z. Travieso, S.
Electron-
Defregger, O.
beam
Tokmak, J.S. Niehaus,
Au- lithograp
CO2/20 M. Deluca, A. Köck.
functionali hy, 300 365% 258 s/264 s 14 days/NA
00 ppm A. CuO thin films
zed CuO thermal
functionalized with
evaporati
gold nanoparticles for
on
conductometric carbon
dioxide gas
sensing. Chemosensor
s, 6, 56. (2018).))

((A. Chapelle, I. El
Younsi, S. Vitale, Y.
Thimont, T. Nelis, L.
Presmanes, A.
Barnabé, P. Tailhades.
RF Improved
CuO/ CO2/50
sputterin 250 40 3300 s/480 s NA/NA semiconducting
CuFe2O4 00 ppm
g CuO/CuFe2O4 nanostr
uctured thin films for
CO2 gas sensing. Sens.
Actuators B
Chem. 204, 407–413.
(2014).))

((G. Joshi, J.K. Rajput,

L.P. Purohit, L.P.
SnO2–Co3O4 pores
composites for CO2
Sol–gel
SnO2– CO2/20 gas sensing at low
spin 30 13.68 2 s/12 s NA/NA
Co3O4 00 ppm operating
coating
temperature. Micropor
ous Mesoporous
Mater. 326, 2–9.
(2021).))

The sensing results reported on p-type metal oxides towards CO 2 are summarized in

Table 1. Compared with other research about CO2 sensors, the K-doped TiO2 nanowire-based

sensor exhibited reasonable performance about both of response/recovery time and sensitivity

because of the increased surface area and the H 2 plasma treatment. In the case of the K-doped

TiO2 nanowire sensor, the 80% response time for the CO 2 gas was about 1~2 min, and the

recovery time was less than 1 min.

Fig 5. Sensitivity towards 500ppm CO 2 for 30 days from different 7 sensors with H 2 plasma
treatments for 20min

To prove the reproducibility of sensor assembly, seven different sensors were

fabricated using the same procedure. Fig. 5 plots the sensitivity values of the sensor with H 2

treatment 20min towards 500ppm CO2 concentration. The maximum sensitivity was 10.1 and

the minimum was 9.1, with a standard deviation of 0.32. To check long-term stability,

sensing behaviors were investigated for 30 days, with measurements taken every 2-3 days.

Fig. 5 shows that the sensitivity values at 500 ppm CO 2 were stable during this test period.

The worst sensor shows about 10% variation, however there is no continuous degradation of

the sensor during the trials. Therefore, we can attribute this to natural variation and assume

the sensor is operating stably for 30 days.

4. Discussion

4.1. Sensing mechanisms

 In resistive TiO2 gas sensors, gas detection is based on the change in electrical

resistance of TiO2 in response to exposure to a targeting gas. Adsorption of gases on the

surface of TiO2 results in the transfer of carrier charges between the gas and the TiO2, which

permits tuning of the electrical resistance. As TiO 2 is exposed to air, oxygen usually can be

adsorbed in three forms: molecular O2−, atomic O− and O2− species ((H.J. Kim, J.H. Lee.

Highly sensitive and selective gas sensors using p-type oxide semiconductors:

Overview. Sens. Actuators B Chem. 192, 607–627. (2014).)). Once oxygen ions are

chemisorbed, the concentration of free charge carriers (holes) increases in p-type materials.

Chemisorbed oxygen species can form a hole-accumulating layer near the TiO 2 surface.

When a targeting gas is introduced after chemisorption of O 2, it reacts with adsorbed oxygen

and may chemisorb by transferring electrons to the TiO 2. Accordingly, the recombination of

electron holes increases, narrowing the accumulation layer. Hence, the resistance of

resistance increases depending on the concentration of a targeting gas.

4.2. Enhancement of sensitivity

In general, one-dimensional nanostructures of metal oxides have led to more efficient

chemical sensors because of their large surface areas. In addition, the sensitivity improvement

also was contributed by the incorporation of the hydroxyl group (OH) on the surface of the

base metal oxide sensor ((Qiang Wang, Zhanhu Guo, Jong Shik Chung. Formation and

structural characterization of potassium titanates and the potassium ion exchange property.

Materials Research Bulletin. 44(10), 1973-1977. (2009).)). Two possible mechanisms could

be explained for the sensitivity improvement by the KOH droplets. First, a local depletion

area around the KOH droplets could have formed on the surface of the nanowire as a result of

a charge transfer between the KOH droplets and the TiO2 nanowire ((M.M. Arafat, B. Dinan,

S.A. Akbar, A.S.M.A.Hseeb. Gas Sensors Based on One Dimensional Nanostructured Metal-

Oxides: A Review. Sensors. 12, 7207 (2012).)). Atomic oxygen species dissociated on the
KOH droplets with strong catalytic abilities could make hydroxyl ions (OH -) as reaction sites

to absorb CO2 onto the surface of the nanowire. The other mechanism could be explained by

the TiO2 surface reaction. The surface of TiO2 can react immediately with hydrogen or humid

air and form hydroxyl groups. When the surface of the TiO 2 is fully hydroxylated, the oxide

ions in the oxide and hydrogen or humid air absorbed on the surface would form hydroxyl

groups ((Chung-Yi Wu, Kuan-Ju Tu, Jin-Pei Deng, Yu-Shiu Lo, and Chien-Hou Wu.

Markedly Enhanced Surface Hydroxyl Groups of TiO2 Nanoparticles with Superior Water-

Dispersibility for Photocatalysis. Materials. 10(5), 566. (2017).)). Surface bridging hydroxyls

could modulate the surface property of TiO 2 significantly, thereby improving its reactivity

with CO2.

4.3. Effect of doping metal on sensing behavior

Introducing noble metals such as Au, Ag, Pt, and Pd is one of the most effective and

widely used methods for improving the gas sensing response in the case of metal oxide

sensing materials. Firstly, adsorption activation energy is decreased, which can improve the

sensing response ((R. Zhou, X. Lin, D. Xue, et al. Enhanced H 2 gas sensing properties by Pd-

loaded urchin-like W18O49 hierarchical nanostructures. Sens Actuators B. 260, 900-907.

(2018).)). Another effect of doping with another metal is that it changes the size, porosity and

surface area of the metal oxide. As a result, gas molecules adsorption sites and diffusion

paths are modified ((H. Ji, W. Zeng, Y. Li. Gas sensing mechanisms of metal oxide

semiconductors: a focus review. Nanoscale. 11, 22664-22684. (2019).)). In most cases,

doping with another metal causes the reduction in grain size. When the size of the grain is

less than twice of the length of the Debye then the entire grain size is occupied by the

electron depletion layer. Therefore, the gas sensing property of metal oxide is improved ((H.

Chen, J. Hu, G-D. Li, Q. Gao, C. Wei, X. Zou. Porous Ga–In bimetallic oxide nanofibers
with controllable structures for ultrasensitive and selective detection of formaldehyde. ACS

Appl Mater Interfaces. 9, 4692-4700. (2017).)).

4.4. Future research

The K-doped TiO2 nanowire-based CO2 microsensor is a resistor-type sensor, which

can be integrated easily into a sensor array to detect various gases and to provide signals for

automatic control systems. To the best of our knowledge, this is the first time it has been

demonstrated that K-doped TiO2 nanowire sensing material can respond to various

concentrations of CO2 gas in a significant and consistent way. This development creates

opportunities for the fabrication of simple, inexpensive CO 2 microsensors with low power

consumption due to their small sizes.

There are still some questions which cannot be explained by using the existing gas

sensing mechanism. Therefore, it is necessary to do more research in this topic for a better

understanding of the topic. In this area, more experiments are required using modern

techniques such as situ analysis to know about the effect of different types of gas sensing

mechanism on the performance of gas sensors. Further exploration will include expanding the

detection range of CO2 concentration, reducing the operational temperature, and improving

the baseline stability by understanding interference among other gases.

5. Conclusion

Traditionally, transition metal oxides, such as TiO2 and SnO2, have been used

extensively in gas sensor applications for reactive gases such as CO and H2 because of the

changes in their electrical conductivities in the presence of a target gas. However, to date,

limited research about a CO2 sensor that uses transition metal oxide has been reported. K-

doped TiO2 nanowires are promising as potential CO2 gas sensors. After treatment with H2
plasma, the sensor showed greater sensitivity to CO 2 gas. Nevertheless, due to the inherent

difficulties associated with working at the nanoscale, the characterization of individual

nanowires are not straightforward tasks and require further research and development.

Although additional research is needed to detect the variable range of CO 2 concentration, and

to stabilize the baseline by investigating the interference with other gases including humidity,

the results of this study indicate that the K-doped TiO 2 nanowires provide a promising

opportunity for the development and use of simple, inexpensive CO2 microsensors.

Nanowires are one-dimensional (1D) nanostructures that have recently gained

immense attention as potential morphologies for the enhancement of chemical sensing

performances. Nanowires have been the subject of considerable attention in the last few

years. Several characteristics of p-type metal oxide nanowires, such as high surface-area-to-

volume ratios and high carrier-charge transport, make them a promising choice for the

fabrication of chemical sensors. Indeed, a few problems remain, such as a lack of selectivity

and high working temperatures, which require surface or bulk modification order to be

resolved.
References

[1] https://gml.noaa.gov/ccgg/, Earth system research laboratories, Global monitoring

laboratory

[2] K. Sahner, A. Schulz, J. Kita, R. Merkle, J. Maier and R. Moos. CO2 Selective

Potentiometric Sensor in Thick-film Technology. Sensors. 8, 4774-4785. (2008)

[3] C. Wang, L. Yin, L. Zhang, D. Xiang and R. Gao. Metal Oxide Gas Sensors: Sensitivity

and Influencing Factors. Sensor. 10, 2088-2106. (2010)

[4] J. Maier. Electrochemical sensor principles for redox–active and acid-base–active gases.

Sensors and Actuators B. 65, 199-203. (2000)

[5] J-S. Lee, J-H. Lee, S-H. Hong. NASICON-based amperometric CO2 sensor using

Na2CO3–BaCO3 auxiliary phase. Sensors and Actuators B. 96, 663–668. (2003)

[6] I. Lee, S.A. Akbar. Potentiometric carbon dioxide sensor based on thin Li 3PO4 electrolyte

and Li2CO3 sensing electrode. Ionics. 20, 563–569. (2014)

[7] G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions. Metal oxide semi-conductor gas

sensors in environmental monitoring. Sensors. 10, 5469–5502. (2010)

[8] T. Zaremba, D. Witkowska. Effect of potassium titanate whiskers on the performance of

vacuum molded glass/epoxy composites. Materials Science-Poland. 28(1), 25-41. (2010)

[9] M Radecka, P Sobaś, M Rekas. Ambipolar Diffusion in TiO2. Solid State Ionics. 119(1–

4), 55-60. (1999)

[10] K. Q. Hong, W. C. Yiu, H. S. Wu, J. Gao M. H. Xie. A simple method for growing high

quantity tungsten-oxide nanoribbons under moist condition. Nanotechnology. 16, 1608-1611.

(2005)

[11] V. Schmidt, J. V. Wittemann, S. Senz, U. Gosele. Silicon Nanowires: A Review on

Aspects of their Growth and their Electrical Properties. Advanced Materials. 21, 2681–2702.

(2009)
[12] V.G. Dubrovskii, G.E. Cirlin, N.V. Sibirev, F. Jabeen, J.C. Harmand, P. Werner. New

Mode of Vapor−Liquid−Solid Nanowire Growth. Nano Letters. 11(3), 1247–1253. (2011)

[13] H. Kim, H. Na, J. Yang, D. Kim. Growth, structural, Raman, and photoluminescence

properties of rutile TiO2 nanowires synthesized by the simple thermal treatment. Journal of

Alloys and Compounds. 504, 217-223. (2010).

[14] A. Wisitsoraat, A. Tuantranont, E. Comini, G. Sberveglieri, W. Wlodarski.

Characterization of n-type and p-type semiconductor gas sensors based on NiO x doped

TiO2 thin films. Thin Solid Films. 517, 2775–2780. (2009)

[15] C.S. Huang, B.R. Huang, C.H. Hsiao, C.Y. Yeh, C.C. Huang, Y.H. Jang. Effects of the

catalyst pretreatment on CO2 sensors made by carbon nanotubes. Diamond & Related

Materials. 17, 624–627. (2008)

[16] A. Prim, E. Pellicer, E. Rossinyol, F. Peiró, A. Cornet, J.R. Morante. A novel

mesoporous CaO-loaded In2O3 material for CO2 sensing. Advanced Functional Materials. 17,

2957–2963. (2007)

[17] Qiang Wang, Zhanhu Guo, Jong Shik Chung. Formation and structural characterization

of potassium titanates and the potassium ion exchange property. Materials Research Bulletin.

44(10), 1973-1977. (2009)

[18] Shang-Chou Chang, Tsung-Han Li, Huang-Tian Chan. Hydrogen Plasma Annealed

Titanium Dioxide Oxide/Aluminum-doped Zinc Oxide Films Applied in Low Emissivity

Glass. Int. J. Electrochem. Sci., 16 1-8. (2021)

[19] H.J. Kim, J.H. Lee. Highly sensitive and selective gas sensors using p-type oxide

semiconductors: Overview. Sens. Actuators B Chem. 192, 607–627. (2014)

[20] M.M. Arafat, B. Dinan, S.A. Akbar, A.S.M.A.Hseeb. Gas Sensors Based on One

Dimensional Nanostructured Metal-Oxides: A Review. Sensors. 12, 7207 (2012)

[21] Chung-Yi Wu, Kuan-Ju Tu, Jin-Pei Deng, Yu-Shiu Lo, and Chien-Hou Wu. Markedly

Enhanced Surface Hydroxyl Groups of TiO2 Nanoparticles with Superior Water-

Dispersibility for Photocatalysis. Materials. 10(5), 566. (2017)

[22] R. Zhou, X. Lin, D. Xue, et al. Enhanced H 2 gas sensing properties by Pd-loaded urchin-

like W18O49 hierarchical nanostructures. Sens Actuators B. 260, 900-907. (2018)

[23] H. Ji, W. Zeng, Y. Li. Gas sensing mechanisms of metal oxide semiconductors: a focus

review. Nanoscale. 11, 22664-22684. (2019)

[24] H. Chen, J. Hu, G-D. Li, Q. Gao, C. Wei, X. Zou. Porous Ga–In bimetallic oxide

nanofibers with controllable structures for ultrasensitive and selective detection of

formaldehyde. ACS Appl Mater Interfaces. 9, 4692-4700. (2017)

[25] R. Wimmer-Teubenbacher, F. Sosada-Ludwikowska, B.Z. Travieso, S. Defregger, O.

Tokmak, J.S. Niehaus, M. Deluca, A. Köck. A. CuO thin films functionalized with gold

nanoparticles for conductometric carbon dioxide gas sensing. Chemosensors, 6, 56. (2018)

[26] A. Chapelle, I. El Younsi, S. Vitale, Y. Thimont, T. Nelis, L. Presmanes, A. Barnabé, P.

Tailhades. Improved semiconducting CuO/CuFe2O4 nanostructured thin films for CO2 gas

sensing. Sens. Actuators B Chem. 204, 407–413. (2014)

[27] G. Joshi, J.K. Rajput, L.P. Purohit, L.P. SnO 2–Co3O4 pores composites for CO2 gas

sensing at low operating temperature. Microporous Mesoporous Mater. 326, 2–9. (2021)