CeO₂ nanofibers for in situ O₂ and CO sensing in harsh environments

Abstract

As the most common type of harsh environment sensors, high temperature gas sensors are of paramount importance to improve combustion efficiency and control emissions. In this study, large scale and polycrystalline CeO₂ nanofibers were prepared with an average diameter of 376 ± 55 nm using a facile two-step synthesis route including electrospinning and calcination. The as-prepared CeO₂ nanofibers show good morphological and structural stability in high temperature environment (800–1000 °C). CeO₂ nanofibers were further employed for in situ, real-time oxygen (O₂) and carbon monoxide (CO) sensing at 800 °C and 1000 °C, respectively, both of which showed sensitive, reversible, and reproducible response. The mechanism for distinguishable response towards O₂ and CO was discussed. These results indicate that electrospun CeO₂ nanofibers are a promising nanomaterial for gas sensing in harsh environments.

---

Introduction

Environmental pollution from combustion engines and power plants has driven an increasing need for reliable gas sensors that can operate in harsh environments especially high temperatures (>500 °C). In particular, high temperature oxygen (O₂) and carbon monoxide (CO) sensors have played a key role in pollution control and fuel consumption efficiency in wide industrial fields, including automobile engine management, control of chemical processes as well as optimization of boilers, steel and cement industries, biological and food processing plants.^1,2 For example, oxygen sensors used in automotive exhaust emission control systems are dominating the applications of solid-state gas sensors worldwide.³ The control of the air/fuel ratio in automobile engines can improve the fuel economy efficiency leading to the reduction of the harmful emissions.⁴ In addition, monitoring all relevant emission components (on-board diagnosis) requires a second oxygen sensor downstream of the catalyst.³ Commercialized oxygen gas sensors are mainly the potentiometric type sensors with a solid electrolyte of ZrO₂, which measure different oxygen concentration between a reference electrode and a sensing electrode as a voltage output. A reference gas and electrode are required for sensor operation, therefore this type sensor usually has a complicated configuration.^5,6 By contrast, resistive oxygen gas sensors usually have a very simple structure, thus allowing for the size to be reduced. Therefore, resistive oxygen gas sensors have drawn a lot of attention recently. It has been reported that TiO₂,^7,8 SrTiO₃,^9,10 Ga₂O₃,^5,11 and CeO₂^6,12,13 can be used as materials for resistive oxygen sensors.

Besides high temperature O₂ sensors, high temperature CO gas sensors are also in high demand for controlling the emission of CO due to the toxicity of CO to human health. Carbon monoxide is a colorless, odorless, poisonous gas formed when carbon in fuels is not fully burned. Automobile exhaust gases can contribute about 60% of all CO emissions. The U.S. Environment Protection Agency (US EPA) health-based national air quality standard for CO is 9 ppm measured as an annual second-maximum 8 hour average concentration. Therefore, in situ detection of CO is essential to its control from combustion processes. However, CO is a strong reducing agent and could irreversibly interact with many sensing materials, resulting in the loss of the sensing function.

Currently, most of the reported high temperature O₂ and CO sensors operate at mild temperatures (<700 °C), and are not suitable for applications at a higher temperature (800–1000 °C). For example, in close proximity to engines, the exhaust gases can reach temperatures close to 1000 °C¹⁴ which require superior thermal stability of the sensor device including the sensing material, electrode and also the supporting substrate. In situ high temperature gas sensors, if successfully employed, could save huge amounts of energy across all energy-consuming industries. Up to date, commercially available sensor technology for such harsh environments is extremely limited due to the high requirements for sensing materials and sensor performance. Therefore, there is an urgent need to develop fast, sensitive, robust, and cost-effective high temperature gas sensors for the control of combustion-related processes.

Various metal oxides have been explored for applications in high temperature environments.^5–13 Among them, CeO₂ is one of the most promising materials in the fabrication of high temperature sensors (especially for oxygen) because CeO₂ possesses good thermal stability and exhibits a large diffusion coefficient for oxygen vacancy compared with TiO₂, resulting in quick changes of oxygen vacancy concentration in equilibrium state and thus fast response to oxygen.^15–17 In addition, CeO₂ has good corrosion resistance to corrosive gases (Cl₂, SO₂ and NO) compared with SrTiO₃ and Ga₂O₃.^16,18 In this study, CeO₂ nanofibers will be prepared for high temperature gas sensing.

In order to improve the sensing performance, the conventional thin film technology faces its fundamental limits¹⁹ because the miniaturization of the sensor device requires simultaneous sensitivity increase to compensate for the loss of total surface area. Therefore, nanostructured materials have emerged as alternatives because they can provide a large specific surface and enhance the sensing performance. Specifically, nanofibers fabricated via electrospinning possess specific surface approximately one to two orders of the magnitude larger than flat films, making them excellent candidates for potential applications in sensors.²⁰ By incorporating metal salts into the polymer, ceramic nanofibers can be fabricated by calcination following electrospinning.^21–23 However, the nanostructured sensors encounter certain challenges (e.g., the structure stability) in high temperature environments, given the fact that thermodynamically minimization of surface area is a spontaneous process.

In this report, novel CeO₂ nanofibers were successfully prepared by a two-step procedure which consists of electrospinning of Ce(NO₃)₃-PVP precursor nanofibers and a subsequent calcination process. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and X-ray diffraction (XRD) were applied to investigate the morphology and structure of the as-prepared sample. The CeO₂ nanofibers exhibited good morphological and structural stability at high temperature and showed sensitive, reversible and reproducible responses towards CO and O₂ at 800 °C and 1000 °C, indicating its potential application in harsh environments.

Experimental

Reagents

Cerium(III) nitrate hexahydrate (Ce(NO₃)₃·6H₂O) and dimethylformamide (DMF) were purchased from Acros Organics. Poly(vinyl pyrrolidone) (PVP, MW = 1 300 000) were obtained from Sigma-Aldrich. For gas sensing studies, high purity nitrogen (N₂, 99.998%), O₂/N₂ gas mixtures (97 ppm oxygen or 10% O₂ in N₂), and CO/Ar gas mixture (1% CO in Ar) were ordered from Airgas.

Preparation of CeO₂ nanofibers

0.386 g Ce(NO₃)₃·6H₂O and 0.386 g PVP were dissolved in 3 mL DMF and then stirred to form a homogeneous solution. The final mixture was loaded into a plastic syringe with a 23-gauge needle for electrospinning. The Ce(NO₃)₃-PVP precursory nanofibers were fabricated by electrospinning with a flow rate of 0.3 mL h⁻¹ at an applied voltage of 20 kV over a collection distance of 15 cm. The precursory nanofibers were then calcined in air as follows: first, calcined at 500 °C for 3 h in order to remove the matrix polymer and generate CeO₂ nanofibers; and then calcined at 1000 °C for 3 h to improve the crystal structure and thermal stability of CeO₂ nanofibers. The samples obtained after the two calcination steps are denoted as CeO₂₍₅₀₀₎ and CeO_2(1000), respectively. To study the thermal stability, the as-prepared CeO₂ nanofibers were treated at 1000 °C in air for 5 h followed by naturally cooling down to room temperature, and the whole process was repeated two more times. After three cycles of heating and cooling treatment, the final product was denoted as CeO_2(1000×3) and then subject to further characterization for chemical and structural stability.

Characterization of CeO₂ nanofibers

A JEOL 6335F field-emission scanning electron microscope (SEM) was employed to examine the morphology and the size of the as-electrospun nanofibers before and after calcination. More detailed morphology and selected area electron diffraction (SAED) patterns were studied using a Tecnai T12 transmission electron microscope (TEM) operated at 120 kV. The crystal structures of the samples were characterized by Oxford diffraction XcaliburTM PX Ultra with ONYX detector.

Sensor device fabrication

Homemade resistor-type CeO₂ nanofibers-based sensor is applied for in situ gas sensing in high temperature environment. Briefly, CeO_2(1000) nanofibers, serving as the sensing material, were cast in a small dent (500 μm in depth) on the top of an Al₂O₃ ceramic tube which has two holes to hold two Pt wires as electrodes. The sensor was connected to a CHI 601C electrochemical analyzer (CH Instruments Inc., USA) through two long Ni–Cr alloy wires and was placed in a furnace with a temperature control to conduct in situ high temperature gas sensing. The current output at a fixed 1 V DC bias was continuously measured.

O₂ and CO sensing at high temperature

The performance of the CeO₂ nanofibers-based gas sensor at high temperatures (800–1000 °C) was evaluated by measuring the resistance/conductance change upon exposure to different concentrations of O₂ and CO in a dynamic gas flow system. High purity nitrogen was used as carrier gas to obtain various O₂ or CO concentrations. Carrying N₂ and gas mixtures with a constant gas flow rate of 1.5 L min⁻¹ were regulated by a computer-controlled gas mixing system (S-4000, Environics Inc., USA). The current in the sensor was continuously measured and the electric resistance of the sensor was calculated by applying Ohm's Law (R = V/I). In a typical O₂ sensing experiment, the sensor placed in furnace at 800 or 1000 °C was first exposed to O₂/N₂ mixture for 7 min, followed by high purity N₂ for 18 min to recover the sensor, and then the “exposure/recovery” cycle was repeated. Upon the exposure to O₂, the resistance of the CeO₂ nanofibers-based sensor increases. The response of the sensor towards O₂ was evaluated by resistance ratio R_g/R₀, where R₀ is the initial electrical resistance of the sensor in high purity nitrogen and R_g is the measured real-time resistance upon exposure to oxygen mixture or pure nitrogen. On the contrary, the resistance of the sensor dramatically drops in the presence of CO, thus reverse responses towards O₂ and CO are obtained. As the resistance of the sensor in CO mixture is one to two orders lower than its original resistance, the reverse resistance ratio (R₀/R_g) was used to evaluate the CO sensing performance of the sensor. Since the response/recovery of the sensor towards “on/off” of CO is very fast, the exposure and recovery time in the CO sensing experiment are adjusted to 5 min and 10 min, respectively.

Results and discussion

Characterization of the material

The morphology changes of the electrospun nanofibers before and after two-step calcination were examined by SEM. As shown in Fig. 1A, the Ce(NO₃)₃-PVP precursory nanofibers displayed uniform and well-distributed fiber structure with an average diameter of 563 ± 66 nm. After calcination in air at 500 °C for 3 h, the PVP matrix was completely degraded and Ce(NO₃)₃ was decomposed to form CeO₂ (Fig. 1B). It can be observed that the average diameter of calcined nanofibers reduced to 374 ± 52 nm and the surfaces of CeO₂₍₅₀₀₎ nanofibers were not as smooth as the precursory nanofibers, which can be attributed to the loss of PVP, the decomposition of Ce(NO₃)₃ and the crystallization of CeO₂. Further calcination at 1000 °C for another 3 h was carried out to eliminate the impurities and improve the CeO₂ crystal phase and stability at high temperature. As shown in Fig. 1C, the CeO_2(1000) nanofibers display rougher surfaces compared to CeO₂₍₅₀₀₎ with a similar average diameter of 376 ± 55 nm, which implies no weight loss and improved crystallinity of CeO₂ during the calcination process at 1000 °C. More detailed morphology of CeO_2(1000) nanofibers was investigated by TEM, as presented in Fig. 2A. The diameter of the nanofiber was ∼386 nm, which was in good agreement with SEM results and small crystallites can be observed in the nanofiber. Selected area electron diffraction (SAED) pattern shows a typical ring pattern, suggesting the polycrystalline structure of the nanofibers (Fig. 2B).

Fig. 1 SEM images of (A) Ce(NO₃)₃-PVP nanofibers; (B) CeO₂₍₅₀₀₎ nanofibers (obtained after calcination of precursor nanofibers at 500 °C); (C) CeO_2(1000) nanofibers (obtained after CeO₂₍₅₀₀₎ was further calcined at 1000 °C).

Fig. 2 (A) A typical TEM image, (B) SAED pattern of CeO_2(1000) nanofibers.

The composition and crystal structure were further characterized by XRD. As shown in Fig. 3, the XRD pattern of CeO_2(1000) nanofibers matches the standard spectrum of JCPDS 65-5923. The diffraction peaks at 2θ values of 28.59, 33.13, 47.56, 56.43, 59.18, 69.53, 76.83, 79.21° correspond to CeO₂ (111), (200), (220), (311), (222), (400), (331) and (420) crystal planes, respectively. No other diffraction peak was observed, indicating the good phase purity. In order to examine the thermal stability of CeO_2(1000) nanofibers, the as-prepared sample was heated/cooled between 1000 °C and room temperature for 3 times. The SEM images (insets of Fig. 3) shows no obvious morphology change between CeO_2(1000) and CeO_2(1000×3), implying the good morphological stability of CeO₂ nanofibers at high temperature. In addition, after thermal treatment at 1000 °C 3 times, the XRD pattern of CeO_2(1000×3) nanofibers showed no peak shift compared with that of CeO_2(1000) nanofibers, indicating the as-prepared CeO₂ nanofibers also possess a good chemical and structural stability at high temperature. Such good thermal stability endows CeO₂ nanofibers with the potential application for high temperature resistive-type gas sensing.

Fig. 3 XRD patterns for the as-prepared CeO_2(1000) nanofibers, CeO_2(1000×3) nanofibers obtained after thermal stability study and the standard values for CeO₂. Insets show the SEM images of CeO_2(1000) and CeO_2(1000×3).

Oxygen and carbon monoxide sensing

The O₂ and CO sensing properties of the CeO₂ nanofibers-based sensor were examined at operation temperatures of 800 °C and 1000 °C, respectively. The I–V profiles of the sensor at 800 °C and 1000 °C are shown in Fig. S1†, indicating that the sensor obeys Ohm's Law at 1 V which was the applied bias in the sensing experiments. Fig. 4 presents the typical responses of the CeO₂ nanofibers as a function of time upon periodic exposure to different concentrations of O₂ and CO, respectively at both operation temperatures. As shown in Fig. 4 (top), upon the exposure of CeO₂ nanofibers to O₂, the resistance ratio R_g/R₀ of CeO₂ nanofibers increases rapidly and displays the concentration-dependent behavior with a response time (t₉₀, defined as the time when the normalized resistance change reached 90% of the maximum response after the exposure to oxygen) of 30 s at 800 °C and 28 s at 1000 °C (towards 800 ppm O₂), respectively. The actual t₉₀ is expected to be shorter by considering the time required to establish the pre-set O₂ concentration in furnace. In addition, the sensor responses towards three-time exposure of 500 ppm oxygen exhibited a very small relative standard deviation (RSD) of 0.85% at 800 °C and 2.6% at 1000 °C, indicating high reproducibility. On the contrary, exposure of the same sensor to CO rapidly decreases the resistance of CeO₂ nanofibers. One can see from Fig. 4 (bottom) that the reverse resistance ratio (R₀/R_g) of the sensor can reach as high as 48.3 towards 100 ppm CO at 800 °C, which means the resistance of the sensor decreases 48.3 folds compared to its original resistance in N₂. By purging with carrying gas, the response of CeO₂ nanofibers can be completely recovered, indicating the good reusability of CeO₂ nanofiber-based sensor. Furthermore, the sensor responses towards three-time exposure of 100 ppm CO also showed good reproducibility with a small RSD of 1.2% at 800 °C and 2.5% at 1000 °C. The reason for initial irregular response to CO can be explained as that the surface may have some unstable species which could interact with CO and finally are stabilized. The CO sensing performance of the CeO₂ nanofibers-based sensor in this work was compared with previous reported resistor-type CO sensors, as shown in Table S1†. To the best of our knowledge, this is one of the best results for resistor-type sensor for CO sensing at such high temperature. For multiple “on-off” cycles of O₂ or CO at both 800 and 1000 °C, the CeO₂ nanofibers show fast, sensitive, reversible, and reproducible responses for all cycles, suggesting that CeO₂ nanofibers are good sensing elements in the detection of O₂ and CO in high temperature harsh environments. The stable performance and fast response/recovery can be attributed to the porous nanofiber structure, which not only allows the free access of analyte molecules to CeO₂ and reduces the diffusion resistance, but also prevents the potential malfunction due to the considerable internal stress caused by thermal-expansion and/or gas molecule adsorption, thus offering an excellent sensing platform with stable performance.²⁴

Fig. 4 Typical responses of CeO₂ nanofibers-based sensor upon periodic exposure to different concentrations of oxygen (top) and carbon monoxide (bottom) at an applied DC bias of 1 V at 800 °C and 1000 °C.

The corresponding calibration curves (response vs. O₂ or CO concentration) are presented in Fig. 5 and Fig. 6, respectively. The gas sensing mechanism of CeO₂ nanofibers can be illustrated based on the electrical properties of the sensing element and the chemical property of the analyte molecules. In this study, the as-prepared CeO₂ is n-type, thus the predominant point defects in CeO₂ are the electron trapped by lattice Ce and oxygen vacancy. The classic mechanism of electrical conductivity for CeO₂ has been reported as follows:^13,25

Fig. 5 (A) Calibration curve for O₂ response of CeO₂ nanofibers-based sensor at an applied DC bias of 1 V at 800 °C and 1000 °C. (B) The normalized electrical conductivity of CeO₂vs. oxygen partial pressure pO₂ at 800 °C and 1000 °C.

Fig. 6 Calibration curve for CO response of CeO₂ nanofibers-based sensor at an applied DC bias of 1 V at 800 °C and 1000 °C.

where Ce^x_Ce and O^x_O represent the regular lattice ions, while Ce^'_Ce and V^••_O are the charged defects. Ce^'_Ce and V^••_O show the electron trapped by Ce^x_Ce and double ionized oxygen vacancy, respectively. Upon the exposure to O₂, oxygen can extract electrons in CeO₂ to form chemisorbed species on CeO₂ surface and adsorbed oxygen species can react with oxygen vacancies to form lattice oxygen resulting in the decrease of electron concentration, thus an increase the resistance of the sensor. Eqn (1) expresses the overall reaction. By considering charge neutrality and ignoring ionic conductivity, the change in the electrical conductivity as a function of oxygen partial pressure can be expressed as follows:

Therefore, the conductivity of CeO₂ nanofibers decreases with the increase of oxygen concentration, in good agreement with the observed results. However, compared to the theoretical value m = 6 in σ ∝ P(O₂)^−1/m, the obtained m values for CeO₂ nanofibers are 6.7 and 8.8 at 800 °C and 1000 °C (Fig. 5B), respectively, which are slightly larger than the theoretical value of 6. Similar phenomenon has been observed for other nanoscale ceria-based oxygen gas sensors.^6,16 The reason is not clear and still under study, but may be ascribed to the unique property (e.g., significant ionic conductivity, populated surface oxygen vacancies, etc.) associated with nanoscale CeO₂ nanofibers.

For CO sensing, the response of the sensor exhibited a linear tendency with increasing CO concentration, as presented in Fig. 6. N₂ is used as the carrying gas because CO and O₂ (or O₂ in air) could react to form CO₂ in the gas phase at the temperature higher than 605 °C which is the autoignition temperature of CO from SIRI Material Safety Data Sheets (MSDS). In the presence of reducing gas such as CO, the resistance of the sensor drops dramatically, which might be ascribed to two effects: First, at the early stage, the surface adsorbed oxygen species will be removed by CO and release the electrons back to the CeO₂; Second, after depletion of adsorbed oxygen species, CO will also extract lattice oxygen to form CO₂ which will generate more oxygen vacancies and electrons according to eqn (1). Upon the purging of high purity N₂, the response can be fully recovered, which is surprising. One possible explanation is the residue oxygen in the high purity N₂ gas can be adsorbed again on the CeO₂ surface and refill the oxygen vacancies, thus shifting eqn (1) to left side and resulting in the conductivity back to original level.

Operating at extremely high temperature (800 and 1000 °C), the CeO₂ nanofibers-based sensors showed very good stability, reversibility, reproducibility and excellent sensitivity towards CO. These good results indicate that electrospun CeO₂ nanofibers are a promising nanomaterial for gas detection in harsh environments.

Conclusions

A facile two-step process (electrospinning followed by calcination) has been successfully applied to fabricate CeO₂ nanofibers with an average diameter of 376 ± 55 nm. The as-prepared CeO₂ nanofibers possess highly porous structure, large surface to volume ratio, and good thermal stability in terms of morphology, chemical structure and crystalline structure, thus offering an excellent sensing material for CO and O₂ gas sensing in harsh environments. The fabricated CeO₂ nanofibers-based resistive type sensor exhibited sensitive, reversible, and reproducible concentration-dependent behavior to reducing gas (CO) and oxidizing gas (O₂) at 800 and 1000 °C. Due to different sensing mechanisms for reducing gas and oxidizing gas, the response to CO and O₂ can be easily distinguished based on the observed resistance increase or decrease. The stable performance and fast response/recovery can be attributed to free access of gaseous molecules to CeO₂ nanofibers. This study provides a promising route for the facile and cost-effective synthesis of high temperature stable nanomaterial which has great potential application in gas sensing in harsh environments.

Acknowledgements

We greatly appreciate the funding support from the Department of Energy (DOE).

References

1. J. W. Schwank and M. DiBattista, MRS Bull., 1999, 24, 44–48.
2. R. Ramamoorthy, P. K. Dutta and S. A. Akbar, J. Mater. Sci., 2003, 38, 4271–4282.
3. J. Riegel, H. Neumann and H. M. Wiedenmann, Solid State Ionics, 2002, 152, 783–800.
4. Y. L. Xu, X. H. Zhou and O. T. Sorensen, Sens. Actuators, B, 2000, 65, 2–4.
5. M. Ogita, K. Higo, Y. Nakanishi and Y. Hatanaka, Appl. Surf. Sci., 2001, 175, 721–725.
6. N. Izu, W. Shin and N. Murayama, Sens. Actuators, B, 2003, 93, 449–453.
7. R. K. Sharma, M. C. Bhatnagar and G. L. Sharma, Sens. Actuators, B, 1997, 45, 209–215.
8. L. Francioso, D. S. Presicce, P. Siciliano and A. Ficarella, Sens. Actuators, B, 2007, 123, 516–521.
9. J. Gerblinger and H. Meixner, Sens. Actuators, B, 1991, 4, 99–102.
10. W. Menesklou, H. J. Schreiner, K. H. Hardtl and E. Ivers-Tiffee, Sens. Actuators, B, 1999, 59, 184–189.
11. C. Baban, Y. Toyoda and M. Ogita, Thin Solid Films, 2005, 484, 369–373.
12. N. Izu, W. Shin, N. Murayarna and S. Kanzaki, Sens. Actuators, B, 2002, 87, 95–98.
13. N. Izu, W. Shin, I. Matsubara and N. Murayama, J. Electroceram., 2004, 13, 703–706.
14. R. Moos, Int. J. Appl. Ceram. Technol., 2005, 2, 401–413.
15. F. Millot and P. Demierry, J. Phys. Chem. Solids, 1985, 46, 797–801.
16. N. Izu, N. Oh-hori, M. Itou, W. Shin, I. Matsubara and N. Murayama, Sens. Actuators, B, 2005, 108, 238–243.
17. M. Kamiya, E. Shimada, Y. Ikuma, M. Komatsu and H. Haneda, J. Electrochem. Soc., 2000, 147, 1222–1227.
18. E. B. Varhegyi, I. V. Perczel, J. Gerblinger, M. Fleischer, H. Meixner and J. Giber, Sens. Actuators, B, 1994, 19, 569–572.
19. A. Kolmakov, Y. X. Zhang, G. S. Cheng and M. Moskovits, Adv. Mater., 2003, 15, 997–1000.
20. B. Ding, M. R. Wang, J. Y. Yu and G. Sun, Sensors, 2009, 9, 1609–1624.
21. D. Li and Y. N. Xia, Nano Lett., 2003, 3, 555–560.
22. X. G. Yang, C. L. Shao, Y. C. Liu, R. X. Mu and H. Y. Guan, Thin Solid Films, 2005, 478, 228–231.
23. Y. Q. Dai, W. Y. Liu, E. Formo, Y. M. Sun and Y. N. Xia, Polym. Adv. Technol., 2011, 22, 326–338.
24. W. Z. Jia, L. Su, Y. Ding, A. Schempf, Y. Wang and Y. Lei, J. Phys. Chem. C, 2009, 113, 16402–16407.
25. H. J. Beie and A. Gnorich, Sens. Actuators, B, 1991, 4, 393–399.

---

Footnote

† Electronic supplementary information (ESI) available: See DOI: 10.1039/c2ra20126g

---