Toward Adequate Operation of Amorphous Oxide Thin-Film

Subscriber access provided by UNIV OF DURHAM

Article

Toward adequate operation of amorphous oxide thin film transistors for low concentration gas detection Kyung Su Kim, Cheol Hyoun Ahn, Sung Hyeon Jung, Sung Woon Cho, and Hyungkoun Cho ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b18657 • Publication Date (Web): 01 Mar 2018 Downloaded from http://pubs.acs.org on March 3, 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.

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 27 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

Toward adequate operation of amorphous oxide thin film transistors for low concentration gas detection Kyung Su Kim‡, Cheol Hyoun Ahn‡, Sung Hyeon Jung, Sung Woon Cho, Hyung Koun Cho* School of Advanced Materials Science and Engineering, Sungkyunkwan University, 2066 Seobu-ro, Jangan-gu, Suwon, Gyeonggi-do, 16419, Republic of Korea ‡

K. S. Kim and C. H. Ahn contributed equally to this work.

*Corresponding author: H. K. Cho; e-mail ([email protected]) Keywords: Low concentration gas detection, low-working temperature, thin-film transistors, amorphous oxide semiconductor, NO2 gas

Abstract We suggest the use of a thin-film transistor (TFT) comprised of amorphous InGaZnO (aIGZO) as a channel and sensing layer for low concentration NO2 gas detection. Although amorphous oxide layers have a restricted surface area when reacting with NO2 gas, such TFT sensors have incomparable advantages in the aspects of the electrical stability, large-scale uniformity, and the possibility of miniaturization. The a-IGZO thin films do not possess typical reactive sites, grain boundaries, so that the variation in drain current of the TFTs strictly originates from oxidation reaction between channel surface and NO2 gas. Especially, the sensing data obtained from the variation rate of drain current makes it possible to monitor efficiently and quickly the variation of the NO2 concentration. Interestingly, we found that enhancement mode-TFT (EM-TFT) allows discrimination of the drain current variation rate at NO2 concentrations ≤ 10 ppm, while a depletion mode-TFT is adequate for discriminating NO2 concentrations ≥ 10 ppm. This discrepancy is attributed to the ratio of charge carriers contributing to gas capture with respect to total carriers. This capacity for the excellent detection of low concentration NO2 gas can be realized through i) three terminal TFT gas sensors using amorphous oxide, ii) measurement of the drain current variation rate for high selectivity, and iii) an EM mode driven by tuning the electrical conductivity of channel layers.

1 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

Page 2 of 27

Introduction Portable and wearable electronics have attracted significant research interest because of their potential application in integrated intelligent devices such as consumer electronics, home healthcare, and environmental monitoring systems.1-3 Recently, renewed interest in toxic gas monitoring has increased due to the focus on environmental pollution stemming from rapid economic growth and the increased use of fossil fuels.4 In addition, the detection of toxic gas can be applied in industrial locations and for personal devices, such as for diagnosing food decay and medical issues,5,6 further accelerating the development of intelligent gas sensors as a key emerging research field. Among many toxic gases, nitrogen dioxide (NO2) that is a prominent air pollutant is one of the most interesting toxic gases, since it mainly originates from internal combustion engines burning fossil fuels, and chronic exposure can bring detrimental effect on human respiratory health.7,8 The European Commission air quality standard states that NO2 gas should not be present in concentrations in excess of 40 µg/m3 as an annual average,9 and some studies have reported that exposure to 5 ppm NO2 gas over 10 min may be dangerous to humans.10 Accordingly, the ability to detect low concentration NO2 under 5 ppm with fast response time is most likely to widen the possibility for the integrated semiconductor gas sensors. To

detect

harmful

gases

using

semiconductor

materials,

two-terminal-type

chemiresistors consisting of metal oxides were increasingly popular and their crystal morphologies including polycrystalline, porous, and particles have been investigated,11-13 as shown in Fig. 1(a). In particular, conventional two-terminal porous and particle semiconductor films are preferred for enhancing reactive sites via prompt gas diffusion, thereby improving response times, while polycrystalline films exhibit low sensitivity and can be operated at high temperature due to sluggish chemical reaction and desorption.14 In common two-electrode metal oxide sensors, reactive harmful oxidation gases such as NO2


Page 3 of 27 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


and CO are adsorbed effectively on the surface of n-type metal oxides, creating oxygen related ion molecules by attracting charged electrons, and resulting in the formation of a depleted region and potential barrier.15,16 Because the grain boundaries in polycrystal films and the enhanced surface area in nanostructures provide dangling bonds for chemical adsorption that change carrier concentrations,17 two-terminal electrode sensors simply monitor the change of electrical resistivity. However, such resistive type structures need either sufficient gas flow or time to generate a distinguishable sensing signal by improving the signal-to-noise ratio, and may use raised operating temperatures in order to promote gas diffusion into inner grain boundaries.18,19 Additionally, delayed recovery times are inevitable in order to promote chemical desorption from deeper regions.20 Therefore, integrated semiconductor sensors using these crystalline metal oxides are intrinsically limited due to their drawbacks of high power consumption as well as the use of complex crystal structures, and thus do not allow large-scale integration, provide excellent uniformity and mobile application. In contrast, three-terminal type sensors based on transistors with sufficient amplification capabilities are fundamental for mass production, low-power consumption, and miniaturization, despite the complex device structures and manufacturing processes expected of practical application to portable/wearable electronics.21,22 Several research groups have already reported NO2 gas sensing properties in three-terminal transistors composed of organic semiconductors, carbon nanotubes, graphene, or oxide nanostructures,23-27 as shown in Fig. 1(b). In addition, novel functional sensors combined with three-electrode transistors have been recently constructed, built from organic or inorganic transistors as a driving device and ionic electrolytes for chemical sensing.28,29 However, these devices combined with separated transistor and sensing element are still not enough in the aspects of device instability, poor reproducibility, and device miniaturization, even though fast response and selectivity.30



Actually, it has been addressed that amorphous InGaZnO (a-IGZO) TFTs attracting attention for display applications showed the unstable electrical properties under environmental and humid conditions.31,32 It is noteworthy that these oxide TFTs show large electrical variation under changing conditions despite using dense amorphous phase channel layers, where the back surface of channel layers in the bottom-gate configuration can actively react with the target gas. Although the sensitivity of electrical properties to a device’s external environments has been studied, little systematic investigation of gas sensors directly utilizing amorphous oxide TFTs has been performed. Most three-terminal gas sensors have focused on one- or two-dimensional field effect transistors, which are not amenable to large-scale and miniaturized production. The use of the amorphous oxide semiconductors in three-terminal gas sensors may be advantageous in terms of responsivity and selectivity as well as large-area uniformity, reproducibility, and miniaturization, as shown in Fig. 1(c). In this work, we apply a-IGZO as a sensing layer combined with three-terminal-type TFTs, where the amorphous oxide film plays the roles of the channel and sensing layers simultaneously. Since an amorphous film has no grain boundaries and is patterned, all chemical sensing is restricted to the film surface. We show that the responsivity of a-IGZO TFT-based gas sensors can be controlled by simply adjusting the conductivity of the active layers. In addition, we illustrate that highly-sensitive, three-terminal gas sensors can be constructed by monitoring the rate of current variation, as an alternative to the simple monitoring of current or resistance change, and this approach is more efficient for detecting extremely low concentrations of harmful gases. Especially, a-IGZO TFT gas sensors exhibit excellent responsivity, sensitivity, and recovery for low NO2 gas concentrations, despite operating at relatively low temperatures.


Page 4 of 27

Page 5 of 27 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


Experimental details Fabrication of TFTs. In order for the active layer to react efficiently with the target gas, aIGZO TFTs with bottom-gate configurations were fabricated on SiO2/Si substrates. To analyze the influence of active layer electrical properties on gas-sensing performance, aIGZO channel (60 nm) TFTs with different operating enhancement and depletion modes were constructed on rigid substrates. Here, a-IGZO channel layers were deposited by radiofrequency (RF) sputter (150 W, 3 mTorr, room temperature). The working modes of the TFTs were artificially managed by controlling the gas ratios [Ar/O2 = 28/2 sccm for depletion-mode (DM) TFT, Ar/O2 = 25/5 sccm for enhancement-mode (EM) TFT] during the sputtering deposition of the a-IGZO. In particular, these TFTs were designed to maintain similar transistor performance. Results are compared in Fig. S1, where all figures of merit except Vth are very similar. Active channel layers were defined by conventional photolithography and a wet-chemical etching process (HCl:DI water = 1:5), and followed by thermal annealing (350°C, air, 1 hr) in a furnace in order to stably re-arrange deposited atoms and provide defect healing. Source/drain Mo (100 nm) electrodes were deposited by direct-current magnetron sputtering (65 W, 3 mTorr) at room temperature, and were defined using a lift-off process. Channel width (W) and length (L) of the fabricated a-IGZO TFTs were 500 µm and 50 µm, respectively. Performance and gas-sensing evaluation of a-IGZO TFTs. We characterized gas-sensing performance at various NO2 gas concentrations in an isolated gas chamber, as shown in Fig. S2. After loading the a-IGZO TFTs into the chamber, it was evacuated to negate any ambient effects. Controlled air and NO2 gas were flown, and the change of drain current at various NO2 gas concentrations (0–1000 ppm) in air balance was monitored (Fig. S2). The electrical performance of a-IGZO TFTs fabricated with bottom-gate configurations was evaluated with a 4145B parameter analyzer. The field effect mobility (µFE), indicating how quickly electrons



Page 6 of 27

move through the channel layer, was calculated from the typical equation μ =

,

where, VD, ID, and VG are the drain voltage, drain current, and gate voltage, respectively. The subthreshold swing (SS) values showing the switching velocity from the off to on states of the TFTs were obtained from

, and the threshold voltage (Vth) was

determined from (ID)1/2–VG plots.

Results and discussion Unlike two-terminal-type chemiresistors showing linear I-V behavior, the I-V transfer curves of typical TFTs electrically exhibit the three states of OFF (high resistive), subthreshold, and ON (high conductive) which can be actively selected by changing the gate voltage as shown in Fig. 2(a). Generally, the Vth of the TFTs is determined by the conductivity (or carrier density) of the channel materials,33 and the modulation of Vth after exposure to target gases leads to an amplified current signal change due to the high transconductance of transistors. Since the active surface area of the a-IGZO TFT is extraordinarily smaller than that of conventional nanostructures with high aspect ratio (Fig. S3), discriminating the sensing signal from the variation of mobility or Vth is not appropriate in detecting low concentration target gases. However, the highest sensibility from the enlarged ∆I(output) that indicates a drain current change before and after gas flow is expected from the subthreshold region, and this region can be recommended for sensor application, as shown in Fig. 2(a). In particular, the TFT devices have a strategic merit to be able to select an initial current value (off gas current) by controlling a gate voltage depending on the types of the reactive target gas, such as reduction or oxidation reaction. Ideally, in order to achieve a high responsivity to low concentration harmful gases via


Page 7 of 27 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


current modulation, all a-IGZO TFT devices must possess excellent electrical performances of high ON/OFF ratio, low SS, and high mobility. Reduced SS values can result in enhanced current change in subthreshold regions by adsorbed gas molecules despite small Vth shifts. The calculated Vth, µFE, SS, and ON/OFF ratio for pristine a-IGZO TFTs, as seen in Fig. 2(b) and Fig. S1(a), were -9.2 V, 11.2 cm2/Vs, 0.36 V/dec., and 6.7×108, respectively. Figures 2(b) and (c) show the gas-sensing performance from a-IGZO TFTs at 50 ppm NO2 gas flow and room temperature. Here, the NO2 was a strong oxidation agent with electrophilic properties. The following relations describe the typical electron capture process in NO2 injection during the formation of ionic molecules:34

or

! "#$%&

+

! "#$%&

! "$(%)*+,(&

→

+ 2, "%.*/$0, )/ )12(,&

! "$(%)*+,(&

+2

"$(%)*+,(& "1&

+ , "%.*/$0, )/ )12(,& →

! "$(%)*+,(& "2&

The adsorption of NO2 gas acts as an acceptor of negative charge carriers on the surface of ntype oxide semiconductor, and the conductance is decreased by the reduction of the mobile electron density in the channel, resulting in a perceptible change in Vth. Thus, the NO2 gas concentration can be evaluated by monitoring the conductance of the oxide semiconductors. As shown in Fig. 2(b), the bottom-gate a-IGZO TFTs showed a distinguishable positive Vth shift with increasing gas exposure times at 50 ppm NO2, due to the continuous oxidation reaction on the channel surface. However, the rate of Vth shift gradually reduced, and the Vth variation nearly saturated after 70 min, as shown in the inset of Fig. 2(b). Compared with polycrystalline or nanoparticle layers, a dense a-IGZO thin film does not possess the potential barriers between crystal grains that are dominant sites for reaction with gas molecules, so that the drain current variation of a-IGZO TFTs resulted primarily from the oxidation reaction between channel surface and NO2 gas. Thus, the saturation of Vth shift is presumed to be



related that the a-IGZO back surface was completely reacted with NO2 gas after a certain period of exposure time due to the limited surface area of active layer. For oxidationproduced gas detection, a relatively conductive off-gas current in the SS region was the most effective, and thus our measurements were begun at Vg = -6 V corresponding to 1.2×10-6 A. As shown in Fig. 2(c), for a gas exposure of 50 ppm NO2, the initial current value quickly dropped and nearly saturated at ~70 min. To analyze the TFT response in detail, the real-time sensing performance of threeterminal a-IGZO transistors was characterized with NO2 gas concentrations of 0–1000 ppm in air balance, as shown in Fig. 3(a). The applied gate and drain voltages were set to VG,working = Vth + ~2.5 V (drain current ~106 A) at VD = 5 V corresponding to a subthreshold region for the high responsivity. Similarly, with increasing the exposure time of NO2 gas, the drain current is continuously decreased with different speeds for current reduction. Typically, it has been accepted that the oxide TFTs can result in Vth shift under positive/negative gate bias stress conditions, related to charge trapping at the channel/dielectric interface, within channel layers, and back-channels.35,36 Among them, a positive gate bias stress may also induce a Vth shift with the same direction as the shift in ambient NO2. However, our a-IGZO TFTs show very slight or negligible Vth variation under positive gate bias stresses unlike the sensing performance with considerable Vth shift, as seen in Fig. 3(a) and S4. For real-time sensing measurements, the drain current of a-IGZO TFTs shifts only slightly under dry air ambient conditions due to adsorption of O2 molecules on the channel surface. On the other hand, the injection of a small amount of NO2 gas caused an extraordinary change in the drain current. Therefore, it can be concluded that the origin of the decrease in drain current under ambient NO2 is entirely related to the decrease of the mobile charge carrier concentration induced from the adsorption of NO2 gas. Furthermore, after NO2 exposure time elapses, the drain current in relatively high gas injection is decreased into the off-current level of the TFTs due 8 ACS Paragon Plus Environment

Page 8 of 27

Page 9 of 27 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


to nearly complete electron trap from sufficient gas adsorption, as shown in Fig. 3(a). However, the timescale for approaching the off-current level appeared to strongly depend on the concentration of the NO2 gas. Additionally, the decrease in drain current was experimentally nonlinear with the NO2 exposure time, probably due to the reduced number of exposed surface sites available to react with NO2 over time, in turn reducing the limited surface area of the a-IGZO active layers. Since NO2 gas can affect human health even over short exposure times and at low concentrations, the development of a highly-responsive sensor is necessary. Unfortunately, the drain current continuously changed at different rates with sensing time, finally reaching a saturation value similarly to the off-current level of the TFTs. This causes an ambiguous dependency on the gas concentration and measurement time, and higher concentrations causes accelerated saturation. As a result, this measurement should be done correctly with radical caution. Consequently, for a-IGZO TFT-based gas sensors, a more efficient characterization method for quick and distinguishable response must be considered, and thus we suggest a new approach for monitoring the sensing signal from the drain current, as shown in Figs. 3(b) and (c), instead of directly comparing the drain current or transconductance. The decrease rate of drain current is strongly dependent on the concentration of reactive NO2 gas at the initial stage, due to the differing amounts of gaseous species arriving on the pristine a-IGZO surface, assumed that their sticking coefficients are identical. As a result, the real-time sensing characteristic curve from a a-IGZO TFT can be described with the drain current variation rate as a function of time under various NO2 concentrations as shown in Fig. 3(c), where the vertical axis differentiates log I as a benchmark value to compare different gas concentrations. From this, we can explicitly distinguish the change of NO2 concentrations. This plot shows steady gas flow values at the same concentration, making it possible to quickly and efficiently monitor the variation of 9 ACS Paragon Plus Environment


NO2 concentration from the oxide TFTs. Nevertheless, this plot fails to recognize the distinct differences of NO2 gas concentrations less than 10 ppm. Thus, the sensing capability of this TFT device appears to be effective with a sufficient flow of NO2 gas, but our hypothesis that three-terminal oxide TFTs can provide a fast response at low concentrations is not achieved. Typical TFT devices have two operating modes depending on their electrical characteristics, which are usually divided by the sign of Vth, with positive Vth indicating the enhancement-mode (EM) and a negative Vth indicating the depletion-mode (DM). The EMTFTs are in the OFF state at zero gate bias, which was produced by using a less-conductive channel (low carrier density). In contrast, the DM-TFTs are normally ON state at zero gate bias due to the more conductive channel properties (high carrier density). For the two types of a-IGZO TFTs, active channel layers possess considerably different electron concentrations that can be chemically captured by NO2 absorption, as shown in Fig. 4(a). In the case of the DM-TFTs, low concentration NO2 gas is insufficient for detecting distinguishable changes of electrical properties, since the small fraction of total electron carriers can contribute to the reaction with NO2 gas due to an excess electron carrier density in the active channel. This behavior is analogous to a p-n diode with a high leakage current that does not show good optical sensing performance under weak illumination due to a high background current.37,38 However, a sufficient injection of NO2 gas provides a meaningful drop of drain current and subsequently allows the discrimination of gas concentration from the variation rate of the drain current. However, EM-TFTs with a relatively low carrier density are expected to show a considerable change in mobile electron density even at low NO2 gas concentrations, and thus it is expected that current variation can be better detected. The carrier density in a-IGZO films can be easily modulated by controlling the density of oxygen-related defects,39 with the subsequent a-IGZO layer deposited at relatively high O2 partial pressure (Ar/O2=25/5 sccm) to produce the EM-TFT. As a result, the device showed typical EM-TFT characteristics with 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27 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


a positive Vth (0.8 V), as shown in Fig. S1. Figure 4(b) demonstrates more clearly the variation of NO2 concentration even at extremely low NO2 concentrations (≤ 10 ppm) from the drain current change of the EM-TFT-based gas sensor, unlike the DM-TFT that requires an NO2 concentration ≥ 10 ppm. In particular, it is noticeable that the EM a-IGZO TFT exhibited good selectivity at low concentration as well as fast response. Consequently, the use of amorphous oxide TFTs as an NO2 gas sensor is promising, as they would distinguish the optimal sensitivity range of the a-IGZO TFTs having simple control of only channel conductivity, and in particular the EM-TFT was better at rapidly detecting low concentration NO2 gas. In general, the sensitivity of conventional chemiresistor-type gas sensors has been expressed as a ratio of resistance variation ["5678 − 5:; &⁄5:; × 100 %] or drain current

variation ["A678 − A:; &⁄A:; × 100 %] after target gas injection.40,41 However, the drain current of our a-IGZO TFTs was dramatically decreased after NO2 gas exposure to the offcurrent level. Thus, the above equation is not effective in the sensing evaluation from a-IGZO TFTs, because all results converge toward 100%. Other groups studying field effect transistor (FET)-based gas sensors have represented the sensitivity with various parameters such as mobility, SS, or transconductance.42,43 However, since our approach utilized the drain current variation rate at a fixed gate voltage in order to distinguish NO2 gas concentration, the comparison of the sensitivity (S) on NO2 gas concentration from both the DM-TFT and EMTFT is performed with following equation, S=C

(D)#AE"FGH& ⁄(I − (D)#AE";J& ⁄(I K × 100 % "3& (D)#AE";J& ⁄(I

where (D)#AE"FGH& ⁄(I and (D)#AE";J& ⁄(I are the drain current variation rate for an NO2 flow and air atmosphere, respectively. These a-IGZO TFTs operating at room temperature



exhibited remarkably high sensitivities proportional to NO2 concentration and expedient discrepancy on gas concentration, as shown in Fig. 4(c). In particular, the EM-TFT devices offer good selectivity even at the lowest NO2 concentration of 0.5 ppm. It should be emphasized that the following three approaches effectively contributed to developing gas sensors that can detect low concentrations of harmful gas: i) three-terminal TFT gas sensors using amorphous oxide active layers, ii) measurement of the drain current variation rate with high selectivity at an initial gas flow, and iii) an EM mode driven by tuning the electrical conductivity of channel layers in order to select an optimal measurement range. In typical chemical gas sensors, a fast recovery behavior is also one of the practically significant key factors determining the efficiency of the sensing capability. It is widely accepted that the balance between slow gas desorption at low temperature and high-power consumption at high temperature is conflicting component for the decision-making of the proper operating temperature, considering the criteria of high sensitivity and fast recovery.44 The normal operating temperatures of gas sensors using oxide semiconductors such as ZnO and SnO2 are fairly high from 200–400°C,45 but such high temperatures are unlikely to operate the a-IGZO TFTs safely.46 After exposure to 5 ppm NO2 gas for 10 min, the recovery behavior of the a-IGZO TFT gas sensor was characterized. As shown in Fig. 5(a), the positively-shifted transfer curve did not recover to its pristine electrical characteristics under sufficient air flow or a high gate bias stress condition for a long time, although some reports suggested that high gate bias could induce effectively electrical recovery.47 The lack of recovery behavior by a high gate bias reveals that the NO2 gas is likely to form strong chemical binds, rather than simply adsorbing on the amorphous IGZO. These results implicitly indicate that such a sensing reaction makes recovery more difficult and that the supply of additional external energy may be needed. Although conventional resistor-type oxide gas sensors are known to require relatively high temperatures for fast recovery, our a12 ACS Paragon Plus Environment

Page 12 of 27

Page 13 of 27 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


IGZO TFT sensors are expected to be processed at relatively low temperatures of ~100°C because the NO2 actively reacts with the limited surface of the active layers. In addition, the a-IGZO TFTs exhibited only a slight change in their electrical characteristics, even when kept at ~100°C at length, implying that the a-IGZO TFTs are very stable at 100°C, as shown in Fig. S5. In order to assess the sensing stability of the devices, we operated the a-IGZO transistors under a repeated on/off pulse of low NO2 gas flow (5 ppm) at constant gate and drain biases (VG = 1.8 V and VD = 5 V) and directly sampled the drain current, as depicted in Fig. 5(b). The operating temperature was relatively low 100°C and gas flow is also low 5 ppm. Nevertheless, due to the high durability of a-IGZO TFTs and acute sensing performance of EM mode, the response to the current variation of the EM a-IGZO TFT sensor was reproducible over repeated testing cycles at this relatively low temperature. Although a slight increase occurred in the current level arriving at the gas-on status, no clear change of sensing shape was demonstrated in the period and off-current level, as seen in Fig. 5(b). The reproducible behavior observed at 100°C is considered to be attributed that basic sensing occurs chemically on the patterned surface area of the channel layers and sensing materials are amorphous thin film without grain boundaries. The typical grain boundaries related gas sensing cause gas diffusion inside the deep film, and conversely inevitably require higher thermal energy for the chemical recovery, but the dense interior of amorphous oxide films does not allow effective chemical reactions.

Conclusions In summary, we systematically investigated chemical sensing characteristics for the NO2 gas of three-terminal amorphous IGZO TFTs that were promising platform materials for 13 ACS Paragon Plus Environment


Page 14 of 27

backplane devices in next-generation displays. The a-IGZO TFTs exhibited a perceptible positive shift in Vth via an oxidation reaction in ambient NO2 gas. However, these TFT gas sensors had restricted surface areas on which the NO2 gas could be adsorbed, so that the Vth shift was insufficient for distinguishing low NO2 gas concentrations, and they quickly saturated at high NO2 concentrations. As an alternative, we suggested that the drain current variation rate in the subthreshold region exhibited remarkably high response rates, providing a novel method for highly-sensitive measurement as a function of NO2 gas concentration. Additionally, the optimally-sensitive region from the a-IGZO TFTs was divided into two categories, depending on the carrier concentration of its active layer. For low and high NO2 gas concentrations near 10 ppm, an EM-TFT with low carrier concentration and a DM-TFT with high carrier concentration, respectively, were more suitable as determined from an optimal combination of carrier and gas concentrations. Finally, the EM a-IGZO TFT exhibited good reproducible behavior at 100°C that may be restricted to the surface of the amorphous structure of sensing materials. The results herein show that sensing recovery at relatively low temperature provides a promising choice of polymer substrates with temperature durability near ~100°C, which may be adopted for fabricating such devices for use as flexible or patchable gas sensors.

Acknowledgments This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (Grant No. NRF-2016R1C1B2012007, and by the Basic

Research

Laboratory

project

of

the

Korea

government

(Grant

No.

2014R1A4A1008474). This research was supported by the Human Resources Program in Energy Technology of the Korea Institute of Energy Technology Evaluation and Planning


Page 15 of 27 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


(KETEP) of the Ministry of Trade, Industry and Energy, Republic of Korea (Grant No. 20174030201800). This work was supported by the Technology Innovation Program (or Industrial Strategic Technology Development Program) (10079974, Development of core technologies on materials, devices, and processes for TFT backplane and light emitting frontplane with enhanced stretchability above 20%, with application to strechable display) funded By the Ministry of Trade, Industry & Energy (MOTIE, Korea)

References (1) Liu, X.; Cheng, S.; Liu, H.; Hu, S.; Zhang, D.; Ning, H. A Survey on Gas Sensing Technology. Sensors 2012, 12, 9635-9665. (2) Pantelopoulos, A.; Bourbakis, N. G. A Survey on Wearable Sensor-Based Systems for Health Monitoring and Prognosis. IEEE Trans. Syst. Man Cybern. C Appl. Rev. 2010, 40, 1-12. (3) Tsow, F.; Forzani, E.; Rai, A.; Wang, R.; Tsui, R.; Mastroianni, S.; Knobbe, C.; Gandolfi, A. J.; Tao, N. J. A Wearable and Wireless Sensor System for Real-Time Monitoring of Toxic Environmental Volatile Organic Compounds. IEEE Sens. J. 2009, 9, 1734-1740. (4) Chatterjee, S. G.; Chatterjee, S.; Ray, A. K.; Chakraborty, A. K. Graphene-Metal Oxide Nanohybrids for Toxic Gas Sensor: A Review. Sens. Actuators, B 2015, 221, 1170-1181. (5) Ryabtsev, S. V.; Shaposhnick, A. V.; Lukin, A. N.; Domashevskaya, E. P. Application



of Semiconductor Gas Sensors for Medical Diagnostics. Sens. Actuator B 1999, 59, 26-29. (6) Turner, A. P. F.; Magan, N. Innovation: Electronic Noses and Disease Diagnostics. Natur Rev, Microbiol 2004, 2, 161-166. (7) Belanger, K.; Gent, J. F.; Triche, E. W.; Bracken, M. B.; Leaderer, B. P. Association of Indoor Nitrogen Dioxide Exposure with Respiratory Symptoms in Children with Asthma. Am. J. Resp. Crit. Care Med. 2006, 173, 297-303. (8) Balazsi, C.; Sedlackova, K.; Liobet, E.; Ionescu, R. Novel Hexagonal WO3 Nanopowder with Metal Decorated Carbon Nanotubes as NO2 Gas Sensor. Sens. Actuator, B 2008, 133, 151-155. (9) http://ec.europa.eu/environment/air/quality/standards.htm. (10) http://www.inspectapedia.com. (11) Barsan, N. & Weimar, U. Conduction Model of Metal Oxide Gas Sensors. J. Electroceram. 2001, 7, 143-167. (12) Korotcenkov, R. Metal Oxides for Solid-State Gas Sensors: What Determines Our Choice? Mater. Sci. Eng. B 2007, 139, 1-23. (13) Wang, C.; Yin, L.; Zhang, L.; Xiang, D.; Gao, Rui. Metal Oxide Gas Sensors: Sensitivity and Influencing Factors. Sensors 2010, 10, 2088-2106. (14) Shimizu, Y.; Jono, A.; Hyodo, T.; Egashira, M. Preparation of Large Mesoporous SnO2 Powder for Gas Sensor Application. Sens. Actuator B 2005, 108, 56-61. (15) Seiyama, T.; Kato, A.; Fujiishi, K.; Nagatani, M. A New Detector for Gaseous Components Using Semiconductive Thin Films. Anal. Chem. 1962, 34, 1502.


Page 16 of 27

Page 17 of 27 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


(16) Barsan, N.; Koziej, D.; Weimar, U. Metal Oxide-Based Gas Sensor Research: How to? Sens. Actuator B 2007, 121, 18-35. (17) Karunagaran, B.; Uthirakumar, P.; Chung, S. J.; Velumani, S.; Suh, E. K. TiO2 Thin Film Gas Sensor for Monitoring Ammonia. Mater. Charact. 2007, 58, 680-684. (18) Patil, S. J.; Patil, A. V.; Dighavkar, C. G.; Thakare, K. S.; Borase, R. Y.; Nandre, S. J.; Deshpande, N. G.; Ahire, R. R. Semiconductor Metal Oxide Compounds Based Gas Sensors: A literature Review. Front. Mater. Sci. 2015, 9, 14-37. (19) Afzal A.; Cioffi, N.; Sabbatini, L.; Torsi, L. NOx Sensors based on Semiconducting Metal Oxide Nanostructures: Progress and Perspectives. Sens. Actuator B 2012, 171172, 25-42. (20) Shen, Y.; Yamazaki, T.; Liu, Z.; Liu, Z.; Meng, D.; Kikuta, T.; Nakatani, N. Influence of Effective Surface Area on Gas Sensing properties of WO3 Sputtered Thin Films. Thin Solid Films 2009, 517, 2069-2072. (21) Nomura, K.; Ohta, H.; Takagi, A.; Kamiya, T.; Hirano, M.; Hosono, H. RoomTemperature Fabrication of Transparent Flexible Thin-Film Transistors Using Amorphous Oxide Semiconductors. Nature 2004, 432, 488-492. (22) Kang, W. J.; Kim, K. S.; Ahn, C. H.; Cho, S. W.; Kim, D. E.; Kim, B. R.; Cho, H. K.; Kim, Y. S. Non-Ideal Current Drop Behavior in Ultra-Thin Inorganic a-InGaZnO Thin Film Transistors. J. Mater Sci: Mater Electron 2017, 28, 8231-8237. (23) Zhang, C.; Chen, P.; Hu, W. Organic Field-Effect Transistor-Based Gas Sensors. Chem. Soc, Rev. 2015, 44, 2087-2107. (24) Wang, Z.; Huang, L.; Zhu, X.; Zhou, X.; Chi, L. An Ultrasensitive Organic



semiconductor NO2 Sensor Based on Crystalline TIPS-Pentacene Films. Adv. Mater. 2017, 29, 1703192 (25) Hu, P.; Zhang, J.; Wang, Z.; O’Neill, W.; Estrela, P. Carbon Nanostructure-based Field-effect Transistors for Label-free Chemical/biological Sensors. Sensors 2010, 10, 5133-5159. (26). Zhang, D. H.; Liu, Z. Q.; Li. C.; Tang, T.; Liu, X. L.; Han, S.; Lei, B.; Zhou, C. W. Detection of NO2 down to ppb Levels using Individual and Multiple In2O3 Nanowire Devices. Nano Lett. 2004, 1919-1924. (27) Andringa, A.; Piliego, C.; Katsouras, I.; Blom, P. W. M.; Leeuw, D. M. NO2 Detection and Real-time Sensing with Field-Effect Transistors. Chem. Mater. 2014, 26, 773-785. (28) Nakato, S.; Shimanoe, K.; Miura, N.; Yamazoe, N. Field Effect Transistor Type NO2 Sensor Combined with NaNO2 auxiliary phase, Sens. Actuator B 2011, 77, 512-561. (29) Butth, F.; Kumar, D.; Stutzmann, M.; Garrido, J. A. Electrolyte-gated Organic Field-effect Transistors for Sensing Applications. Appl. Phys. Lett. 2011, 98, 15332. (30) Korotcenkov, G. Handbook of Gas Sensor Materials. Springer 2014, 209-222. (31) Raja, J.; Jang, K.; Balaji, N.; Hussain, S. Q.; Velumani, S.; Chatterjee, S.; Kim, T.; Yi, J. Aging Effects on the Stability of Nitrogen-Doped and Un-Doped InGaZnO Thin-Film Transistors. Mater. Sci. Semicond. Process. 2015, 37, 129-134. (32) Kim, K. S.; Ahn, C. H.; Kang, W. J.; Cho, S. W.; Jung, S. H.; Yoon, D. H.; Cho, H. K. An All Oxide-Based Imperceptible Thin-Film Transistor with Humidity Sensing Properties. Materials 2017, 10, 530.


Page 18 of 27

Page 19 of 27 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


(33) Cho, S. W.; Kim, D. E.; Kim, K. S.; Jung, S. H.; Cho, H. K. Toward Environmentally Stable Solution-Processed Oxide Thin-Film Transistors: A RareMetal-Free Oxide-Based Semiconductor/Insulator Heterostructure and Chemically Stable Multi-Stacking. J. Mater. Chem. C 2017, 5, 10437-10670. (34) Belysheva, T. V.; Bogovtseva, L. P.; Kazachkov, E. A.; Serebryakova, N. V. Gassensing Properties of doped In2O3 Films as Sensors for NO2 in Air, J. Anal. Chem. 2003, 58, 583-587. (35) Shankar, P.; Rayappan, J. B. B. Gas Sensing Mechanism of Metal Oxides: The Role of Ambient Atmosphere, Type of Semiconductor and Gases - A Review. Sci. Lett. J. 2015, 4, 1-18. (36) Sung, S. Y.; Choi, J. H.; Han, U. B.; Lee, K. C.; Lee, J. H.; Kim, J. J.; Lim, W.; Pearton, S. J.; Norton, D. P.; Heo, Y. W. Effects of Ambient Atmosphere on the Transfer Characteristics and Gate-Bias Stress Stability of Amorphous IndiumGallium-Zinc Oxide Thin-Film Transistors. Appl. Phys. Lett. 2010, 96, 102107. (37) Gorrn, P.; Holzer, P.; Riedl, T.; Kowalsky, W.; Wang, J.; Weimann, T.; Hinze, P.; Kipp, S. Stability of Transparent Zinc Tin Oxide Transistors under Bias Stress. Appl. Phys. Lett. 2007, 90, 063502. (38) Ohba, S.; Naakai, M.; Ando, H.; Hanamura, S.; Shimada, S.; Satoh, K.; Takahashi, k.; Kubo, M.; Fujita, T. MOS Area Sensor: Part ІІ – Low-Noise MOS Area Sensor with Antiblooming Photodiodes. IEEE Trans. Electron Devices 1980, ED-27, 16821687. (39) Fu, X.; Yao, S.; Xu, J.; Lu, Y.; Zheng, Y. Study on High Signal-to-Noise Ratio (SNR) Silicon P-N Junction Photodetector. Opt. Appl. 2006, 36, 421-428.



(40) Yao, J.; Xu, N.; Deng, S.; Chen, J.; She, J.; Shieh, H. P. D.; Liu, P. T.; Huang, Y. P. Electrical and Photosensitive Characteristics of a-IGZO TFTs Related to Oxygen Vacancy. IEEE Trans. Electron Devices 2011, 58, 1121-1126. (41) Choi, Y. J.; Hwang, I. S.; Park, J. G.; Choi, K. J.; Park, J. H.; Lee, J. H. Novel Fabrication of An SnO2 Nanowire Gas Sensor with High Sensitivity. Nanotechnology 2008, 19, 095508. (42) Kwon, O. S.; Hong, J. Y.; Park, S. J.; Jang, Y.; Jang, J. Resistive Gas Sensors Based on Precisely Size-Controlled Polypyrrole Nanoparticles: Effects of Particle Size and Deposition Method. J. Phys. Chem. C 2010, 114, 18874-18879. (43) Torsi, L.; Dodabalapur, A.; Sabbatini, L.; Zambonin, P. G. Multi-Parameter Gas Sensors Based on Organic Thin-Film-Transistors. Sens. Actuators, B 2000, 67, 312316. (44) Lu, G.; Ocola, L. E.; Chen, J. Reduced Graphene Oxide for Room-Temperature Gas Sensors. Nanotechnology 2009, 20, 445502. (45) Chang, J. F.; Kuo, H. H.; Leu, I. C.; Hon, M. H. The Effects of Thickness and Operation Temperature on ZnO:Al Thin Film CO Gas Sensor. Sens. Actuators B 2002, 84, 258-264. (46) Kohl, D. Function and Applications of Gas Sensors. J. Phys. D: Appl. Phys. 2001, 34, 125-149. (47) Takechi, K.; Nakata, M.; Eguchi, T.; Yamaguchi, H.; Kaneko, S. TemperatureDependent Transfer Characteristics of Amorphous InGaZnO4 Thin-Film Transistors. Jpn. J. Appl. Phys. 2009, 48, 011301.


Page 20 of 27

Page 21 of 27 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 1. Classification of terminals (a point of connection to external circuits) and material geometries for typical semiconductor-based sensor devices, and their characteristics from the point of view of sensing performances. 500x310mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 22 of 27

Page 23 of 27 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. (a) Variation of transfer characteristics from the a-IGZO TFT under 50 ppm NO2 gas concentration and the drain current change (before and after NO2 gas flow) depending on VG. Here, typical TFT exhibits three regions of OFF (high resistive), subthreshold, and ON (high conductive), and the highest sensibility indicating a change in the drain current of before and after gas flow is the subthreshold region. (b) Variation of transfer curves from the a-IGZO TFT under 50 ppm NO2 gas concentration (VD = 5 V, 120 min) and Vth shift as a function of time. (c) Variation of drain current from the a-IGZO TFT under 50 ppm NO2 gas concentration at VD = 5 V, VG = -6 V. 419x534mm (300 x 300 DPI)



Figure 3. (a) Real-time sensing performances of a-IGZO TFTs with NO2 gas concentration of the 0–1000 ppm in air balance. (b) Dependency of reactive NO2 gas concentration on typical drain current decrease rate of oxide TFTs. (c) Drain current variation rate from a-IGZO TFTs as a function of time under various NO2 concentrations. All data were obtained at VG,working = Vth + ~2.5 V for an initial drain current of ~106 A and VD = 5 V. 510x963mm (300 x 300 DPI)


Page 24 of 27

Page 25 of 27 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. (a) Schematic models for the reason showing different NO2 gas sensing performances from the EM- and DM-TFTs under the low and high gas concentrations. (b) Drain current variation rate from an EMTFT with high sensitivity under various NO2 concentrations (VG = 3 V and VD = 5 V). (c) Sensitivities proportional to NO2 concentration for the EM- and DM-TFTs. 489x440mm (300 x 300 DPI)



Figure 5. (a) Transfer curves showing recovery behavior via different methods from a-IGZO TFTs. Here, a relatively low temperature of ~100 °C resulted in complete recovery. (b) Sensing performance of an EM aIGZO TFT under repeated on/off pulse of low NO2 gas flow (5 ppm) at constant gate and drain bias (VG = 1.8 V and VD = 5 V) at 100 °C. 419x550mm (300 x 300 DPI)


Page 26 of 27

Page 27 of 27 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


Table of Contents/Abstract Graphic 17x7mm (600 x 600 DPI)


Toward Adequate Operation of Amorphous Oxide Thin-Film

Recommend Documents