Facile Formation of Metal–Oxide Nanocraters by Laser Irradiation for Highly Enhanced Detection of Volatile Organic Compounds

Although various fabrication methods for metal–oxide nanostructures have been well developed for enlarged surface area, numerous efforts to further enhance the effective surface area for their chemical sensor applications are still being studied. Herein, a high‐power laser is irradiated on the existing metal–oxide nanostructures to expose the hidden inner surface of the nanostructures for full participation in the surface gas‐sensing reactions, resulting in extraordinary gas‐sensing performance. In addition, noble metal catalyst decoration at both the inner and outer surfaces of the nanostructures records extremely high gas response and selectivity to volatile organic compounds. The numerical simulation and experimental verification of the effects of high‐power laser irradiation for morphological evolution of the metal–oxide nanostructures can provide a new perspective toward the time‐efficient development of nanostructure‐based electronic devices.


Introduction
To realize seamless interaction between ambient environment and human activities in the internet of everything era, various electronic devices have been developed based on metal-oxide thin films.The metal-oxide thin films have been required to be morphologically engineered into highly effective nanostructures for such electronic device applications.Especially, chemoresistive gas sensors based on metal-oxides have increasing demands for effective nanostructures with enlarged surface area to facilitate more chemical reactions at the surface. [1]Depending on Although various fabrication methods for metal-oxide nanostructures have been well developed for enlarged surface area, numerous efforts to further enhance the effective surface area for their chemical sensor applications are still being studied.Herein, a high-power laser is irradiated on the existing metal-oxide nanostructures to expose the hidden inner surface of the nanostructures for full participation in the surface gas-sensing reactions, resulting in extraordinary gas-sensing performance.In addition, noble metal catalyst decoration at both the inner and outer surfaces of the nanostructures records extremely high gas response and selectivity to volatile organic compounds.The numerical simulation and experimental verification of the effects of high-power laser irradiation for morphological evolution of the metal-oxide nanostructures can provide a new perspective toward the time-efficient development of nanostructure-based electronic devices.
how the nanostructures are prepared and fabricated, various subsequent modifications like catalyst loading may be done to further improve surface activity.Therefore, precise design and control of metal-oxide nanostructures are the important keys to high-performance chemoresistive gas sensors.
In the last decades, numerous methods for the fabrication of metal-oxide nanostructures have been developed.Physical vapor deposition including e-beam evaporation, radio-frequency sputter, or pulsed laser deposition with control of deposition angle, [2] substrate rotation, [3] oxygen partial pressure, [4] or temperature [5] can yield various porous nanostructures.The solution-based methods including hydrothermal, [6] electrodeposition, [7] or sol-gel methods [8] can produce nanostructures through electrical and thermal controls or precursor compositions that can decide preferential crystallographic growth direction. [9]Combining precursor solution with polymers can result in fibrous nanostructures, [10] and anodization of metal film in ethylene glycol under high voltage is studied for nanotubular structures. [11]ompared to the previous bottom-up methods, the photolithography and etching process can be utilized for top-down fabrication of the nanostructures. [12]Recent studies reported phase mask-induced 3D nanopatterning of photoresists as a template for highly periodic porous structures. [13]The such templatebased procedures can easily build complicated 3D nanostructures.Especially, polystyrene (PS) microsphere is a popular template for the fabrication of dome-shaped or inverse opal nanostructures. [14]Such hollow nanostructures possess extra inner surfaces, which has a good utility function for facilitated electrical modulation, but so far it has not been successful in fully utilizing them. [15]Despite that numerous approaches have been successful in enlarging the surface area, new opportunities can be made if unexposed hidden surfaces inside the nanostructures are exposed for additional chemical-sensing reactions. [16]In this aspect, high-power laser irradiation on the existing nanostructures can easily tune the morphologies and provide a promising approach to achieve such utilization of hidden surface area.Also, the laser irradiation method can be highly beneficial in terms of processing time due to the rapid transformation of nanostructures, being a highly cost-effective and time-saving process. [17]o far, laser irradiation has been widely utilized for metal-oxide particle synthesis [18] or selective area hydrothermal synthesis [19] mostly through localized heating effect, but only a few reports focused on its high power that is capable of a morphological reshaping of the existing metal-oxide thin films. [20]High-power laser irradiation with enough thermal energy can melt down and deform existing metal-oxide films but excessively destructive power and random controllability have limited its transition into the desired morphology.Although Wang et al. successfully reformed the sharp top of ZnO nanorods into hollow nanospheres through high-power laser irradiation, [20b] a high potential of high-power laser irradiation for hidden surface exposure and application to effective chemical sensing should further be explored.
In this study, KrF excimer laser (248 nm) irradiation on SnO 2 nanodomes resulted in the exposure of the hidden inner surface of SnO 2 nanodomes to participate in gas-sensing chemical reactions.Through precise control of the laser parameters, highly reproducible nanocrater-like SnO 2 nanostructures (nc-SnO 2 ) could be achieved.Dome-shaped nanostructures provided natural gradation of laser intensities along the spherical surface for selective melting of the top surface.Just a single laser pulse irradiation on SnO 2 nanodomes resulted in two times the higher gas response to ethanol (C 2 H 5 OH).Furthermore, Au catalyst decoration at the inner and outer surfaces of SnO 2 nanostructures through the methods already developed by the authors [21] exhibited extraordinary gas sensitivity and selectivity toward volatile organic compounds.Our numerical simulation and experimental demonstration of superior gas-sensing performance achieved by nc-SnO 2 can suggest a versatile platform for the development of electronic devices based on metal-oxide nanostructures.

Results and Discussion
In Figure 1a, schematic illustrations of the fabrication procedure are presented.The PS monolayer is spin-coated on O 2 -plasma-treated substrates with Pt-interdigitated electrodes (IDEs).For samples without Au catalysts, only SnO 2 was deposited on the PS monolayer using an e-beam evaporator to yield SnO 2 nanodomes.For samples with Au catalysts, Au was first deposited on a PS monolayer, followed by SnO 2 and another Au deposition using an e-beam evaporator to be both sides Au-decorated SnO 2 nanodomes.After e-beam evaporation, the samples were placed under a KrF excimer laser irradiation system equipped with a homogenizer which allows uniform laser intensity within the laser spot area.With precisely controlled laser intensities, SnO 2 nanodomes were irradiated with a single laser pulse (25 ns width) to result in nanocrater-like SnO 2 nanostructures (nc-SnO 2 ) after thermal annealing.Figure 1b shows schematic drawings for surface evolution from SnO 2 nanodomes to nc-SnO 2 under irradiation of different laser intensities.As laser intensity increases, the top surface of SnO 2 nanodomes starts to get cracked and finally gets ablated under further higher laser intensity.The detailed fabrication processes are available in Experimental Section.
To investigate heat-transfer distribution on SnO 2 nanodomes and resulting morphological changes induced by laser irradiation, numerical simulation using COMSOL Multiphysics was conducted.As shown in Figure S1, Supporting Information, 3D models for amorphous SnO 2 nanodomes were constructed on SiO 2 substrate with a boundary heat source at the top of nanodome.Using a time-dependent heat-transfer module, the heat gradient on amorphous SnO 2 nanodomes was simulated as shown in Figure 2a.As clearly indicated in red color, the top center region of SnO 2 nanodomes experiences the highest temperature under simulated laser irradiation with an intensity of 110 mJ cm À2 .The 2D temperature profile mapping through the A-B intersection indicated in Figure 2a is shown in Figure 2b.In simulated temperature distribution, the top center of SnO 2 nanodomes (point 1) reaches a temperature as high as 2747 K, and the temperature decreases for the points away from the center.Since the model of as-deposited SnO 2 is amorphous, there is no definite melting point value due to the random arrangement of the atomic lattices but glass-transition temperature which is generally lower than the melting point of crystalline materials. [22]The literature value of the melting point of bulk crystalline SnO 2 is 1903 K. [23] Considering the extremely thin thickness of SnO 2 in the model, a further lower melting point can be expected.Therefore, somewhere between points 4 and 5 in Figure 2b has the highest chance of amorphous SnO 2 getting start to melt upon laser irradiation (Figure 2c).The SnO 2 surfaces lower than point 5 experience not enough heat transfer for morphological changes.In other words, the nanodome shape provides a natural laser intensity gradient along the spherical surface of the SnO 2 nanodomes suitable for top surface selective ablation.In Figure 2d, the maximum temperature elevations upon different laser intensities (50, 70, 90, 110, and 130 mJ cm À2 ) are calculated.The SnO 2 melting is calculated to happen under laser intensity higher than 70 mJ cm À2 and should have optimal laser intensity for desired morphological changes before it becomes too destructive, which needs to be experimentally verified.
The top-view and cross-sectional-view scanning electron microscopy (SEM) images of SnO 2 nanodomes and their evolution into nc-SnO 2 under different laser intensities (0, 70, 90, and 110 mJ cm À2 ) are shown in Figure 3a-h.As illustrated in schematics of Figure 1b, the top surface of SnO 2 nanodomes exhibits a little crack under irradiation of laser with 70 mJ cm À2 , indicating a little melting of SnO 2 .As laser intensity increases, cracks at the top surface get larger and at an intensity of 110 mJ cm À2 , the top surface of SnO 2 nanodomes is completely ablated, and the inner surface of SnO 2 nanodomes is fully exposed to the atmosphere.In line with simulation data, the edge side of SnO 2 nanodomes remained almost undamaged due to negligible heat transfer from laser irradiation.The hollow volume inside the nanostructures was formed after burning PS microspheres during thermal annealing of the laser-irradiated SnO 2 nanodomes.As a result, nanocrater-like nanostructures could be fabricated.Also, a little increase in overall nanostructure heights could be observed due to the resulting particulate products from the laser irradiation being attached to the top of the nanostructures.Figure 3i-k shows top-view SEM images of the fabricated nc-SnO 2 on IDEs.The nc-SnO 2 is uniformly distributed in between each electrode finger with 5 μm spacing.Also, the fabrication procedures of nc-SnO 2 were compatible with other metal-oxides including ZnO, WO 3 , NiO, and Fe 2 O 3 as shown in Figure S2, Supporting Information, indicating high uniformity, reproducibility, and universal versatility of the fabrication procedures.In Figure 4a, cross-sectional transmission electron microscopy (TEM) images of both sides Au-decorated nc-SnO 2 irradiated.The uniformly decorated Au catalysts at both the inner and outer surfaces of nc-SnO 2 can be found, which can also be clearly indicated in SEM images in Figure S3, Supporting Information.Figure 4b,c shows high-resolution TEM (HRTEM) images of nc-SnO 2 at the top region and bottom region, respectively.A fringe interval of 0.3347 nm corresponds to a d-spacing of (110) for SnO 2 and that of 0.2642 nm corresponds to a    intensities decreased due to less signal from the destroyed top surface of the nanostructures.Interestingly, peak sharpness or full width at half maximum of SnO 2 peaks became smaller upon stronger laser irradiation.This can be attributed to instant crystallization that occurs during the instant melting and solidifying process of laser irradiation.Although all samples were treated at 550 °C after laser irradiation for the crystallization, laser-induced crystallization effects also contributed and yielded even higher crystallinity.
The fabricated gas sensors based on nc-SnO 2 under irradiation of different laser intensities were exposed to 50 ppm C 2 H 5 OH at 300 °C to evaluate their gas-sensing properties as shown in Figure 5a.21a] Compared to SnO 2 plain film, SnO 2 nanodomes (without laser irradiation) and nc-SnO 2 exhibited higher base resistance due to limited electrical conduction path in between each unit nanostructure.The difference between base resistance values of nc-SnO 2 under irradiation of different laser intensities was not significant.It can be attributed to almost undamaged edge regions of the nanostructures, which is the current-necking region that determines overall base resistance.The gas response, which is defined as the ratio between saturated resistance value under air exposure (R a ) and target gas exposure (R g ), can be calculated ([R a ÀR g ]/R g ) from the gas-sensing curves of Figure 5a.Among the tested sample conditions, nc-SnO 2 under irradiation of 90 and 110 mJ cm À2 exhibited the highest gas responses of 1077 and 1030, respectively, toward 50 ppm C 2 H 5 OH.Although the gas response of nc-SnO 2 under irradiation of 90 mJ cm À2 is slightly higher than that of 110 mJ cm À2 , the background noise of nc-SnO 2 under irradiation of 90 mJ cm À2 is higher (δ = 16.98)than that of 110 mJ cm À2 (δ = 11.19) as shown in Figure S6, Supporting Information.This can be attributed to the partially opened top surface with many cracks and ruptured particles for nc-SnO 2 under 90 mJ cm À2 as shown in Figure 3e,f, while nc-SnO 2 under 110 mJ cm À2 exhibits a clearly opened top surface with fewer cracks and particulates as shown in Figure 3g,h.The signal-to-noise ratio, therefore, is calculated as 63.44 and 92.05 for nc-SnO 2 under 90 and 110 mJ cm À2 , respectively.Therefore, we decided to use nc-SnO 2 undergone laser irradiation of 110 mJ cm À2 for further gas-sensor measurements.The response time, which is defined as the time required to reach the 90% of the difference between R a and R g , is another important parameter for the evaluation of gas-sensing properties.As shown in Figure 5b, all tested gas sensors exhibited an extremely fast response time of 2 s.The fast response time can be attributed to the natural surface characteristics of SnO 2 .No significant difference in response time between SnO 2 plain film and SnO 2 nanostructures (SnO 2 nanodomes without laser irradiation and nc-SnO 2 with laser irradiation) indicates that morphological evolution was the only change happened during laser irradiation without affecting surface characteristics.For example, broken stoichiometry during laser irradiation might have increased the number of surface defects, and resulting highly active sites could have contributed to changes in response time.However, the post-annealing process had reconstructed broken stoichiometry to original SnO 2 crystallographic lattices, taking only morphological evolution into account for the contribution to gas-sensing performance.Figure 5c shows the gas-sensing curves of both sides Au-decorated nc-SnO 2 irradiated with a laser intensity of 110 mJ cm À2 to 50 ppm C 2 H 5 OH at 300 °C.The base resistance further increased compared to nc-SnO 2 without Au decoration due to the formation of depletion at the interface between SnO 2 and Au.The gas response after Au decoration to 50 ppm C 2 H 5 OH further increased to 3613 due to the catalytic effect.The gas response values of the tested samples are summarized in Figure 5d.
To further investigate gas-sensing properties of the fabricated nc-SnO 2 , gas selectivity toward various gases (C 2 H 5 OH, C 7 H 8 , CH 3 COCH 3 , C 6 H 6 , CH 3 SH, and CO) was evaluated as shown in Figure S6, Supporting Information.The calculated gas responses from gas-sensing curves of SnO 2 nanodomes (without laser irradiation) and Au-decorated nc-SnO 2 as shown in Figure S7, Supporting Information, are summarized in Figure 6a,b.While selective behaviors toward C 2 H 5 OH and CH 3 COCH 3 of SnO 2 nanodomes were kept even after Au decoration, gas response toward CH 3 SH increased dramatically after Au decoration.For a clear comparison, the ratio between gas responses of SnO 2 nanodomes (S a ) and Au-decorated nc-SnO 2 (S b ) was calculated (S b /S a ) and plotted as shown in Figure 6c.The calculated response ratio toward CH 3 SH was the highest among other gases, indicating extremely selective detection of CH 3 SH achieved.Such CH 3 SH selectivity can be utilized for air-quality monitoring, [24] halitosis monitoring, [25] or detection of biomarkers. [26]In Figure 7a-c, Au-decorated nc-SnO 2 was exposed to 1, 2.5, 5, and 10 ppm of C 2 H 5 OH, CH 3 COCH 3 , and CH 3 SH, respectively, to evaluate linearity between gas concentration and gas response.The linear relationship is very important considering the practical application of the gas sensors for the evaluation of target gas with random concentration.The resulting gas responses in relation to gas concentrations  are plotted in Figure 7d.A clear linear relationship for all testes gases can be found and the slope of the linear fitting curve for CH 3 SH was higher than the other two gases, C 2 H 5 OH and CH 3 COCH 3 , indicating Au-decorated nc-SnO 2 exhibited its highest sensitivity toward CH 3 SH.Moreover, the theoretical detection limit can be calculated by extrapolating linear fitting curves to the lowest gas concentration where the signal-to-noise ratio is 3. [27] The theoretical detection limit of Au-decorated nc-SnO 2 to C 2 H 5 OH, CH 3 COCH 3 , and CH 3 SH was calculated to be 1.8, 1.6, and 0.037 ppt, respectively.Compared to previously reported studies of chemoresistive gas sensors based on metaloxides for various mercaptan detection, Au-decorated nc-SnO 2 in this study marked the highest gas response to CH 3 SH to the best of the author's knowledge as summarized in Table S1, Supporting Information. [28]hemisorption of O 2 molecules on the sensor surface at the operating temperature of 300 °C is a fundamental process for chemical gas-sensing reactions with target gas molecules.The O 2 molecules are ionized at 300 °C and adsorbed at the surface of SnO 2 .In this process, electrons are transferred from n-type SnO 2 to adsorbed oxygen molecules, forming an electron depletion region at the SnO 2 surface.Since SnO 2 nanodomes are serially connected, electronically depleted inter-grain contact between adjacent nanodomes forms double Schottky barriers as shown in Figure 8a. [15]The height of the double Schottky barriers can be modulated upon exposure to target gas molecules, enabling chemoresistive gas responses of SnO 2 nanostructures.For nc-SnO 2 , the inner surface of the nanostructures is exposed for additional O 2 adsorption and resulting in a larger depletion region at the inter-grain contacts.This further increases the height of the double Schottky barrier, enabling a more sensitive chemoresistive gas response to target gas molecules (Figure 8b).When Au catalysts are decorated on both the inner and outer surfaces of nc-SnO 2 , additional electrons are transferred to the Au catalysts and the depletion region gets further larger.This elevates the height of the double Schottky barrier and further sensitive modulation upon gas molecule adsorption and desorption can happen (Figure 8c).The enlarged depletion region for each morphology evolution can also be noticed in the base resistance values in Figure 5a,c.The SnO 2 nanostructures with a larger depletion region exhibited higher base resistances.At the same time, the catalytic effects of Au also promote overall gas-sensing response through the spillover effect where target gas molecules are first adsorbed on the Au catalysts and migrate toward the SnO 2 surface.The exceptional CH 3 SH selectivity of nc-SnO 2 can be attributed to the strong selective interaction between Au and sulfur compounds as reported by previous studies. [29]Thus, the excellent response and selectivity of Au-decorated nc-SnO 2 can be understood by the unique nanocrater-shaped nanostructure and the catalytic effect of Au catalysts.Figure 8d-f summarizes the aforementioned O 2 molecule accessibility to the nanostructures and the resulting thickness of depletion layers before and after Au catalysts loading.

Conclusion
The high-power excimer laser was used to selectively ablate the top surface of the SnO 2 nanodome taking advantage of natural laser intensity gradation on spherical surface morphologies.The numerical simulation revealed that a critical temperature for selective ablation at the desired location can be controlled through a proper choice of laser intensity.The laser irradiation with an extremely short processing time resulted in the exposure of the hidden inner surface of the SnO 2 nanodomes, forming nanocrater-like nanostructures.The participation of both inner and outer surfaces in the gas-sensing reaction could already achieve high-performance gas sensing, but subsequent Au catalysts loading at both surfaces demonstrated extraordinary gas response and selectivity toward target gas molecules.The fabrication method developed in this study could be done in an extremely short processing time and was proved to be compatible also with other metal-oxides showing potential for a versatile and universal cost-effective fabrication process.This study is strongly believed to provide good insight into the development of electronic devices based on metal-oxide nanostructures.

Experimental Section
Fabrication: The IDEs were fabricated using conventional photolithography and e-beam evaporation.After cleaning the Si wafer with acetone and isopropanol, hexamethyldisilane and photoresist (AZ5214, MicroChemicals) were spin-coated on Si wafer at 3500 rpm for 40 s.After baking at 110 °C for 1 min, the photolithography of the IDEs pattern was done using Mask Aligner (MA-6, Karl Suss).After baking the patterned Si wafer at 110 °C for 2 min, the photoresist was image reversed under flood exposure using the Mask Aligner.Then, the patterned Si wafer was developed (AZ 300MIF, MicroChemicals) and completely rinsed with deionized water.After photolithography, the wafer was loaded into an e-beam evaporator (EL-5, ULVAC) for the deposition of Pt/Ti (100 nm/30 nm).The subsequent liftoff process finalized the fabrication of Pt-IDEs for gas-sensor substrates.The fabricated Pt-IDEs were treated with O 2 plasma and PS microsphere solution (Polysciences) was spincoated.The samples were then completely dried at 50 °C.The prepared PS-coated samples were loaded into an e-beam evaporator (Korea Vacuum) for the deposition of SnO 2 (Kojundo Chemistry).The as-deposited SnO 2 nanodome samples were placed in the excimer laser irradiation system (COMPexPro 201, Coherent) equipped with a homogenizer and homebuilt optic systems.The laser intensities of 50, 70, 90, and 110 mJ cm À2 were used and the laser pulse width was 25 ns.After irradiating a single pulse of laser on the as-deposited SnO 2 nanodome samples, the samples were annealed at 550 °C for 1 h for the crystallization of SnO 2 and complete burning out of PS microspheres.
Characterization: The top-view and cross-sectional morphologies were analyzed using field-emission SEM (FE-SEM, Inspect F, FEI).Further detailed cross-sectional morphologies, crystallographic information, and atomic composition data were obtained using TEM (TitanTM80-300, FEI) equipped with EDS detector.Additional crystallographic information was acquired using an (XRD, ATX-G, Rigaku).The COMSOL Multiphysics software was used for the simulation of heat-transfer distribution on the nanostructures.
Gas-Sensing Measurement: The gas-sensing characteristics were evaluated using a homebuilt gas-mixing system.The gas sensors were located on the sample holder with Pt wires for electrical contact.After forming electrical contact with silver paste, the sample holder was placed inside the quartz tube surrounded by an external heater (Lindberg/Blue M Mini-Mite Tube Furnaces, Thermo Fisher Scientific).After elevating the furnace temperature to 300 °C, the synthetic air (N 2 80% and O 2 20%, Sinjin Gas) and target gas (balanced with N 2 , Sinjin Gas) were injected in sequential order using the automated mass flow control system.The flow rate of the gases was fixed at 1000 sccm and the concentration of the tested gases was all 50 ppm except CO which was 100 ppm.

Figure 1 .
Figure 1.a) Schematic illustrations of fabrication procedure for the laser-irradiated nanocrater-like SnO 2 nanostructures (nc-SnO 2 ).b) Schematic illustrations of surface evolution from SnO 2 nanodomes to nc-SnO 2 as a function of irradiated laser intensity.

Figure 2 .
Figure 2. Simulated heat-transfer distribution of the amorphous SnO 2 nanodomes during 21.3 ns with a laser intensity of the 110 mJ cm À2 using COMSOL Multiphysics.a) The 3D mapping of temperature distribution on amorphous SnO 2 nanodomes and b) cross-sectional 2D mapping at the A-B intersection indicated in (a).c) Simulated temperature profiles at specific points (indicated in (b)) on the SnO 2 nanostructure as a function of time.d) Maximum temperatures under laser irradiation with various laser intensities (50, 70, 90, 110, and 130 mJ cm À2 ).Inset numbers indicate the maximum temperature at each laser intensity.

Figure 3 .
Figure 3. Scanning electron microscopy (SEM) characterization of the SnO 2 nanostructures and their morphological evolution as a function of laser intensity: a,b) before laser irradiation, c,d) 70 mJ cm À2 , e,f ) 90 mJ cm À2 , and g,h) 110 mJ cm À2 .i) An SEM micrograph of a Pt-interdigitated electrodes (IDEs) substrate with 5 μm spacing for gas-sensing measurement.j) Top-view SEM micrograph of the laser-irradiated nc-SnO 2 between Pt IDEs.k) Magnified SEM micrograph of the laser-irradiated nc-SnO 2 in the selected area.
d-spacing of (101) for SnO 2 .The stoichiometry of SnO 2 could have been destroyed when under laser irradiation, but since we annealed nc-SnO 2 at high temperature to remove residual PS, broken stoichiometry could recover and nc-SnO 2 was recrystallized to exhibit no clear difference between the top and bottom regions of nc-SnO 2 .Figure4d,eshows HRTEM images of Au catalysts at the inner and outer surfaces of nc-SnO 2 , respectively.A fringe interval of 0.2033 nm corresponds to a d-spacing of (200) for Au and that of 0.2346 nm corresponds to a d-spacing of (111) for Au.Along with the HRTEM of Au, energy-dispersive X-ray spectroscopy (EDS) mapping for Sn, O, and Au shown in Figure 4f-h also clearly support well-distributed Au catalysts at the inner and outer surfaces of nc-SnO 2 .The EDS line profile shown in Figure S4, Supporting Information, shows a sudden drop of Au intensity at the center region, indicating missing Au catalysts at the ablated nc-SnO 2 top regions.In X-ray diffraction (XRD) data shown in Figure S5, Supporting Information, polycrystalline SnO 2 with (110), (101), and (200) peaks (JCPDS No. 41-1445) for all nc-SnO 2 and not-irradiated SnO 2 samples was clearly indicated.After laser irradiation, overall peak

Figure 4 .
Figure 4. a) Cross-sectional transmission electron microscopy (TEM) image of nc-SnO 2 with Au catalyst decoration.High-resolution TEM (HR-TEM) images of b) SnO 2 of nc-SnO 2 at the top and c) bottom region.HR-TEM images of Au catalysts decorated at d) outer and e) inner surfaces of nc-SnO 2 .f-h) Energy-dispersive X-ray spectroscopy (EDS) element mapping of f ) Sn, g) O, and h) Au for Au-decorated nc-SnO 2 .

Figure 6 .
Figure 6.Polar plot of responses of a) SnO 2 nanodomes (without laser irradiation) and b) Au-decorated nc-SnO 2 to 50 ppm C 2 H 5 OH, C 7 H 8 , CH 3 COCH 3 , C 6 H 6 , CH 3 SH, and 100 ppm CO. c) The ratio of responses (S b /S a ) between SnO 2 nanodomes and Au-decorated nc-SnO 2 .S a and S b represent the gas responses of SnO 2 nanodomes and Au-decorated nc-SnO 2 .

Figure 8 .
Figure 8. Schematic illustrations for the current path of a) SnO 2 nanodomes without laser irradiation, b) nc-SnO 2 , and c) Au-decorated nc-SnO 2 .The conduction band edges correspond to potential barriers in the air.d-f ) Schematic illustrations for O 2 adsorption and enlarged depletion region on the surface of d) SnO 2 nanodomes without laser irradiation, e) nc-SnO 2 , and f ) Au-decorated nc-SnO 2 .Note that the width of the surface depletion region is dramatically increased after inner surface exposure and Au decoration.