Oxygen Vacancy Enhanced Gas-Sensing Performance of CeO2

bCollege of Physics and Information Technology, Shaanxi Normal University, Xi'an ... concentration of oxygen vacancies in CeO2 is needed to enhance th...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Oxygen Vacancy Enhanced Gas Sensing Performance of CeO/Graphene Heterostructure at Room Temperature 2

Li zhai Zhang, Qinglong Fang, Yuhong Huang, Kewei Xu, Paul K Chu, and Fei Ma Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b01768 • Publication Date (Web): 23 Jul 2018 Downloaded from http://pubs.acs.org on July 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Oxygen Vacancy Enhanced Gas-Sensing Performance of CeO2/Graphene Heterostructure at Room Temperature Lizhai Zhang a, Qinglong Fang a, Yuhong Huang b, Kewei Xu a,c,*, Paul K. Chu d,*, Fei Ma a,* a

State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, Shaanxi, China College of Physics and Information Technology, Shaanxi Normal University, Xi’an 710062, Shaanxi, China c Department of Physics and Opt-electronic Engineering, Xi’an University of Arts and Science, Xi’an 710065, Shaanxi, China d Department of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China b

ABSTRACT: Oxygen vacancies (Ov) as the active sites have significant influences on the gas sensing performance of metal oxides, and self-doping of Ce3+ in CeO2 might promote the formation of oxygen vacancies. In this work, hydrothermal process is adopted to fabricate the composites of graphene and CeO2 nanoparticles, and the influences of oxygen vacancies as well as Ce3+ ions on the sensing response to NO2 are studied. It is found that the sensitivity of the composites to NO2 increases gradually, as the proportion of Ce3+ relative to all the cerium ions is increased from 14.6% to 50.7%, but decreases after that value. First-principles calculations illustrate that CeO2 becomes metallic at the Ce3+ proportion of < 50.7%, the chemical potential of electrons on surface decreases and the Fermi level shifts upwards due to the existence of low-electronegativity Ce3+ ions, resulting in reduced Schottky barrier height (SBH) at the CeO2/Graphene interface, enhanced interfacial charge transfer and high gas sensing performance. However, deep energy level will be induced at the Ce3+ proportion of > 50.7%, and the Fermi level is pinned at the interface. As a result, the density of free electrons is reduced, leading to increased SBH and poor gas sensing response. It demonstrates that an appropriate concentration of oxygen vacancies in CeO2 is needed to enhance the gas sensing performance to NO2.

1. INTRODUCTION Rapid economic development during the past decades has dramatically increased the worldwide energy consumption. Most of the energy originates from combustion of fossil fuels, and the toxic gases are released into the atmosphere.[1] Nitrogen dioxide (NO2) is one of the harmful gases.[2] Short-term exposure to NO2 (≥10 ppm) can induce immediate distress, including edema, nose and throat irritation. If the concentration of NO2 is higher than 100 ppm, short-term exposure will lead to death by asphyxiation.[3] Furthermore, NO2 will cause acid rain, accelerate the formation of microscopic particles and the photochemical smog in the air.[3] Therefore, simple and cost-effective chemical sensors with excellent sensing performance to NO2 for health and environmental monitoring is greatly imperative. Metal oxides exhibit promising applications in the field of gas sensing due to low power consumption, high sensitivity, superior selectivity, low detection limit, rapid response and recovery rate, simplicity in fabrication, and high compatibility with microelectronic processing.[4-6] For example, WO3 nanograins,[7] ZnO nanowires,[8] SnO2 nanoribbons,[9] and In2O3 nanoparticles[10] exhibit good gas sensing performances. But some disadvantages limit their applications: (1) the conductivity of metal oxides is poor at room temperature;[11] (2) operation temperature is high (423.15-673.15 K), resulting in high power consumption and lowered stability and shortened life;[11] (3) agglomeration often takes place in the synthesis process, and thus the specific surface area certainly will be substantially reduced.[11] In recent years, it was found that the electrical

conductivity and dispersive feature could be well improved if metal oxide nanoparticles were deposited on conductive matrix, for example, carbon materials. Graphene has been widely explored for that purpose because of high carrier mobility, large specific surface area and inherently low electrical noise.[12-14] The composites of graphene and SnO2,[15] CuxO,[16] Fe2O3,[17] Co3O4,[18] and Cu2O[19] have been illustrated to deliver good sensing performance at room temperature. Ceria (CeO2) is one of the promising candidates for gas sensing because of the abundance and particularly the multi-valence characteristics.[20] Fu et al. synthesized the composites of CeO2 nanoparticles and graphene and illustrated the fast response (7.33 s) to 100 ppm NO2, but the sensitivity is only 10.39% at room temperature.[21] Therefore, it is mandatory to improve the gas sensing response. Recently, it was found that the gas sensing performance could be improved through controlling the surrounding facets. For example, the composites of graphene and CeO2 nanocubes with {100} facets exhibit high gas sensing performance, the sensitivity to 100 ppm NO2 at room temperature is 33%, which is enhanced by 200% as compared to the reported results (10.39%).[22] However, the {100} facets with high surface energy are usually replaced by low-energy {111} facets during growth. Therefore, how to enhance the gas sensing activity of CeO2 nanoparticles enclosed by the most stable {111} facets is crucial for the practical application. Nanostructured metal oxides with different morphologies could enhance the gas sensing performance, but the intrinsic electronic properties change little.[23] Doping might improve the electronic properties of metal oxides, but it

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 10

will deteriorate the crystalline stability due to the size discrepancy between the dopant and the host element.[23] As well known, oxygen vacancies in metal oxides affect the electronic states and physical properties substantially, and they commonly serve as the shallow donors to increase the carrier density and thus to improve the electrical conductivity, moreover, the crystal structure is still stable with oxygen vacancies.[23] For instance, Pang et al demonstrated that mesoporous tungsten oxide with oxygen vacancies exhibits superior gas sensing performance.[24] In this paper, hydrothermal process is adopted to fabricate the composites of graphene and CeO2 nanoparticles. H2O2 as the oxidant is adopted to reduce the proportion of Ce3+ ions in the composites, while L(+)-ascorbic acid (AA) as the reducing agent is used to increase the proportion of Ce3+ ions. Based on this, the influence of oxygen vacancies as well as Ce3+ ions on the sensing performance to NO2 is studied. It is found that the sensitivity of the composites to NO2 increases gradually, as the proportion of Ce3+ is increased from 14.6% to 50.7%, but decreases after that value. First-principles calculations are performed to study the electronic states and understand the mechanism for the enhanced gas sensing performance.

voltage (I-V) curves. The test parameters of Electron paramagnetic resonance, photoluminescence and ultraviolet photoelectron spectra are shown in the supporting information. 2.3. Gas sensing measurements. As shown in Figure S1, gold interdigitated electrodes with a finger width were used as the substrates. 5 mg of CeO2/graphene powder was uniformly dispersed in ethyl alcohol by ultrasonication. A few drops of the suspension were deposited onto the Au interdigitated electrode. After the ethanol was evaporated, the CeO2/Graphene powder was left on the electrode, and then heated to 333 K for 1 h to evaporate the solvent. The detail information is shown in the supporting information. 2.4. Theoretical model and calculations. Densityfunctional theory (DFT) calculation was conducted with the Vienna ab initio simulation package (VASP).[26-28] The kinetic energy cutoff for the plane wave basis was 500 eV. Brillouin zone integration on grids with 3×3×1 and 5×5×1 Monk horst−Pack k-points was implemented for geometrical optimization and calculation of density of states, respectively. The detail calculation is shown in the supporting information.

2. EXPERIMENTAL AND THEORETICAL SECTION

3.1. Morphological and structural characteristics.

2.1. Material Preparation. Graphene oxide (GO) was synthesized from pure graphite powders by the modified Hummers method [25] and graphene was prepared by reduction of GO with H2 at 673.15 K for 1 h. The CeO2/Graphene nanocomposites were synthesized via a hydrothermal method. In a typical process, 10 mg of graphene were dispersed in 30 mL mixture solvent consisting of triethylene glycol (TEG) and deionized water (H2O) with different volume ratios (TEG+H2O= 0 mL+ 30mL, 10 mL+20 mL, 15 mL+15 mL, 20 mL+10 mL, or 30 mL+0 mL) under sonication for 2 h. Then, 150 mg Ce(NO3)3·6H2O, 41.46 mg NaOH, and 0.3 g PVP were added into the solvent with continuous stirring for 3h. After that, the mixture was transferred to a 100 mL Teflonlined autoclave, kept at 473.15 K for 24 h, and then naturally cooled to room temperature. Afterwards, the precipitate was washed three times with distilled water and ethanol and dried at 343.15 K in a vacuum oven overnight to obtain the CeO2/graphene nanocomposite. The H2O2 and L(+)-ascorbic acid (AA) were adopted to control the oxygen vacancy of CeO2/Graphene nanocomposites. The detail process is shown in the supporting information. 2.2. Structural and electrical characterizations. X-ray diffraction (XRD) patterns were acquired between 10° and 90° (SHIMADZU XRD-7000S) with Cu Kα radiation. Raman spectra were measured on a Horiba HR800 spectrometer with a 633 nm laser as the excitation source. The morphology of the samples was characterized by scanning electron microscopy (SEM, FEI Quanta 600S) and high-resolution transmission electron microscopy (HR-TEM, JEOL JEM-2100Plus). The Mott-Schottky curves were measured on an electrochemical analyzer (CHI760E, Shanghai), the voltage was scanned from -0.4 V to 0.4 V versus SCE (saturated calomel reference electrode) at a frequency of 10,000 Hz and 0.5 M NaSO2 was used as the electrolyte. XPS spectra were measured on a Thermo Fisher ESCALAB Xi+. The Keithley 4200 Semiconductor Characterization System was used to obtain the current-

CeO2/Graphene composites were synthesized via hydrothermal reactions as shown in Figure S2. Figure S3 shows the XRD patterns of the as-prepared CeO2/Graphene composites. All the diffraction peaks can be indexed to the (111), (200), (220), (311), (222), (400), (331), (420) and (422) planes of CeO2 in a cubic fluorite-type structure [space group: Fm-3m (225), a=5.415 Å, JCPDS No. 2-1306]. No other peaks were observed in the XRD patterns, indicating high purity of the assynthesized samples. The peak intensity decreases gradually from samples (a) to (e) and the full width at half-maximum (FWHM) increases due to suppressed growth of CeO2 nanoparticles in TEG solution. According to Scherer’s equation, the average sizes of CeO2 nanoparticles in the composites are calculated to be about 20 nm, 19 nm, 16 nm, 13 nm, and 4 nm for samples (a) to (e), respectively. Figure S4 displays the Raman spectra of graphene[22], CeO2, and CeO2/graphene composites, the evident D and G bands confirm the presence of graphene.[29, 30] The G-band is attributed to the first-order scattering of the E2g mode,[31] while the D-band is associated with the structural defects or functional groups adsorbed on the carbon basal plane.[32] Compared with graphene, the G bands of the CeO2/Graphene composites exhibit red shift owing to p-type doping effect.[33] The intensity ratios of D to G bands (ID/IG) of the CeO2/Graphene composites are 1.148, 1.11, 1.157, 1.156 and 1.151, respectively, which is higher than that of graphene (0.97) as a result of partial modification by functional groups, formation of new sp2 clusters and high concentration of defects caused by increased vacancies, grain boundaries, amorphous carbon species, and CeO2 nanoparticles inserted between graphene sheets.[34] Moreover, the triply degenerate F2g mode of the fluorite CeO2, red shifts from 465 cm-1 to 452 cm-1 as a result of increased concentration of Ce3+ and Ov with increasing volume ratio of TEG and H2O.[35] According to the DFT calculation, the graphene sheet is quite flat upon structural optimization, indicating that the CeO2{111}/graphene interaction is indeed vdW rather than

3. RESULTS AND DISCUSSION

2 ACS Paragon Plus Environment

Page 3 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

covalent one, in accordance with reported results.[36, 37] Hence, the structure of graphene is indeed not changed. Figure S5 and Figure 1 display the SEM, TEM, HRTEM images and the fast Fourier transformation (FFT) patterns of the CeO2/Graphene composites. If deionized water (30 mL H2O) is used, irregular tetragonal CeO2 nanoparticles with two pyramidal tips and the size of about 20 nm are mainly formed (Figure S5a1,a2 and Figure 1a1,a2). As shown in the HRTEM image (Figure 1a2) and FFT (Figure 1a3), the interplanar spacing of the exposed facets is 0.32 nm corresponding to {111} facets of CeO2. When the volume ratio of TEG/H2O is increased to 10:20, the CeO2 nanoparticles evolve into irregularly elongated octahedral with an average size of 18 nm (Figure S5b1, b2 and Figure 1b1, b2). The HRTEM image (Figure 1b2) and FFT pattern (Figure 1b3) demonstrate that the middle parts of these elongated octahedra are bound by low-energy {111} facets. When the volume ratio of TEG/H2O is further adjusted to 15:15, the CeO2 nanoparticles become truncated octahedral morphology with an average size of 15 nm (Figure S5c1, c2 and Figure 1c1, c2). Regular hexagonal CeO2 nanocrystals emerge. The HRTEM image (Figure 1c2) and FFT patterns (Figure 1c3) indicate that the exposed lattice planes are predominately {111} facets with the interplanar spacing of 0.32 nm. If the volume ratio of TEG/H2O is increased to 20:10, the CeO2 nanocrystals evolve into irregular elongated octahedral with an average size of 12 nm (Figure S5d1, d2 and Figure 1d1, d2). The HRTEM image (Figure 1d2) and FFT patterns (Figure 1d3) indicate that the exposed facets are almost {111} planes and a very small portion of {100} facets at the edges and corners. If the TEG/H2O volume ratio is increased to 30:0, that is, only TEG in the solution, the composites of graphene and CeO2 quantum dos (QDs) 3-5 nm in size is formed (Figure S5e1, e2 and Figure 1e1, e2). As shown in the HR-TEM image (Figure 1e2) and SAED pattern (Figure 1e3), the fringe spacing is about 0.32, 0.27 and 0.19 nm, corre-

Figure 1. TEM, HRTEM images and FFT patterns of CeO2/graphene composites prepared with different volume ratio of TEG/H2O (a: 0:30, b: 10:20, c: 15:15, d: 20:10 and e: 30:0).

sponding to the (111), (200), and (220) crystal planes of cubic CeO2. The tiny nanoparticles have larger surface-to-volume ratio. The EDS maps show that C, O, and Ce atoms are homogeneously distributed throughout the sample e, confirming the formation of CeO2/graphene nanocomposites (Figure S6). Figure S7a shows the XPS spectra, in which the peaks of carbon, oxygen and ceria are clearly identified. Once Ce3+ ion appears in fluorite-structure ceria, oxygen vacancies will be produced to preserve charge neutrality through the transformation from Ce4+ to Ce3+ ions (2Ce4+ + O2- = 2Ce3+ + Vo+ 0.5O2, Vo is the abbreviation of O vacancy).[39-41] The higher the Ce3+ proportion is, the more oxygen vacancies are formed. Figure S7b-f display the spectra of Ce 3d in the CeO2/Graphene hybrid. The peaks at 885.9 eV and 904.5 eV can be ascribed to the Ce3+ ions, suggesting the presence of oxygen vacancies in CeO2.[39-41] The proportion of Ce3+ with respect to all the cerium ions in the CeO2/Graphene nanocomposites is calculated from the integrated peak area as following: [Ce 3+ ] =

ACe 3+ ACe 3+ + ACe 4 +

,

(4)

in which ACe3+ and ACe 4+ are the integrated area of the XPS peaks of Ce3+ ions and Ce4+ ions, respectively. Six binding energy peaks labeled as U1 , U 2 , U 3 , U′1 , U 2′ and U 3′ are attributed to Ce 3d5/2 and 3d3/2 in CeO2, while V and V ′ peaks are similar to Ce 3d5/2 and 3d3/2 for Ce3+ ions.[39-41] The proportion of Ce3+ ions with respect to all the cerium ions in the CeO2/Graphene composites increases with reducing size of CeO2 nanoparticles. Especially, as the size of CeO2 nanoparticles is reduced down to 3-5 nm, the proportion of Ce3+ is increased up to 32.45%, accompanied by increasing concentration of oxygen vacancies. As shown in Figure S8a-e, the XPS peaks at the binding energies of 529.8, 532.1, and 533.7 eV can be attributed to oxygen ions in the lattice of CeO2 (O0), the absorbed oxygen and the oxygen ions in the lattice of Ce2O3 (O1),[42-44], respectively. The proportion of oxygen ions in the lattice of Ce2O3 increases with reducing size of CeO2 nanoparticles, it is even up to 20.83% when the particle size is reduced down to 3-5 nm, which is consistent with the XPS spectra of Ce3+ ions. In essential, as the particle size is reduced, the fraction of surface atoms is increased and, the surface energy and surface tension increase resulting in more active surface atoms and more oxygen vacancies, accompanied by lattice strain.[45] Ce3+ is generated through the reduction of Ce4+ by the electrons left behind owing to oxygen vacancies. Figure S8f shows the peak of C. It can be deconvoluted into two subpeaks, the large peak at 284.6 eV is attributed to C-C bonding in the graphene scaffold, and the smaller peak at 286.0 eV is attributed to C-O bonding at the interface between CeO2 and graphene.[46] The formation of CeO2 nanocrystals from Ce3+ precursor proceeds in two stages: nucleation and ripening. In water, the nucleation process involves precipitation of Ce(OH)3 and transition into CeO2 nuclei 2-3 nm in size through oxidation and rapid dehydration. Then, CeO2 nanocrystals grow larger and larger through Ostwald ripening in the hydrothermal process. The ripening process might be suppressed if TEG is introduced into the solution. At the same time, a small amount of chelating Ce3+ alkoxide complexes are formed by the terminal hydroxyl groups in TEG. Hence, the size of CeO2 is substantially reduced. If only TEG is in the solution, a great num-

3 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ber of chelating Ce3+ alkoxide complexes instead of Ce(OH)3 will be generated, and then they are oxidized and dehydrated into Ce4+ alkoxide complexes. Finally, the Ce4+ alkoxide complex es are decomposed into CeO2 QDs (Figure S9).[47] 3.2. The sensing characteristics of CeO2/graphene devices. Figure S10a and b display the Mott-Schottky (MS) curves of both CeO2 nanoparticles and CeO2/graphene composites, they are approximately linear with a negative slope, indicative of p-type behavior. The carrier density can be calculated by the Mott-Schottky equation: 1 2 CSC

=

2  KT  ,  E − E fb −  eεε0 N  e 

(5)

in which Csc is the capacitance of the space charge region, e is the electronic charge, ε is the relative dielectric constant of the CeO2, ε is the permittivity in vacuum, N is the carrier 0

density, E is the applied potential, Efb is the flat band potential, K is the Boltzmann constant, and T is the absolute temperature.[48] The density of carriers is increased from 9.5×1018 cm-3 in the CeO2 nanoparticles to 4.939 × 1020 cm-3 in CeO2/Graphene nanocomposites by 5000%. The sensitivity of the samples to NO2, (Rgas-Rair)/Rair, is determined, in which Rgas and Rair denote the resistance of the sample exposed to NO2 and air, respectively, at room temperature. Figure 2a show the sensing response of the as-fabricated CeO2/graphene nanocomposites to 200 ppm, 100 ppm, 50 ppm, 25 ppm, and 10 ppm NO2. Apparently, the resistance decreases drastically upon exposure to NO2 and returns back to its initial value after exposure to air. This can be ascribed to the direct charge transfer between NO2 and CeO2/graphene composites. NO2 is an electron acceptor due to the unpaired electron on the nitrogen atom.[46] Since CeO2/graphene composites

Page 4 of 10

exhibit the P-type characteristics, the electron transfer from CeO2/graphene to NO2 increases the concentration of holes, resulting in decreased resistance. During subsequent exposure to air, the resistance will recover as a result of desorption of NO2. The response and recovery time to 10 ppm NO2 is about 181 s and 246 s, respectively (Figure 2b). The slow recovery at room temperature suggests the defect-dominated adsorption on CeO2 [46], which is further confirmed by DFT calculation. Figure 2c shows the sensitivity of pure CeO2 and CeO2/graphene composites as the concentration of NO2 varies from 200 ppm to 10 ppm. The response degree increases with gas concentration. The sensitivity of samples a-e to 100 ppm NO2 is about 16.83% , 19.24% , 23.03% , 24.67% and 32.7%, respectively, being 9.3, 10.69, 12.79, 13.7, and 18.17 times of pure CeO2 nanoparticles (1.8%). In particular, sample e (CeO2 QDs/graphene composites) delivers the best sensing performance. 10 ppm NO2 can be detected with a large resistance variation of 12.7%, which is indeed enhanced by 1170% as compared to pure CeO2 nanoparticles (1%). As shown in Figure 2d, the two cycling curves nearly coincide with each other, and a deviation of only 10% is observed after ten-day exposure to 5 ppm NO2, demonstrating excellent stability (Figure S10c). Selectivity is also an important parameter for gas sensors. The responses of CeO2/Graphene composites (Sample e) to 200 ppm Alcohol, Acetone, NH3, Toluene and Lsopropyl alcohol have been measured. The results are shown in Figure2e. It is clearly that the CeO2 /Graphene composites exhibits the highest sensitivity to NO2 among them, and thus an excellent selectivity toward NO2 gas at room temperature. In fact, a gas can be detected only when it can react with anion oxygen on the composites. The adsorption abilities and activities of target gases are different at a given temperature, which may lead to the selectivity. Moreover, oxygen vacancies and antisite defects can act as channel entrances for the small gas

(a)

(b)

(c)

(d)

(e)

(f)

200 ppm

Figure 2. (a) Response and recovery curves of CeO2/graphene-a composites with respect to 10-200 ppm NO2 at room temperature; (b) Response and recovery time of CeO2/graphene-a to 10 ppm NO2 gas at room temperature; (c) Sensitivity of pure CeO2, CeO2/graphene composites a-e for 10-200 ppm NO2 at room temperature; (d) The stability of CeO2/Graphene-e composites upon exposure to 5 ppm NO2 for two cycles; (e) Response of the CeO2/Graphene-e sensors in various gases, of which the concentration of Alcohol, Acetone, NH3, Toluene, and Lsopropyl alcohol is 200 ppm, and NO2 is 5 ppm; (f) The responses of the CeO2/Graphene-e sensors towards 5 ppm NO2 at different humidity.(G: graphene)

4 ACS Paragon Plus Environment

Page 5 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

molecules but not for the large gas ones, resulting in selective sensing.[11] Furthermore, the effects of environmental humidity on the gas sensing performances are also studied, and the results are shown in Figure 2f. The sensitivity to NO2 gas decreases rapidly as the atmosphere humidity is increased. In higher humidity, the water adsorbed on CeO2/Graphene composites acts as a barrier and inhibits the absorption of NO2 gas.[49] Furthermore, NO2 can react with water, forming HNO3, which will deteriorate the gas sensing response. Hence, the humidity has substantial effect on gas sensing performance. 3.3. The effects of Ov on the gas sensing performance. It is well accepted that self-doping of Ce3+ in CeO2 can promote the formation of stable oxygen vacancies owing to the small difference in ionic radius between Ce3+ (0.1283 nm) and Ce4+ (0.1098 nm).[45] The concentration of oxygen vacancies can be tuned by changing the concentration of Ce3+ ions. Hence, to further evaluate the influences of oxygen vacancy on the gas sensing performance, H2O2 and L(+)-ascorbic acid (AA) are adopted to change the proportion of Ce3+ ions with respect to all the cerium ions as well as the concentration of oxygen vacancies.[50-54] In fact, H2O2 can oxidize Ce3+ into Ce4+, while AA can reduce Ce4+ into Ce3+, the concentration of oxygen vacancies changes correspondingly. [50-54] 2Ce 3+ + H 2 O2 → 2Ce 4+ + 2OH −

(6)

2Ce 4+ + C6 H 8O6 → 2Ce3+ + C6 H 6O6 + 2H +

(7) Figure S11 shows the XRD patterns of the CeO2/graphene composites prepared with H2O2 or L(+)-ascorbic acid (AA), all the peaks can be indexed as the face-centered cubic phase [space group: Fm-3m (225)] of CeO2 with a cell constant of

(a)

(b)

a = 5.41 Å (JCPDS No. 65-2975). As the concentration of H2O2 is increased from 0 % to 1.7 %, the XRD peaks become higher and shaper owing to reduced defects and improved crystallization of CeO2 nanoparticles (Figure S11a). But the intensity of XRD peaks is reduced with increasing density of AA, indicating more defects (Figure S11b). Figure S12 displays the Raman spectra of the CeO2/Graphene composites, in which the D and G bands could be evidenced at 1348 cm-1 and 1595 cm-1.[29,30] The Raman peak of the triply degenerate F2g mode of the fluorite CeO2 is blue shifted with increasing concentration of H2O2 (Figure S12a), but red shifted with increasing concentration of AA (Figure S12b). This can be ascribed to the decreased or increased Ce3+ ions in the reaction environment of H2O2 and AA, respectively. The XPS spectra of Ce 3d in the CeO2/Graphene composites are displayed in Figures S13 and S14. Six binding energy peaks, U1 , U2 , U 3 , U′1 , U 2′ and U 3′ are attributed to Ce 3d5/2 and 3d3/2 of Ce4+, but V and V ′ peaks to Ce 3d5/2 and 3d3/2 of Ce3+ ions.[39-41] As shown in Figure S13, the proportion of Ce3+ ions decreases from 32.45% to 14.6% gradually as the concentration of H2O2 is increased from 0% to 1.7%. But the Ce3+ fraction is increased from 32.45% to 60.6% as the concentration of AA is increased from 0 g/L to 6.67 g/L (Figure S14). The XPS spectra of O 1s are also measured. the proportion of oxygen ions in the lattice of Ce2O3 (O1) decreases from 20.83% to 17.02% gradually as the concentration of H2O2 is increased from 0% to 1.7% (Figure S15). But the fraction of oxygen ions in the lattice of CeO2 (O0) decreases from 7.71% to 1.47% as the concentration of AA is increased from 0.167 g/L to 6.67 g/L (Figure S16), which is consistent with the outcome of Ce3+ ions. So the proportion of Ce3+ ions and the concentration of oxygen va-

g=~1.97

(c)

g=~2.002

(d)

(e)

(f)

Figure 3. (a) PL spectra of CeO2/Graphene composites prepared with 0.5 mL H2O2, 0 mL H2O2 (0 mg AA), 5 mg AA and 10 mg AA; (b) EPR spectra of CeO2/Graphene composites prepared with 0.5 mL H2O2, 0 mL H2O2 (0 mg AA), and 10 mg AA. (AA: Ascorbic acid); (c) The fraction of Ce3+ in the as-synthesized CeO2/graphene composites prepared with different concentration of H2O2 and AA, respectively; (d) The sensitivity of the as-prepared CeO2/graphene composites prepared with different fraction of Ce3+ for 10-200 ppm NO2 at room temperature; (e) Response and recovery curves of CeO2/graphene composites (Ce3+: 50.7%) to 3-10 ppm NO2 at room temperature, and (f) the response time of CeO2/graphene (Ce3+: 50.7%) to 3-200 ppm NO2 gas at room temperature.

5

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cancies can be changed through adding H2O2 and AA of different concentration into the reaction solution. (Figure 3c). In order to further confirm the presence of oxygen vacancy, the photoluminescence (PL) and Electron paramagnetic resonance (EPR) are measured. Figure 3a shows the PL spectra of CeO2/Graphene composites. The PL intensity is higher if the CeO2/Graphene composites with lower Ce3+ concentrations are prepared with H2O2 in the reaction solution, but it becomes lowered if the composites with higher Ce3+ concentrations are prepared with AA in the reaction solution. In fact, at higher concentrations of Ce3+ and oxygen vacancies, more valence electrons can be excited and trapped onto the defect levels, so that the recombination between electrons and holes might be suppressed, resulting in lowered PL intensity.[55] Figure 3b displays the measured EPR spectra at room temperature. A stronger EPR peak which can be assigned to the Ce3+ ions is observed at g= 1.97 if the composites are prepared with AA in the reaction solution. Moreover, the EPR signal at g=2.002 is also observed as a result of O2- reduced from the O2 moleculars in atmosphere by Ce3+ ions.[56] Terefore, both PL and EPR confirm the oxygen vacancies in CeO2/Graphene composites, which is tunable by changing the concentration of H2O2 and AA in the reaction solution (Figure 3c). Figure 3d plots the sensitivity to NO2 gas as a function of the proportion of Ce3+ ions. As the proportion of Ce3+ ions is increased from 14.6% to 50.7%, the sensitivity increases gradually, but then decreases dramatically after that value. Figure 3e shows the response and recovery curves of the CeO2/graphene composite with Ce3+ proportion of 50.7%. The slow recovery at room temperature suggests the defectdominated adsorption on CeO2[46], which is further confirmed by DFT calculation, and as displayed in Figure 3f, the response time is elongated from 53 s to 380 s as the concentration of NO2 is decreased from the 200 ppm to 3 ppm. The sensing performances to NO2 in the this work and the reported results are summarized in Table S1 for comparison.[57-60, 21] It is noteworthy that the as-fabricated composites in this work exhibit better sensing performances. Although the SnO2/RGO,

Page 6 of 10

WO3/RGO, and NiO/RGO composites exhibit higher sensitivity to NO2, the high operating temperature limits their practical applications.[57-59] This issue is overcome by the CeO2/Graphene composites, which is a promising candidate for detecting NO2 gas at room temperature. 3.4 Gas-sensing mechanism Graphene plays an important role in enhancing the gassensing performance. Firstly, it can inhibit the agglomeration and recrystallization of CeO2 nanoparticles (Figure S17b-d). During the synthetic process, the oxygen-containing groups on graphene are negatively charged, so that Ce3+ ions can be easily attached onto graphene through electrostatic attraction. Uniformly dispersed nucleation and growth of CeO2 nanoparticles take place on graphene via oxidation of Ce3+ ions and rapid dehydration.[61] Figure S17a shows the SEM image of CeO2 without graphene, in which agglomeration and recrystallization take place. The uniform distribution of CeO2 nanoparticles on the graphene sheets guarantees the good gas penetration and transport in the composites, enhancing the gas sensing performance;[62] Secondly, the CeO2 nanostructures exhibit good gas sensing performance only at higher temperature (200 ℃),[63] but graphene in the hybrid architectures improves the electrical conductivity of CeO2 significantly, and thus lower the operating temperature; Thirdly, graphene sheets have extremely large surface area, thus facilitating the access of gases onto the CeO2/Graphene composites;[12-14] Finally, the graphene/CeO2 interface can be adopted to modulate the charge transfer during gas-sensing process, which is further confirmed by the first-principles calculation. 3.5. The mechanism for the oxygen vacancy dependent gassensing performance. Figure S18a shows the I-V curves of CeO2/graphene composites with different concentration of Ce3+ ions, and the nonlinear characteristics indicate the Schottky contact. Based on

(a)

(b)

(c)

(d)

(e)

(f)

Figure 4. The DOS and PDOS (a, b) CeO2{111} facets, (c, d) CeO2 {111} facets containing one Ov and two Ce3+ ions (Ce:O=0.51), and (e, f) CeO2 {111} with Ce:O = 0.6, respectively. The dashed lines indicate the Fermi level.

6 ACS Paragon Plus Environment

Page 7 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the thermionic-emission theory, under a forward bias, the current passing through the Schottky barrier can be described as:[64]  qφ    qv  (8) I TE = SA∗T 2 exp  − SB  exp  − 1  ,  kT   kT   and

A∗ =

4πqm∗ k 2

, (9) h3 in which S is the area of the Schottky contact, A∗ is the effective Richardson constant, T is the temperature, q is the unit electronic charge, k is the Boltzmann constant, V is the applied voltage and φ SB is the Schottky barrier height. Under a reverse bias, the current can be described as:[64]  4 q 7 N (V + V − kT / q) / 8π 2ε 3  qφ   D bi s I TED = SA∗∗T 2 exp − SB  × exp kT   kT  

and

(

)

Vbi = φ SB − Ec − E f ,

 ,   

(10)

(11)

in which A∗∗ is the effective Richardson constant, ND is the donor impurity density, Vbi is the built-in potential, and ε s is the permittivity. The depletion layer formed at the Schokky barrier area has a width of:[64] WD =

2ε s qN D

 kT  ,  Vbi − V −  q  

(12)

According to Eqs. 8, 10 and 12, the current passing through the Schottky contact is very sensitive to the Schottky barrier height (SBH) and barrier width. At large SBH, the transfer of electrons between the graphene and metal oxides becomes extremely difficult, the adsorption gas induced change of electrical current is negligible, resulting in poor gas sensing performance.[64] In principle, the original SBH is mainly determined by the difference of the work function between metal oxides and graphene.[64] Based on DFT calculations, the work function of CeO2 (6.14 eV) is much larger than that of graphene (4.54 eV), large Schottky barrier height exists between graphene and CeO2, making it difficult for the electrons to transfer between them, and the system exhibits poor gas sensing performance (Figures S18b-c).[22] Since the electronegativity of Ce3+ ions is lower, the introduction of Ce3+ ions in CeO2 will lower the work function and thus the difference of work function between CeO2{111} facets and graphene decreases. Hence, the Schottky barrier height (SBH) is lower.[65] Furthermore, oxygen vacancy (Ov) will be produced owing to the existence of Ce3+ ions, and give rise to donor states. The ionization at donor states near the conduction band will increase the density of carriers in the conduction band, and give rise to an upward Fermi-level shift and electron tunneling.[65] The contact resistance and Schottky barrier height (SBH) decrease. As a result, it becomes easier for the electrons to travel across the interface between CeO2 and graphene, and thus the sensing performance of CeO2/Graphene composites is improved. Hence, the Ce3+ ions might decrease the Schottky barrier height of CeO2/Graphene heterojunction. To further reveal the dependence of sensitivity on Ce3+ ions and Ov, the work function, total and projected density of states of CeO2 and Ce3+-doped CeO2 are calculated by density functional theory (DFT). As shown in Figure S18(c-g), the work function decreases, and thus the Fermi level is up-shift if Ce3+

ions and oxygen vacancy coexist. Figures 4a and 4b show the total and projected density of states of CeO2 without oxygen vacancies (Ce:O = 0.5). The valence states are predominantly from the O 2p orbitals, while the conduction states originate from the Ce 4f orbitals. The Fermi level is located at the valence band maximum (VBM), characteristic of P-type feature. If Ce3+ ions and Ov coexist (Figure S19), the VBM and conduction band minimum (CBM) of CeO2{111} facets are mainly dominated by the O 2p and Ce 4f states, respectively (Figure 4c and 4d), and the Fermi level is up-shifted towards the CBM. Consequently, the surface chemical potential of CeO2 and the SBH of CeO2/graphene interface are lowered. But, if heavily doped with Ce3+ ions (Ce:O = 0.6), the donor states will be located at the mid of bandgap, leading to a deep level (Figure 4e and 4f). Owing to the larger ionization energy, the donor state on the deep level is generally not ionized and contributes little to the carriers. But it is the recombination center of electrons and holes, so the life of non-equilibrium minority carriers is shortened, the Fermi level is pinned, and the SBH of CeO2/Graphene interface is increased, resulting in poor gas sensing performance. To further study the influences of Ce3+ ions on Fermi level, ultraviolet photoelectron spectroscopy (UPS) are measured, and the results are shown in Figure S20. The work function is obtained through subtraction of the excited photon energy by the energy of the overflowed electrons, that is, the intercept of linear part on the curves. As the proportion of Ce3+ ions is increased from 47.71% up to 50.7%, the work function of CeO2 decreases from 3.6 eV to 2.7 eV firstly, and then increases up to 3.6 eV at the proportion of Ce3+ ions of 52.83%. So it can be deduced that the Fermi level is up-shifted at small Ce3+ proportion and down-shifted at large Ce3+ proportion, which is consistent with DFT calculation. At high Ce3+ ions concentrations, phase transition might take place, resulting in different bond length, coordination symmetry and electronic states. So the influences of high-proportion Ce3+ ions on Fermi level and work function become more complex.[65] Finally, the binding energies of NO2 at different adsorption sites on CeO2/graphene heterojunction is also calculated. The configurations of NO2 at different adsorption sites on CeO2/graphene heterojunction are shown in the Figure S21. As shown in Figure 5, all the Eads values are negative, indicat ing that NO2 molecules are stably absorbed. The absolute value of adsorption energy is large enough to withstand the thermal disturbance at room temperature (0.026 eV). Particularly, the absolute value of adsorption energy at oxygen va-

Figure 5. The binding energies of NO2 at different adsorption sites on CeO2/graphene heterojunction.

7 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cancies is higher than that at any other sites. Hence, oxygen vacancies are the best active sites for chemisorption of NO2. It is easier for NO2 to be adsorbed on the composites. This is also crucial to the improvement of gas sensing performance. 4. CONCLUSIONS In summary, hydrothermal process is adopted to fabricate the composites of graphene and CeO2 nanoparticles, of which the proportion of Ce3+ ions is changed through adding H2O2 and L(+)-ascorbic acid (AA) as the oxidant and reducing agent in the reaction solution. It is found that the gas sensing per formance of the composites to NO2 is highly dependent on the proportion of Ce3+ ions. The gas sensing performance is improved as the proportion of Ce3+ ions is increased gradually up to 50.7%, but decreases after that value. Based on the measured I-V curves, the Schottky barrier exists between CeO2 and graphene. However, the work function of CeO2 and the Schottky barrier height of CeO2/graphene heterojunction is lowered upon introduction of Ce3+ ions owing to the lower electronegativity, and thus the gas sensing performance is enhanced. This is also confirmed by the DFT calculations. But it is a different case for the composites with large proportion of Ce3+ ions, in which the deep level is produced and the Fermi level is pinned at the interface. As a result, the density of electrons decreases, and the SBH is increased, resulting in poor gas sensing response. The results present us a clear phys ical picture for the oxygen vacancy dependent gas sensing performances of CeO2/graphene composites.

ASSOCIATED CONTENT Supporting Information Experimental and DFT calculation details. XRD, Raman spectra, XPS, SEM and TEM of CeO2 and CeO2/graphene composites, respectively. Schematic mechanism for the formation of CeO2 at the TEG. Mott-Schottky plots. This material is available free of charge via the Internet at http://pubs.acs.org

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] (Fei Ma) *E-mail: [email protected] (Kewei Xu) *E-mail: [email protected] (Paul K. Chu)

Notes The authors declare no competing financial interest. ACKNOWLEDGMENT This work was jointly supported by National Natural Science Foundation of China (Grant No. 51771144, 51471130), Natural Science Foundation of Shaanxi Province (No. 2017JZ015), the fund of the State Key Laboratory of Solidification Processing in NWPU (SKLSP201708), City University of Hong Kong Applied Research Grant (ARG) No. 9667122 and the Instrument Analysis Center of Xi'an Jiao tong University.

REFERENCES (1) Nejat, P.; Jomehzadeh, F.; Taheri, M. M.; Gohari, M.; Abd. Majid, M. Z. Renewable Sustainable Energy Rev. 2015, 43, 843-862. (2) Hamra, G. B.; Laden, F.; Cohen, A. J.; Raaschou-Nielsen, O.; Brauer, M.; Loomis, D. Environ. Health Perspect. 2015, 123, 11071112.

Page 8 of 10

(3) Pohanish, R. Sittig’s Handbook of Toxic and Hazardous Chemicals and Carcinogens, 6th ed., William Andrew, New York, 2011. (4) Gu, F. B.; Nie, R.; Tian, Z. W.; Han, D. M.; Wang, Z. H. RSC Adv. 2015, 5, 99018–99022. (5) Zhang, J.; Song, P.; Li, J.; Yang, Z. X.; Wang, Q. Sens. Actuators B. 2017, 249, 458–466. (6) Cao, S. X.; Chen, H.; Han, T.; Zhao, C.; Peng, L. L. Mater. Lett. 2016, 180, 135–139. (7) Lee, D. S.; Han, S. D.; Huh, J. S.; Lee, D. D. Sens. Actuators B. 1999, 60, 57–63. (8) Ahn, M. W.; Park, K. S.; Heo, J. H.; Park, J. G.; Kim, D. W.; Choi, K. J.; Lee, J. H.; Hong, S. H. Appl. Phys. Lett. 2008, 93, 263103. (9) Maiti, A.; Rodriguez, J. A.; Law, M.; Kung, P.; McKinney, J. R.; Yang, P. D. Nano Lett. 2003, 3, 1025–1028. (10) Soulantica, K.; Erades, L.; Sauvan, M.; Senocq, F.; Maisonnat, A.; Chaudret, B. Adv. Funct. Mater. 2003, 13, 553–557. (11) Li, L.; He, S. J.; Liu, M. M.; Zhang, C. M.; Chen, W. Anal. Chem. 2015, 87, 1638−1645. (12) Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P.; Katsnelson, M. I.; Novoselov, K. S. Nat. Mater. 2007, 6, 652-655. (13) Wehling, T. O.; Novoselov, K. S.; Morozov, S. V.; Vdovin, E. E.; Katsnelson, M. I.; Geim, A. K.; Lichtenstein, A. I. Nano Lett. 2008, 8, 173-177. (14) Li, W. W.; Geng, X. M.; Guo, Y. F.; Rong, J. Z.; Gong, Y. P.; Wu, L. Q.; Zhang, X. M.; Li, P.; Xu, J. B.; Cheng, G. S.; Sun, M. T.; Liu, L. W. ACS Nano. 2011, 5, 6955-6961. (15) Mao, S.; Cui, S.; Lu, G.; Yu, K.; Wen, Z.; Chen, J. J. Mater. Chem. 2012, 22, 11009-11013. (16) Yang, Y.; Tian, C.; Wang, J.; Sun, L.; Shi, K.; Zhou, W.; Fu, H. Nanoscale, 2014, 6, 7369-7378. (17) Dong, Y. L.; Zhang, X. F.; Cheng, X. L.; Xu, Y. M.; Gao, S.; Zhao, H.; Huo, L. H. RSC Adv. 2014, 4, 57493-57500. (18) Chen, N.; Li, X. G.; Wang, X. Y.; Yu, J.; Wang, J.; Tang, Z. N; Akbar, S. A. Sens. Actuators B. 2013, 188, 902-908. (19) Deng, S. Z.; Tjoa, V.; Fan, H. M.; Tan, H. R.; Sayle, D. C.; Olivo, M.; Mhaisalkar, S.; Wei, J.; Sow, C. H. J. Am. Chem. Soc. 2012, 134, 4905-4917. (20) Wu, K.; Sun, L. D.; Yan, C. H. Adv. Energy Mater. 2016, 6, 1600501. (21) Yang, Y.; Tian, C. G.; Sun, L.; Lv, R. J.; Zhou, W.; Shi, K. Y.; Kan, K.; Wang, J. C.; Fu, H. G. J. Mater. Chem. A. 2013, 1, 1274212749. (22) Zhang, L. Z.; Fang, Q. L.; Huang, Y. H.; Xu, K. W.; Ma, F.; Chu, P. K. J. Mater. Chem. C. 2017, 5, 6973—6981. (23)Wang, G. M.; Ling, Y. C.; Li, Y. Nanoscale, 2012, 4, 6682. (24) Wang, D.; Sun, J. B.; Cao, X.; Zhu, Y. H.; Wang, Q. X.; Wang, G. C.; Han, Y.; Lu, G. Y.; Pang, G. S.; Feng, S. H. J. Mater. Chem. A. 2013, 1, 8653–8657. (25) Marcano, D. C.; Kosynkin, D. V.; Berlin, J. M.; Sinitskii, A.; Sun, Z. Z.; Slesarev, A.; Alemany, L. B.; Lu, W.; Tour, J. M. ACS Nano. 2010, 4, 4806-4814. (26) Kresse, G.; Furthmüller, J. Phys. Rev. B. 1996, 54, 1116911186. (27) Kresse, G.; Hafner, J. Phys. Rev. B. 1993, 48, 13115-13118. (28) Kres, G.; Joubert, D. Phys. Rev. B. 1999, 59, 1758-1775. (29) Kim, Y. A.; Fujisawa, K.; Muramatsu, H.; Hayashi, T.; Endo, M.; Fujimori, T.; Kaneko, K.; Terrones, M.; Behrends, J.; Eckmann, A.; Casiraghi, C.; Novoselov, K. S.; Saito, R.; Dresselhaus, M. S. ACS Nano. 2012, 6, 6293-6300. (30) Qu, J. Y.; Gao, F.; Zhou, Q.; Wan, Z. Y.; Hu, H.; Li, B. B.; Wan, W. B.; Wang, X. Z.; Qiu, J. S. Nanoscale. 2013, 5, 2999. (31) Guan, Q.; Cheng, J. L.; Wang, B.; Ni, W.; Gu, G. F.; Li, X. D.; Huang, L.; Yang, G. C.; Nie, F. D. ACS Appl. Mater. Interfaces. 2014, 6, 7626−7632. (32) Das, A. K.; Srivastav, M.; Layek, R. K.; Uddin, M. E.; Jung, D.; Kim, N. H.; Lee, J. H. J. Mater. Chem. A. 2014, 2, 1332−1340. (33) Xu, C. H.; Sun, J.; Gao, L. J. Mater. Chem. 2012, 22, 975−979.

8 ACS Paragon Plus Environment

Page 9 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(34) Gu, F. B.; Nie, R.; Han, D. M.; Wang, Z. H. Sens. Actuator B. 2015, 219, 94-99. (35) Kosacki, I.; Suzuki, T.; Petrovsky, V.; Anderson, H. U.; Colomban, P. Solid State Ionics. 2002, 149, 99-105. (36) Xu, L.; Huang, W. Q.; Wang, L. L.; Huang, G. F.; Peng, P. J. Phys. Chem. C. 2014, 118, 12972−12979. (37) Xu, P.; Tang, Q.; Zhou, Z. Nanotechnology, 2013, 24, 305401. (38) Liu, X. W.; Zhou, K. B.; Wang, L.; Wang, B. Y.; Li, Y. D. J. Am. Chem. Soc. 2009, 131, 3140–3141. (39) Tsunekawa, S.; Fukuda, T.; Kasuya, A. Surf. Sci. 2000, 457, L437–L440. (40) Mahammadunnisa, S.; Manoj Kumar Reddy, P.; Lingaiah, N.; Subrahmanyam, C. Catal. Sci. Technol. 2013, 3, 730–736. (41) Acharya, S. A.; Gaikwad, V. M.; D'Souza, S. W.; Barman, S. R. Solid State Ionics. 2014, 260, 21–29. (42) Vinodkumar, T.; Durgasri, D. N.; Reddy, B. M.; Alxneit, I.; Catal. Lett. 2014, 144, 2033. (43) Zou, H.; Dong, X.; Lin, W.; Appl. Surf. Sci. 2006, 253, 2893. (44) Ilieva, L.; Pantaleo, G.; Ivanov, I.; Venezia, A. M.; Andreeva, D.; Appl. Catal. B: Environ. 2006, 65, 101. (45) Choudhury, B.; Choudhury, A. Materials Chemistry and Physics. 2012, 131, 666–671. (46) Long, H.; Trochimczyk, A. H.; Pham, T.; Tang, Z.; Shi, T. L.; Zettl, A.; Carraro, C.; Worsley, M. A.; Maboudian, R. Adv. Funct. Mater. 2016, 26, 5158–5165. (47) Xin, Y.; Yang, X.; Jiang, P.; Zhang, Z. L.; Wang, Z. P.; Zhang, Y. H. ChemCatChem. 2011, 3, 1772 – 1778. (48) Zhang, W. W.; Guo, H. L.; Sun, H. Q.; Zeng, R. C. Applied Surface Science. 2016, 382, 128-134. (49) Wang, C. ; Yin, L.; Zhang, L.; Xiang, D.; Gao, R. Sensors. 2010, 10, 2088-2106. (50) Perez, J. M.; Asati, A.; Nath, S.; Kaittanis, C. Small. 2008, 4, 552–556. (51) Kitsuda, M.; Fujihara, S. J. Phys. Chem. C. 2011, 115, 8808– 8815. (52) Chen, G. Z.; Zhao, H. G.; Rosei, F.; Ma, D. L. J. Phys. Chem. C. 2013, 117, 10031–11038. (53) Lv, C. J.; Di, W. H.; Liu, Z. H.; Zheng, K. Z.; Qin, W. P. Analyst. 2014, 139, 4547–4555. (54) Gao, W.; Li, J.; Zhou, X. M.; Zhang, Z. Y.; Ma, Y. Y.; Qu, Y. Q. J. Mater. Chem. C. 2014, 2, 8729–8735. (55) Shehata, N.; Meehan, K.; Hudait, M.; Jain, N. J. Nanopart. Res., 2012, 14, 1173–1183. (56) Wang, L.; Luan, Q. F.; Yang, D.; Yao, X.; Zhou, K. B. RSC Adv., 2013, 3, 6339–6342. (57) Srivastava, S.; Jain, K.; Singh, V. N.; Singh, S.; Vijayan, N.; Dilawar, N.; Gupta, G.; Senguttuvan, T. D. Nanotechnology. 2012, 23 205501. (58) Hoa, L.T.; Tien, H. N.; Luan, V. H.; Chung, J. S.; Hur, S. H. Sens. Actuators B. 2013, 185, 701–705. (59) Zhang, H.; Feng, J.; Fei, T.; Liu, S.; Zhang, T. Sens. Actuators B. 2014, 190, 472–478. (60) Cho, B.; Yoon, J.; Hahm, M. G.; Kim, D. H.; Kim, A. R.; Kahng, Y. H.; Park, S. W.; Lee, Y. J.; Park, S. G.; Kwon, J. D.; Kim, C. S.; Song, M.; Jeong, Y.; Nam, K. S.; Ko, H. C. J. Mater. Chem. C, 2014, 2, 5280−5285. (61) Tricoli, A.; Righettoni, M.; Teleki, A. Angew. Chem., Int. Ed. 2010, 49, 7632−7659. (62) Li, S. J.; Bao, D.; Shi, M. M.; Wulan, B. R.; Yan, J. M.; Jiang, Q. Adv. Mater. 2017, 29, 1700001-1700006. (63) Novoselo, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science, 2004, 306, 666. (64) Monch, W. Electronic Properties of Semiconductor Interfaces, Springer-Verlag, New York, 2004. (65) Greiner, M. T.; Chai, L.; Helander, M. G.; Tang, W. M.; Lu, Z. H. Adv. Funct. Mater. 2012, 22, 4557-4568.

9 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 10

The graph of the Table of Contents (TOC)

Ce

3+

H2O2

Ascorbic acid

Ce

4+

10 ACS Paragon Plus Environment