In Situ Monitoring of the Deposition of Flame-Made Chemoresistive

Jun 16, 2017 - Flame-deposited semiconducting nanomaterials on microelectronic circuitry exhibit exceptional performance as chemoresistive gas sensors...
1 downloads 17 Views 2MB Size
Subscriber access provided by UNIV OF ARIZONA

Article

In situ monitoring the deposition of flamemade chemoresistive gas sensing films Christoph Oliver Blattmann, Andreas T. T Güntner, and Sotiris E Pratsinis ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b04530 • Publication Date (Web): 16 Jun 2017 Downloaded from http://pubs.acs.org on June 20, 2017

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 28

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

ACS Applied Materials & Interfaces

In situ monitoring the deposition of flame-made chemoresistive gas sensing films

Christoph O. Blattmann, Andreas T. Güntner, Sotiris E. Pratsinis*

Particle Technology Laboratory, Institute of Process Engineering, Department of Mechanical and Process Engineering, ETH Zürich, Sonneggstrasse 3, 8092 Zürich, Switzerland.

8. June 2017

Keywords tin oxide, antimony, flame-spray pyrolysis, thin film, elongated, surface growth, monitor, fabrication control

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces

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 28

Abstract Flame-deposited semiconducting nanomaterials on microelectronic circuitry exhibit exceptional performance as chemoresistive gas sensors. Current manufacturing technology, however, does not monitor in situ the formation of such nanostructured films even though this can facilitate the controlled and economic synthesis of these sensors. Here, the resistance of such growing films is measured in situ during fabrication to monitor the creation of a semiconducting nanoparticle network for gas sensors. Upon formation of that network, the film resistance drops drastically to an asymptotic value that depends largely on film structure or morphology rather than on its thickness and nanoparticle size. Precursor solutions of various concentrations enable the flame deposition of Sb-doped SnO2 sensing films of different morphologies, each of which exhibit a characteristic in situ resistance pattern. Low precursor concentrations (1 mM) lead to thin (circa 0.16 µm) films with slender columnar structures of increasing diameter (up to 25 nm) after prolonged deposition (up to 6 min) and show an oscillating in situ resistance during their fabrication. On the other extreme, high precursor concentrations (100 mM) lead to thick (up to 80 µm) dendritic and porous films consisting of nanoparticles with relatively small primary particle diameter (around 7 nm) that remain invariant of deposition duration, in agreement with the stable in situ resistance. Such dendritic films exhibit an order of magnitude longer sensor recovery time than those made at lower concentrations. The above understanding enables the rapid and economic flame-synthesis of thin gas sensors consisting of minimal semiconducting nanomaterial mass, possessing a tuned baseline resistance and exhibiting excellent response to ethanol vapor.

ACS Paragon Plus Environment

2

Page 3 of 28

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

ACS Applied Materials & Interfaces

1. Introduction Gas sensors made by direct flame-deposition are attractive for their highly porous, crack-free1 and mechanically stable2 sensing films. They are composed of high surface area nanoparticles with diverse composition tuned for ppb-level sensitivity and high selectivity towards key analytes such as Si-doped WO3 and MoO3 for acetone3 and ammonia,4 respectively, or Ti-doped ZnO for isoprene.5 Such sensors are attractive for healthcare monitoring, disease diagnosis and even screening of patients by non-invasive breath analysis. Their outstanding performance is attributed to the stable synthesis of metastable phases6 and nanostructured and porous1 film morphology. Fabrication development has also led to the ability to process flame-made sensors by techniques compatible with standard silicon-wafer micromachining. This enables the preparation of sensors with low power consumption7 and recently to highly sensitive sensor arrays8 that are attractive for portable hand-held devices.9 So far the development of such sensors has been focused mostly on material composition and to a lesser extent on film morphology. However, optimizing the latter could boost analyte response even further as well as reduce the corresponding response/recovery times.10 Thinner structures, especially at and below twice the Debye length, bring along significant improvement in sensor sensitivity as was shown for pure and doped SnO2.11 This can be done with smaller nanoparticles12 which can bring along enhanced performance also with their greater interaction area.13 The sensor response, nevertheless, is governed by the finest structures within the sensing film, e.g., the necks between constituent nanoparticles.11 The use of thin and highly porous films in gas sensors therefore is considered ideal for high response and brings along the additional benefit of short response and recovery times.14 Measuring the in situ resistance during flame-deposition of nanostructured films offers fabrication control by providing immediate feedback when an interconnected film is created.15 Thus by monitoring the resistance of semiconducting films during their deposition, it is possible to prepare gas sensing films of minimal thickness and optimal resistance in a reproducible and economical manner. Recently, such in situ measurements have been conducted also during preparation of CuO nanowires16 ACS Paragon Plus Environment

3

ACS Applied Materials & Interfaces

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 4 of 28

and tetrafurcated 3D bridges17 for gas sensors by thermal oxidative growth and charged aerosol deposition, respectively. Single nanowires act as interconnections between substrate electrodes so that their growing number and size increase the sensor conductivity.16 The in situ resistance measurements during tetrafurcated 3D bridge preparation, on the other hand, enabled the precise definition of the required fabrication time for the formation of these structures.17 This work focuses on in situ monitoring the resistance of nanostructured films fabricated by direct deposition of flame-made Sb:SnO2 nanoparticles. This material, a good transparent conductor and infrared reflective coating,18 is ideal for gas sensors exhibiting a non-deteriorating low baseline resistance19 even below 100 °C20 and has been considered beneficial for its superior humidity stability.21 Here, the Sb:SnO2 film morphology was tuned by a substrate-impinging particle flame fed by a precursor solution containing varying metal ion concentrations. In situ resistance monitoring enabled control over final film resistance as well as direct insight into network formation by material deposition and nanoparticle growth, necking and coalescence. The performance of these films as ethanol vapor sensors was assessed and related to their in situ resistance and nanoparticle film characteristics. As shown here, the understanding gained by monitoring the in situ resistance of gas sensors led to their faster preparation by conventional2 flame-deposition.1 It is quite likely to be beneficial also for the development and optimization of thin continuous films made by flame aerosol technologies for batteries,22 solar23 and chemical24 energy conversion, UV-light detectors25 and diffusion barriers26 as shown for conductive polymer nanocomposites.15

2. Experimental 2.1 Film fabrication Gas sensing nanoparticle films were prepared by direct deposition of flame-made27 Sb:SnO2 by dispersing a combustible organometallic precursor (11 ml/min) by 5 L/min oxygen (PanGas, technical grade) ignited with a methane pilot flame (1.5 L/min CH4 (PanGas, 3.5) and 3.2 L/min O2). A ACS Paragon Plus Environment

4

Page 5 of 28

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

ACS Applied Materials & Interfaces

standard28 flame-spray pyrolysis (FSP) burner was used. The precursors contained 1, 10 and 100 mM of total metallic atoms (i.e., Sn and Sb) by dissolving tin(II) 2-ethylhexanoate (Sigma Aldrich, 92.5100%) and antimony(III) propoxide (Sigma Aldrich, 98%) in xylene (Sigma Aldrich, puriss). The ratio Sn:Sb was fixed at 7.333 in order to obtain 12 at% Sb:SnO2 with low resistance.27 The Sb:SnO2 was deposited on alumina substrates having a pair of interdigitated platinum electrodes, a resistance temperature detector (RTD) and back-heater (Electronics Design Center (Cleveland, OH), sensor #103). The substrate was fixed to a water-cooled (10 L/min, ~20 °C inlet temperature) copper holder at 14.5 cm height above the burner (HAB). Just prior to deposition the substrate was preheated for 20 s by exposure to an identical flame fed only by xylene (i.e., metal-free). Deposition was carried-out for up to 6 min. Substrate preheating, film deposition and cooling to room temperature was closely monitored by the in situ resistance (Ri)15 between the interdigitated electrodes with a multimeter (Tektronix DMM4050). No further processing steps, such as film calcination,1 were conducted. For comparison conventional sensing film fabrication was done with a 500 mM precursor solution concentration (PSC):2 nanoparticle deposition for up to 3 min at 20 cm HAB and in situ annealing for 30 s at 14.5 cm HAB.

2.2 Gas sensing Sensor testing was conducted in a chamber29 fed by 1 L/min dry synthetic air (PanGas, hydrocarbon-free grade). For analyte sensing, 1 ppm ethanol (Pan Gas, 10.4 ppm ethanol in dry synthetic air) was admixed to the carrier gas.4 The inlet gas was preheated to 55 °C. The substrate backheater was supplied with an electrical current (power supply: EA-PS 2384-05 B) to obtain an operating temperature of 300 °C. This temperature was attained by increasing the heater inlet voltage by 1 V/min during sensor booting. Resistance of the sensing film was measured by a Keithley 2700 multimeter. The sensor response was computed from the baseline resistance in dry air (R0) and the response

ACS Paragon Plus Environment

5

ACS Applied Materials & Interfaces

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 6 of 28

resistance to ethanol (R1):29 (R0/R1)-1. The response and recovery time towards ethanol exposure was determined as the duration when 90% of the change in resistance (i.e., ǀ0.9·(R0-R1)ǀ) was reached.

2.3 Nanoparticle film and powder characterization Nanoparticle sensing films were viewed by scanning electron microscopy (Hitachi S-4800 FESEM) at the Optical Materials Engineering Laboratory (OMEL) at ETH Zürich. Cross-section images were acquired by viewing along the fracture of a substrate divided in half. Film thickness was computed with the software ImageJ. Nanoparticle size and morphology at various HAB was determined by thermophoretic sampling30 on carbon-coated copper grids (Plano GmbH (Wetzlar, Germany), S162-4). Their particle size was measured from transmission electron microscopy (TEM) images (FEI Tecnai F30 FEG at ScopeM ETH Zürich) with the software ImageJ. X-ray diffraction (XRD) was conducted with a Bruker D2 Phaser 2nd Generation on nanoparticle powder collected on a water-cooled glass fiber filter downstream of the deposition substrate. The lower precursor solution concentrations (i.e., 1 and 10 mM) resulted in insufficient amount of filter-collected material so that a cutout of the particle-loaded filter was placed on the XRD dish instead. The software Topas and EVA were used to determine the crystal size (dXRD) and composition, respectively.

3. Results and Discussion Films of Sb:SnO2 nanoparticles (for XRD characterization see Supporting Information (SI), Fig. S1) are deposited during flame-spray pyrolysis (FSP) at high temperatures so that the particle-laden flame engulfs the back-cooled alumina substrate (Fig. 1). This is made possible by the selection of a high FSP precursor solution feed-rate (i.e., 11 ml/min) and low HAB (i.e., 14.5 cm) exposing the substrate front to temperatures just below 1200 °C ( SI, Fig. S2). Conventionally, flame-deposited nanoparticle films are prepared at much lower precursor solution feed-rates and greater HAB,1 but are ACS Paragon Plus Environment

6

Page 7 of 28

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

ACS Applied Materials & Interfaces

frequently also exposed to similarly high temperatures during subsequent in situ flame annealing to increase their cohesion and adhesion to the substrate.2 During such high-temperature nanoparticle deposition (SI, Fig. S2) the formation of an interconnected network is monitored in situ by the resistance15 (Ri) between a pair of interdigitated electrodes on the substrate (Fig. 1). The prevailing deposition temperatures were sufficient also to monitor the Ri of non-doped SnO2 (SI, Fig. S3) but employing room-temperature conductive18 Sb:SnO2 was of additional benefit.

Figure 1. Schematic of the direct deposition of FSP-made Sb:SnO2 with a substrate-impinging nanoparticle flame. The nanoparticle network formation on the substrate is continuously monitored during preparation by measuring the in situ resistance (Ri) between the interdigitated electrodes.

3.1 In situ resistance monitored deposition Figure 2 shows the Ri as a function of Sb:SnO2 deposition duration on alumina substrates with interdigitated Pt-electrodes using 0 (diamonds), 1 (squares), 10 (circles) and 100 mM (triangles) FSP ACS Paragon Plus Environment

7

ACS Applied Materials & Interfaces

precursor solution concentrations (PSC). During preheating, the Ri drops by an order of magnitude upon exposure to the flame even in the absence of particles.

1.E+08 108

PSC, mM: 0

1.E+07 107

1

In situ resistance Ri, Ω

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 8 of 28

10

1.E+06 106

100

1.E+05 105 1.E+04 104 1.E+03 103 1.E+02 102 1.E+01 101 0 0

60 1

120 2

180 3

240 4

300 5

360 6

420

Deposition duration, min Figure 2. The in situ resistance Ri of Sb:SnO2 nanoparticle films during their deposition at 0 (diamonds), 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentrations (PSC). The 0 mM solution concentration refers to a pure solvent (i.e. metal-free) precursor. Prior to deposition, the substrate is preheated for 20 s as depicted by the vertical dashed line.

At the start of deposition (Fig. 2, t =0 min) Ri exhibits a PSC-dependent reduction with high reproducibility (SI, Fig. S4). At low PSC (1 mM), the Ri remains at first (up to ca. 1 min) hardly distinguishable from the particle-free conditions as it takes some time to build the interconnected particle bridges. But as deposition is prolonged (t >1.5 min), the Ri starts to drop. This clearly indicates ACS Paragon Plus Environment

8

Page 9 of 28

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

ACS Applied Materials & Interfaces

the formation of an interconnected network15 of nanoparticles as shown in top-view SEM images (Fig. 3) of such films deposited for 1 (a), 2 (b) and 4 min (c) when using a 1 mM PSC. After 1 min only very small, spherical and well separated nanoparticles can be seen (bright spots). Until then no interconnected network exists so that Ri remains unchanged (Fig. 2). At 2 min, the nanoparticles on the substrate more than double in size from below 10 nm to about 20 nm (squares in Fig. 3k). Such growth leads to some necking with their surrounding nanoparticles (Fig. 3b), similar to nanosilver,15 and results in an interconnected network reducing the Ri (Fig. 2). This becomes even more apparent after 4 min of particle deposition. But the increment in deposited nanoparticle size between 2 and 4 min is not as large as between 1 and 2 min (Fig. 3k) when determined from top-view SEM images. By viewing their cross-section on the other hand (Fig. 3d), one can notice that these films consist of elongated structures perpendicular to the substrate, in particular after 4 min deposition. It must be noted, though, that these structures form on top of a very thin film of Sb:SnO2 nanoparticles (SI, Fig S5) that most likely acts as a seed layer.31 These elongated Sb:SnO2 structures form primarily at the top of substrate alumina grains rather than between them (SI, Fig. S5). By using a planar substrate, such as a polished Si-wafer (SI, Fig. S6), such growth occurs homogeneously over the entire substrate making them potentially attractive for integration in thin film devices especially in combination with other metal oxides, such as ZnO (Fig. S3b). Also quite notable is the strong oscillation of Ri, by more than an order of magnitude (Fig. 2), during network formation. This oscillation is measured reproducibly at such low PSC (SI, Fig. S4) and is present also during pure SnO2 and ZnO deposition (SI, Fig. S3). This stems from the constant formation (by deposition) and break-up of necked nanoparticles (by coalescence32) on the substrate (e.g., repeatedly from lace-like to cauliflower-like structures2) during flame-deposition. Particle growth, necking and coalescence constantly alter the network and therefore its Ri. This becomes especially pronounced at low PSC due to the thin substrate surface adhering nanoparticle layer as well as the fine elongated structures (Fig. 3d). ACS Paragon Plus Environment

9

ACS Applied Materials & Interfaces

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 28

Figure 3. Top (a-c,e-j) and cross-section (d) SEM images of Sb:SnO2 nanoparticle films deposited with 1 (top row, red), 10 (middle row, blue) and 100 mM (bottom row, green) PSCs for 1 (a,e,h), 2 (b,f,i) and 4 minutes (c,d,g,j). All images are at the same magnification as in j, except d. The particle size on the substrate was measured from these top view images (k). The evolving nanoparticle size (i.e. diameter) as a function of deposition duration for 1 (squares), 10 (circles) and 100 mM (triangles) PSC.

ACS Paragon Plus Environment

10

Page 11 of 28

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

ACS Applied Materials & Interfaces

Films of aligned and elongated SnO2 have been considered beneficial for field emission applications by improving the current density and reducing the turn-on field.33 Their formation on substrates has been obtained by classic chemical vapor deposition (CVD)34 of TiO2 and thermal evaporation of WO3 in a flame35 as well as during FSP-deposition of SiO2 36 and nanosilver.15 The latter two require slow deposition rates that are realized, for example, by low PSCs or precursor solution feed-rates. Thermophoretic sampling of Sb:SnO2 (SI, Fig. S7) confirm that nanoparticles are present in the off-gas at the deposition HAB even with a 1 mM PSC which, nevertheless, does not disqualify also the presence of gaseous precursor. In fact, it is quite likely since the formation of such elongated structures usually is traced to columnar growth by gas-to-particle conversion36 with possible dependency on the temperature,37 its gradient towards the cooled substrate35 and gas compounds present during growth.37 The geometric standard deviation below the threshold of a self-preserving size distribution (SI, Fig. S8) supports this hypothesized gas-to-particle conversion. Increasing the PSC from 1 to 10 mM leads naturally to a higher generation rate of nanoparticles and therefore faster deposition on the substrate (Fig. 2). This is reflected in the Ri where the drop in resistance occurs rapidly, way before 1 min (circles in Fig. 2). The Ri attains a relatively constant value after a decrease of over five orders of magnitude within 2 min deposition. The quicker deposition and network formation on the substrate in comparison to the 1 mM PSC is also optically confirmed from top-view SEM images (Fig. 3e-g) where already after 1 min the nanoparticles are assembled more densely (compare Fig. 3a and e). Figure 4 shows the average film thickness as a function of nanoparticle deposition duration for 1 (squares), 10 (circles) and 100 mM (triangles) PSCs. Only tens of nanometer thickness are achieved in the first 2 min of deposition at the lowest PSC, whereas for the 10 mM PSC already about 100 nm are obtained after 1 min. The latter thickness is barely achieved by the compact Sb:SnO2 film for the 1 mM PSC even for >4 min deposition duration. The absolute decrease of Ri is significantly greater for the 10 mM PSC than that obtained with the 1 mM (Fig. 2) which is clearly related to the differentiating film ACS Paragon Plus Environment

11

ACS Applied Materials & Interfaces

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 12 of 28

thickness and morphology. While the former is intuitive (i.e., thicker films have lower resistance38), the latter can be deduced from the SEM images (Fig. 3a-g). There one can see that Sb:SnO2 deposition with 10 mM PSC leads to formation of thickly necked nanoparticle clusters already after 1 min (Fig. 3e). These clusters become more pronounced after 2 and 4 min deposition (Fig. 3f and g, respectively) due to slight nanoparticle growth (Fig. 3k) and especially coalescence of neighboring clusters to larger ones. Such clusters exhibit a compact morphology (SI, Fig. S9 for SEM of film cross-sections) enabling the entire film to contribute to electron conduction. This contrasts to the elongated structures obtained with the 1 mM PSC (Fig. 3d) where complex electron paths between them most likely restrict electron flow.39 At 10 mM PSC, the Ri still oscillates by an order of magnitude during deposition which again reflects the formation and break-up of nanoparticle networks by deposition and coalescence. It occurs at a higher rate than for the lowest PSC, due to faster Sb:SnO2 deposition, and thus cannot be resolved distinctly by the relatively large sampling intervals of the multimeter. The earlier drop of Ri with increasing PSC is also confirmed for the highest PSC (100 mM) where the network has formed already after about 30 s (Fig. 2). However, the Ri levels off between 10 and 100 kΩ which differs markedly from that (0.1-1 kΩ) obtained with the 10 mM PSC. Its greater value can be related to film morphology. Fig. 3h-j show top-view images of such films after 1, 2 and 4 min deposition, respectively. These fragile and porous films possess a dendritic structure and are composed of loosely connected nanoparticles of about 7 nm diameter independent of duration (Fig. 3k). In fact, their small size is only slightly larger than that of thermophoretically sampled aerosols (3-4 nm) at the same HAB and PSC (SI, Fig. S7). This suggests that limited nanoparticle growth occurs upon deposition on the substrate and thus hindering necking. As a result, Ri is strongly influenced by the numerous thin nanoparticle necks and remains much higher40 than that made at 10 mM PSC.

ACS Paragon Plus Environment

12

Page 13 of 28

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

ACS Applied Materials & Interfaces

Figure 4. Nanoparticle film thickness (from SEM cross-section images) as a function of flame deposition duration when using a 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentration (PSC). Insets are cross-section images of films prepared for 2 min deposition with 1 (b) and 100 mM (a) PSC. The film thickness is indicated by yellow arrows. Note the significantly different magnification of (a) and (b). Cross-section images of films prepared with 10 mM PSC can be found in Fig. S9)

At high PSC, the rapid Sb:SnO2 deposition exceeds nanoparticle sintering- and growth rates at the substrate resulting in much thicker (tens of micrometer) but highly porous and dendritic films (Fig. 4a). Cross-section images indicate that such films are similar to lace-like2 flame-deposited nanoparticle

ACS Paragon Plus Environment

13

ACS Applied Materials & Interfaces

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 14 of 28

films without in situ annealing. There the nanoparticle size does not vary between bottom and top of the film indicating the absence of sintering. In other words, the earlier deposited nanoparticles do not grow by sintering with neighboring ones while the film continues to build up above them. In contrast to the lowest PSC (1 mM), the highest PSC used here (100 mM) results in a rather stable Ri (Fig. 2, triangles) confirming that nanoparticle growth and necking as well as breakup of nanoparticle bridges are insignificant at this high deposition rate.

3.2 Gas sensing Figure 5a shows the baseline resistance (R0) as a function of deposition duration at 300 °C in dry air for sensing films prepared from 1 (squares), 10 (circles) and 100 mM (triangles) PSCs. The diamonds represent sensors prepared with the same composition but by conventional flame-deposition followed by in situ flame-annealing.2 Films with insufficient deposited material (e.g., 1 and 2 min with 1 mM PSC) could not be employed as gas sensors since they exhibited too high R0 due to insufficient formation of interconnected networks (Fig. 2 & 3). The R0 is lower for films with prolonged deposition as is expected due to greater deposited mass, nanoparticle growth and sintering (Fig. 3) resulting in thicker films (Fig. 4). In general R0 is lower for these Sb:SnO2 sensors19 than that of pure SnO2 gas sensors by about three orders of magnitude41 independent of PSC. The R0 is in fair agreement with Ri at the end of the deposition process (Fig. 2). For example, Ri is around 100 kΩ after 6 min of deposition with a 1 mM PSC (Fig. 2, squares) while the average R0 lays just above 100 kΩ (Fig. 5a, squares). Slightly larger R0 than Ri is expected due to the lower sensing temperature than during deposition (300 versus 1200 °C) and the absence of possibly interacting flue gases within the sensing chamber. The R0 of sensing films prepared with 10 mM PSC barely changes for longer deposition duration (Fig. 5a, circles). This agrees with Ri which also exhibits no significant decrease after 1 min of deposition (Fig. 2). A similar agreement emerges for the 100 mM

ACS Paragon Plus Environment

14

Page 15 of 28

ACS Applied Materials & Interfaces

1 2 PSC. In sum, the R0 is consistent with the in situ measured resistance for all PSCs. This enables precise 3 4 tuning of its final value already during fabrication in order to meet the specifications compatible with 5 6 device electronics. 7 8 9 1000000008 10 15 10 PCS, mM: dry air @ 300°C 11 1 ppm ethanol 1 12 14 7 10000000 10 10 13 14 100 13 15 dep. + in situ annealing 106 16 1000000 17 18 8 19 100000 105 20 6 21 22 10000 4 10 23 4 24 25 1000 103 26 2 27 Shao, et al. [20] 28 Wang, et al. [42] 2 100 0 10 29 0 2 4 6 0 2 4 6 30 31 Deposition duration, min Deposition duration, min 32 33 Figure 5. The baseline resistance R0 in dry air at 300 °C as a function of FSP deposition duration (a) 34 for Sb:SnO2 nanoparticle films made with 1 (squares), 10 (circles) and 100 mM (triangles) precursor 35 36 solution concentration (PSC). Sensors prepared by flame-deposition with in situ annealing2 are 37 38 represented as diamonds. The corresponding response towards 1 ppm of ethanol in dry air at 300 °C is 39 40 shown in (b). Extrapolated data of Sb:SnO2 sensors from Wang et al.42 and Shao et al.20 are indicated 41 42 by dashed lines. 43 44 45 46 The low R0 obtained by ordinary deposition with in situ annealing2 is comparable to that obtained 47 48 in sensing films prepared from 10 mM PSC. This is reasonable since they exhibit a similar compact 49 50 morphology (compare SI, Fig. S10 with Fig. 3e-g) indicative of low resistance. But due to inconclusive 51 52 53 Ri measurements during nanoparticle deposition (SI, Fig. S11), resistance-based film optimization 54 55 becomes a tedious trial-and-error procedure. For example, the sensor prepared with 5 s of deposited 56 57 58 59 60

b

Response (R0/R1)-1, -

Baseline resistance R0, Ω

a

ACS Paragon Plus Environment

15

ACS Applied Materials & Interfaces

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 16 of 28

material exhibits a large R0 (Fig. 5a) which could not be foreseen during fabrication. In situ Ri measurements therefore do not lead to insight into the fabrication process. Figure 5b shows the response to 1 ppm ethanol in dry air at 300 °C by sensors made at various deposition durations at 1 (squares), 10 (circles) and 100 mM (triangles) PSCs as well as that response from sensors made by conventional flame-deposition with in situ annealing2 (diamonds). All sensors prepared (while in situ monitoring their Ri) exhibit an average response between 2 and 3. This is significantly greater than other Sb-doped SnO2 gas sensors at 1 ppm ethanol (dashed lines in Fig. 5b).20, 42

In fact, such responses are even quite comparable, if not better, than other SnO2-based and

commercial MOx gas sensors for ethanol detection (SI, Table S1). As some FSP-made gas sensors exhibit greater responses than the present ones, there is potential for their further optimization using the on-line resistance monitoring during their synthesis. The analyte response data suggest an increasing average response for finer nanoparticles (i.e., smaller size) as achieved, for example, at shorter deposition duration and greater PSCs (Fig. 3k). More specifically, the films prepared with 10 and 100 mM PSC at the shortest deposition duration (1 min) exhibit the greatest average response (Fig. 5b). The elongated structures43 in films prepared with the 1 mM PSC (Fig. 3d) may have led to the slightly larger average response44 for longer deposition duration. The response of the Ri-monitored films are lower than that of sensors prepared by flamedeposition with in situ annealing where values up to about 14 are achieved (Fig. 5b). This deviation may arise from differences in morphology10 and/or surface species/oxygen vacancies.45 Actually, sensors prepared by flame-deposition with in situ annealing also suggest an increasing response with decreasing deposition duration. Thinner films are therefore considered beneficial for improving performance as has been noted previously46 and even observed for flame-deposited SnO2.47 It is yet unclear, though, which properties, for example, film morphology10 or Pt-electrodes from the

ACS Paragon Plus Environment

16

Page 17 of 28

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

ACS Applied Materials & Interfaces

substrate,48 contribute to this enhancement. A more detailed study dedicated to this aspect may enable more insight. Figure 6 shows the response (a) and recovery times (b) to 1 ppm ethanol at 300 °C in dry air as a function of deposition duration of Sb:SnO2 sensors prepared from 1 (squares), 10 (circles) and 100 mM (triangles) PSC. These times for sensors made by deposition and in situ annealing2 are depicted by diamonds. A quicker response is indicated for thinner films (i.e., lower PSC and/or shorter deposition duration), in agreement with In2O3 sensing films of decreasing thickness.49 The Sb:SnO2 nanoparticle films prepared from 1 mM PSCs thus exhibited a response time of less than 10 s. Such short response times can be attributed to the reduced total adsorption surface area and/or quicker diffusion of ethanol to all active sites. Sensors prepared by flame-deposition with in situ annealing exhibit similar decrease of response time for shorter deposition duration. A response time down to 5 s for 0.5 min deposition duration could be achieved. Such short response times are superior to flame-deposited SnO2 films for ethanol detection.47 The recovery time of the sensors prepared from 1 and 10 mM PSCs ranges from 15 to about 20 min (Fig. 6b) and decreases for thinner films (i.e., shorter deposition durations) similar to the response time (Fig. 6a). This compares well with that of sensors made by flame-deposition with in situ annealing (diamonds). The rather long times (multiple minutes) are in fair agreement with flame-made Sb:SnO2 used for methane detection19 and most likely are due to the slow desorption50 of ethanol51 and/or its derivatives from the nanoparticle surface as well as the re-adsorption48 of oxygen. The much thicker nanoparticle films prepared from a 100 mM PSC possess recovery times above 2.5 h. Such long times may be attributed to their large thickness46 (Fig. 4), highly porous morphology (Fig. 3h-j) and high surface area. This could lead to large adsorption of ethanol and/or its derivatives within the film that are outgased only slowly during recovery.

ACS Paragon Plus Environment

17

ACS Applied Materials & Interfaces

100

100000

a

10

Recovery time, s

Response time, s

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

1

0

2

PCS, mM: 1 10 100

b

deposition + in situ annealing

10000

1000

1 ppm ethanol dry air @ 300°C

0.1

Page 18 of 28

100

4

6

0

Deposition duration, min

2

4

6

Deposition duration, min

Figure 6. The Sb:SnO2 sensor response (a) and recovery times (b) as a function of nanoparticle deposition duration for films prepared with 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentrations (PSC) as well as such made by deposition with in situ annealing2 (diamonds).

4. Conclusions The in situ resistance of Sb:SnO2 nanoparticle films was monitored during their direct flamedeposition onto interdigitated alumina substrates. Using different precursor solution concentrations (PSC) enabled the preparation of small nanoparticles in the range of 3 to 7 nm and, more importantly, varied the obtained nanoparticle film morphology and thickness due to particle growth at the substrate. Low PSCs resulted in compact films down to 10 nm thickness. Tens of micrometer thick films with porous dendritic morphology were prepared at greater PSC. The onset or creation of the nanoparticle network in these films could be followed in real-time during their fabrication by conducting in situ resistance measurements. This enabled the ability to optimize material quantity, fabrication time and also the final film resistance. The nanoparticle film resistance could be linked with its evolving particle size and necking. ACS Paragon Plus Environment

18

Page 19 of 28

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

ACS Applied Materials & Interfaces

Reproducible fabrication of such in situ resistance-monitored nanoparticle films was key for the optimization of Sb:SnO2 gas sensors. Compact and thin morphology were best for low baseline resistance and faster response/recovery time towards ethanol. This led to rapid synthesis of thin flamemade gas sensors with improved analyte response, reduced response/recovery times and economized fabrication following conventional flame-deposition and in situ annealing.

Associated Content Supporting Information is available free of charge on the ACS Publication website: PDF document containing extended experimental results in the form of figures and descriptive text referred to in this main document. In brief, these results show XRD particle characterization, substrate temperature evaluation, Ri measurement during SnO2 deposition, multiple reproductions of Ri measurements during Sb:SnO2 deposition, Ri measurements during conventional flame aerosol deposition with in situ annealing, electron microscopy images of nanoparticle films on alumina and silicon substrates, characterization of aerosol particle size and standard deviation from thermophoretic sampling, comparative table of published SnO2-based gas sensors for ethanol detection.

Author Information Corresponding Author * E-mail: [email protected] Tel.: +41(0)446323180

Acknowledgements We thank Dr. Frank Krumeich (Electron Microscopy Center) for TEM analysis and the Optical Materials Engineering Laboratory (Prof. D.J. Norris) for access to SEM (both at ETH Zurich). Financial support from the ETH Zürich Research Grant (ETH-08 14-2) and the Swiss National Science Foundation (grant 200021_159763/1) are kindly acknowledged.

References 1. Mädler, L.; Roessler, A.; Pratsinis, S. E.; Sahm, T.; Gurlo, A.; Barsan, N.; Weimar, U., Direct Formation of Highly Porous Gas-Sensing Films by In-Situ Thermophoretic Deposition of FlameMade Pt/SnO2 Nanoparticles. Sens. Actuator. B-Chem. 2006, 114 (1), 283-295.

ACS Paragon Plus Environment

19

ACS Applied Materials & Interfaces

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 20 of 28

2. Tricoli, A.; Graf, M.; Mayer, F.; Kuhne, S.; Hierlemann, A.; Pratsinis, S. E., Micropatterning Layers by Flame Aerosol Deposition-Annealing. Adv. Mater. 2008, 20 (16), 3005-3010. 3. Righettoni, M.; Tricoli, A.; Pratsinis, S. E., Si:WO3 Sensors for Highly Selective Detection of Acetone for Easy Diagnosis of Diabetes by Breath Analysis. Anal. Chem. 2010, 82 (9), 35813587. 4. Güntner, A. T.; Righettoni, M.; Pratsinis, S. E., Selective Sensing of NH3 by Si-Doped α-MoO3 for Breath Analysis. Sens. Actuator. B-Chem. 2016, 223, 266-273. 5. Güntner, A. T.; Pineau, N. J.; Chie, D.; Krumeich, F.; Pratsinis, S. E., Selective Sensing of Isoprene by Ti-Doped ZnO for Breath Diagnostics. J. Mater. Chem. B 2016, 4 (32), 5358-5366. 6. Wang, L.; Teleki, A.; Pratsinis, S. E.; Gouma, P. I., Ferroelectric WO3 Nanoparticles for Acetone Selective Detection. Chem. Mater. 2008, 20 (15), 4794-4796. 7. Kühne, S.; Graf, M.; Tricoli, A.; Mayer, F.; Pratsinis, S. E.; Hierlemann, A., Wafer-Level FlameSpray-Pyrolysis Deposition of Gas-Sensitive Layers on Microsensors. J. Micromech. Microeng. 2008, 18 (3), 35040. 8. Güntner, A. T.; Koren, V.; Chikkadi, K.; Righettoni, M.; Pratsinis, S. E., E-Nose Sensing of Lowppb Formaldehyde in Gas Mixtures at High Relative Humidity for Breath Screening of Lung Cancer? ACS Sens. 2016, 1 (5), 528-535. 9. Righettoni, M.; Ragnoni, A.; Güntner, A. T.; Loccioni, C.; Pratsinis, S. E.; Risby, T. H., Monitoring Breath Markers Under Controlled Conditions. J. Breath Res. 2015, 9 (4), 047101. 10. Barsan, N.; Weimar, U., Conduction Model of Metal Oxide Gas Sensors. J. Electroceram. 2001, 7 (3), 143-167. 11. Xu, C.; Tamaki, J.; Miura, N.; Yamazoe, N., Grain Size Effects on Gas Sensitivity of Porous SnO2Based Elements. Sens. Actuator. B-Chem. 1991, 3 (2), 147-155. 12. Tamvakos, A.; Calestani, D.; Tamvakos, D.; Mosca, R.; Pullini, D.; Pruna, A., Effect of Grain-Size on the Ethanol Vapor Sensing Properties of Room-Temperature Sputtered ZnO Thin Films. Microchim. Acta 2015, 182 (11), 1991-1999. 13. Zhou, Q.; Chen, W.; Li, J.; Tang, C.; Zhang, H., Nanosheet-Assembled Flower-Like SnO2 Hierarchical Structures with Enhanced Gas-Sensing Performance. Mater. Lett. 2015, 161, 499502. 14. Zhang, W.; Zeng, W.; BinMiao; Wang, Z., Effect of the Sheet Thickness of Hierarchical SnO2 on the Gas Sensing Performance. Appl. Surf. Sci. 2015, 355, 631-637. 15. Blattmann, C. O.; Pratsinis, S. E., In Situ Measurement of Conductivity during Nanocomposite Film Deposition. Appl. Surf. Sci. 2016, 371, 329-336. 16. Steinhauer, S.; Chapelle, A.; Menini, P.; Sowwan, M., Local CuO nanowire growth on microhotplates: In situ electrical measurements and gas sensing application. ACS Sens. 2016, 1 (5), 503-507. 17. Bae, Y.; Pikhitsa, P. V.; Cho, H.; Choi, M., Multifurcation Assembly of Charged Aerosols and its Application to 3D Structured Gas Sensors. Adv. Mater. 2017, 29 (2), 1604159. 18. Gordon, R. G., Criteria for Choosing Transparent Conductors. MRS. Bull. 2000, 25 (8), 52-57. 19. Grossmann, K.; Kovács, K. E.; Pham, D. K.; Mädler, L.; Barsan, N.; Weimar, U., Enhancing Performance of FSP SnO2-Based Gas Sensors Through Sb-Doping and Pd-Functionalization. Sens. Actuator. B-Chem. 2011, 158 (1), 388-392. 20. Shao, S.; Koehn, R.; Wu, H.; Wu, T.; Rao, W.-F., Generation of Highly Ordered Nanoporous SbSnO2 Thin Films with Enhanced Ethanol Sensing Performance at Low Temperature. New J. Chem. 2016, 40 (6), 5523-5530. ACS Paragon Plus Environment

20

Page 21 of 28

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

ACS Applied Materials & Interfaces

21. Suematsu, K.; Sasaki, M.; Ma, N.; Yuasa, M.; Shimanoe, K., Antimony-doped tin dioxide gas sensors exhibiting high stability in the sensitivity to humidity changes. ACS Sensor. 2016, 1 (7), 913-920. 22. Chew, S. Y.; Patey, T. J.; Waser, O.; Ng, S. H.; Büchel, R.; Tricoli, A.; Krumeich, F.; Wang, J.; Liu, H. K.; Pratsinis, S. E.; Novák, P., Thin Nanostructured LiMn2O4 Films by Flame Spray Deposition and In Situ Annealing Method. J. Power Sources 2009, 189 (1), 449-453. 23. Tricoli, A.; Nasiri, N.; Chen, H.; Wallerand, A. S.; Righettoni, M., Ultra-Rapid Synthesis of Highly Porous and Robust Hierarchical ZnO Films for Dye Sensitized Solar Cells. Sol. Energy 2016, 136, 553-559. 24. Cho, K.; Oh, J.; Lee, T.; Shin, D., Fabrication of Solid Thin Film Electrolyte and LiCoO2 by Aerosol Flame Pyrolysis Deposition. J. Anal. Appl. Pyrolysis 2007, 80 (2), 502-506. 25. Nasiri, N.; Bo, R.; Wang, F.; Fu, L.; Tricoli, A., Ultraporous Electron-Depleted ZnO Nanoparticle Networks for Highly Sensitive Portable Visible-Blind UV Photodetectors. Adv. Mater. 2015, 27 (29), 4336-4343. 26. Nédélec, R.; Neagu, R.; Uhlenbruck, S.; Maric, R.; Sebold, D.; Buchkremer, H. P.; Stöver, D., Gas Phase Deposition of Diffusion Barriers for Metal Substrates in Solid Oxide Fuel Cells. Surf. Coat. Technol. 2011, 205 (16), 3999-4004. 27. Bubenhofer, S. B.; Schumacher, C. M.; Koehler, F. M.; Luechinger, N. A.; Sotiriou, G. A.; Grass, R. N.; Stark, W. J., Electrical resistivity of assembled transparent inorganic oxide nanoparticle thin layers: Influence of silica, insulating impurities, and surfactant layer thickness. ACS Appl. Mater. Interfaces 2012, 4 (5), 2664-2671. 28. Mädler, L.; Stark, W. J.; Pratsinis, S. E., Simultaneous Deposition of Au Nanoparticles During Flame Synthesis of TiO2 and SiO2. J. Mater. Res. 2003, 18 (01), 115-120. 29. Righettoni, M.; Tricoli, A.; Gass, S.; Schmid, A.; Amann, A.; Pratsinis, S. E., Breath Acetone Monitoring by Portable Si:WO3 Gas Sensors. Anal. Chim. Acta 2012, 738 (0), 69-75. 30. Dobbins, R. A.; Megaridis, C. M., Morphology of Flame-Generated Soot as Determined by Thermophoretic Sampling. Langmuir 1987, 3 (2), 254-259. 31. Ye, Z.; Ji, X.; Zhang, Q., Effect of Buffer Layer on Growth and Properties of ZnO Nanorod Arrays. J. Mater. Sci. Mater. Electron. 2015, 26 (7), 5232-5236. 32. Eggersdorfer, M. L.; Pratsinis, S. E., Agglomerates and Aggregates of Nanoparticles Made in the Gas Phase. Adv. Powder Technol. 2014, 25 (1), 71-90. 33. Luo, S. H.; Wan, Q.; Liu, W. L.; Zhang, M.; Di, Z. F.; Wang, S. Y.; Song, Z. T.; Lin, C. L.; Dai, J. Y., Vacuum Electron Field Emission from SnO2 Nanowhiskers Synthesized by Thermal Evaporation. Nanotechnology 2004, 15 (11), 1424. 34. Byun, D.; Jin, Y.; Kim, B.; Lee, J. K.; Park, D., Photocatalytic TiO2 Deposition by Chemical Vapor Deposition. J. Hazard. Mater. 2000, 73 (2), 199-206. 35. Xu, F.; Tse, S. D.; Al-Sharab, J. F.; Kear, B. H., Flame Synthesis of Aligned Tungsten Oxide Nanowires. Appl. Phys. Lett. 2006, 88 (24), 243115. 36. Tricoli, A.; Righettoni, M.; Krumeich, F.; Stark, W. J.; Pratsinis, S. E., Scalable Flame Synthesis of SiO2 Nanowires: Dynamics of Growth. Nanotechnology 2010, 21 (46), 465604. 37. Xu, F.; Liu, X.; Tse, S. D.; Cosandey, F.; Kear, B. H., Flame Synthesis of Zinc Oxide Nanowires. Chem. Phys. Lett. 2007, 449 (1–3), 175-181. 38. Kim, H.; Gilmore, C. M.; Piqué, A.; Horwitz, J. S.; Mattoussi, H.; Murata, H.; Kafafi, Z. H.; Chrisey, D. B., Electrical, Optical, and Structural Properties of Indium-Tin-Oxide Thin Films for Organic Light-Emitting Devices. J. Appl. Phys. 1999, 86 (11), 6451-6461. ACS Paragon Plus Environment

21

ACS Applied Materials & Interfaces

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 22 of 28

39. Bukauskas, V.; Olekas, A.; Senuliené, D.; Strazdiené, V.; Setkus, A.; Kaciulis, S.; Pandolfi, L., Effect of Thickness of Ultra-Thin Tin Oxide Film Based Gas Sensors. Lith. J. Phys. 2007, 47 (4), 475-483. 40. Goyal, D. J.; Agashe, C.; Marathe, B. R.; Takwale, M. G.; Bhide, V. G., Effect of Dopant Incorporation on Structural and Electrical Properties of Sprayed SnO2-Sb Films. J. Appl. Phys. 1993, 73 (11), 7520-7523. 41. Tricoli, A.; Righettoni, M.; Pratsinis, S. E., Minimal Cross-sensitivity to Humidity During Ethanol Detection by SnO2-TiO2 Solid Solutions. Nanotechnology 2009, 20 (31), 315502. 42. Wang, M. Y.; Zhu, L. F.; Zhang, C. Y.; Gai, G. S.; Ji, X. W.; Li, B. H.; Yao, Y. W., Lanthanum Oxide @ Antimony-Doped Tin Oxide with High Gas Sensitivity and Selectivity Towards Ethanol Vapor. Sens. Actuator. B-Chem. 2016, 224, 478-484. 43. Korotcenkov, G., The Role of Morphology and Crystallographic Structure of Metal Oxides in Response of Conductometric-Type Gas Sensors. Mater. Sci. Eng. R-Rep. 2008, 61 (1–6), 1-39. 44. Yu, X.; Zeng, W., Fabrication and Gas-Sensing Performance of Nanorod-Assembled SnO2 Nanostructures. J. Mater. Sci.-Mater. Electron. 2016, 27 (7), 7448-7453. 45. Wang, X.; Ren, P.; Tian, H.; Fan, H.; Cai, C.; Liu, W., Enhanced Gas Sensing Properties of SnO2: The Role of the Oxygen Defects Induced by Quenching. J. Alloy. Compd. 2016, 669, 29-37. 46. Korotchenkov, G.; Brynzari, V.; Dmitriev, S., SnO2 Films for Thin Film Gas Sensor Design. Mater. Sci. Eng. B 1999, 63 (3), 195-204. 47. Tricoli, A.; Pratsinis, S. E., Dispersed Nanoelectrode Devices. Nat. Nanotechnol. 2010, 5 (1), 5460. 48. Shimizu, Y.; Maekawa, T.; Nakamura, Y.; Egashira, M., Effects of Gas Diffusivity and Reactivity on Sensing Properties of Thick Film SnO2-Based Sensors. Sens. Actuator. B-Chem. 1998, 46 (3), 163-168. 49. Korotcenkov, G.; Brinzari, V.; Cerneavschi, A.; Ivanov, M.; Golovanov, V.; Cornet, A.; Morante, J.; Cabot, A.; Arbiol, J., The Influence of Film Structure on In2O3 Gas Response. Thin Solid Films 2004, 460 (1–2), 315-323. 50. Kida, T.; Kuroiwa, T.; Yuasa, M.; Shimanoe, K.; Yamazoe, N., Study on the Response and Recovery Properties of Semiconductor Gas Sensors Using a High-Speed Gas-Switching System. Sens. Actuator. B-Chem. 2008, 134 (2), 928-933. 51. Teleki, A.; Pratsinis, S. E.; Kalyanasundaram, K.; Gouma, P. I., Sensing of Organic Vapors by Flame-Made TiO2 Nanoparticles. Sens. Actuator. B-Chem. 2006, 119 (2), 683-690.

TOC image

ACS Paragon Plus Environment

22

Page 23 of 28

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

ACS Applied Materials & Interfaces

Figure 1. Schematic of the direct deposition of FSP-made Sb:SnO2 with a substrate-impinging nanoparticle flame. The nanoparticle network formation on the substrate is continuously monitored during preparation by measuring the in situ resistance (Ri) between the interdigitated electrodes. 179x150mm (96 x 96 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Figure 2. The in situ resistance Ri of Sb:SnO2 nanoparticle films during their deposition at 0 (diamonds), 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentrations (PSC). The 0 mM solution concentration refers to a pure solvent (i.e. metal-free) precursor. Prior to deposition, the substrate is preheated for 20 s as depicted by the vertical dashed line. 200x154mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28

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

ACS Applied Materials & Interfaces

Figure 3. Top (a-c,e-j) and cross-section (d) SEM images of Sb:SnO2 nanoparticle films deposited with 1 (top row, red), 10 (middle row, blue) and 100 mM (bottom row, green) PSCs for 1 (a,e,h), 2 (b,f,i) and 4 minutes (c,d,g,j). All images are at the same magnification as in j, except d. The particle size on the substrate was measured from these top view images (k). The evolving nanoparticle size (i.e. diameter) as a function of deposition duration for 1 (squares), 10 (circles) and 100 mM (triangles) PSC. 269x339mm (96 x 96 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Figure 4. Nanoparticle film thickness (from SEM cross-section images) as a function of flame deposition duration when using a 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentration (PSC). Insets are cross-section images of films prepared for 2 min deposition with 1 (b) and 100 mM (a) PSC. The film thickness is indicated by yellow arrows. Note the significantly different magnification of (a) and (b). Cross-section images of films prepared with 10 mM PSC can be found in Fig. S9. 215x189mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28

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

ACS Applied Materials & Interfaces

Figure 5. The baseline resistance R0 in dry air at 300 °C as a function of FSP deposition duration (a) for Sb:SnO2 nanoparticle films made with 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentration (PSC). Sensors prepared by flame-deposition with in situ annealing2 are represented as diamonds. The corresponding response towards 1 ppm of ethanol in dry air at 300 °C is shown in (b). Extrapolated data of Sb:SnO2 sensors from Wang et al.37 and Shao et al.19 are indicated by dashed lines. 250x124mm (96 x 96 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Figure 6. The Sb:SnO2 sensor response (a) and recovery times (b) as a function of nanoparticle deposition duration for films prepared with 1 (squares), 10 (circles) and 100 mM (triangles) precursor solution concentrations (PSC) as well as such made by deposition with in situ annealing2 (diamonds). 212x124mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 28 of 28