Boosting Carrier Mobility in Zinc Oxynitride Thin-Film Transistors via

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Functional Inorganic Materials and Devices

Boosting Carrier Mobility in Zinc Oxynitride ThinFilm Transistors via Tantalum Oxide Encapsulation Taeho Kim, Min Jae Kim, Jiwon Lee, and Jae Kyeong Jeong ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b03865 • Publication Date (Web): 02 Jun 2019 Downloaded from http://pubs.acs.org on June 5, 2019

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

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

Page 1 of 30 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

Boosting Carrier Mobility in Zinc Oxynitride Thin-Film Transistors via Tantalum Oxide Encapsulation Taeho Kim, Min Jae Kim, Jiwon Lee, and Jae Kyeong Jeong* Department of Electronic Engineering, Hanyang University, Seoul 133-791, South Korea AUTHOR EMAIL ADDRESS: J. K. Jeong ([email protected]) KEYWORDS: Zinc oxynitride; tantalum oxide; thin-film transistor; encapsulation; scavenge effect; device stability

ABSTRACT

Novel TaOx encapsulation was presented to enhance the field-effect mobility (FE) of ZnON thin-film transistors (TFTs), consisting of a metallic Ta film deposited onto the ZnON surface followed by a modest annealing process. The resulting TaOx/ZnON film stack exhibited a more uniform distribution of nanoscale ZnON crystallites with increased stoichiometric anion lattices compared to the control ZnON film. The control ZnON TFTs exhibited a reasonable FE, subthreshold gate swing (SS), and ION/OFF ratio of 36.2 cm2/Vs, 0.28 V/decade, and 2.9  108, respectively. A significantly enhanced FE value of 89.4 cm2/Vs was achieved for ZnON TFTs with a TaOx encapsulation layer, whereas SS of 0.33 V/decade and ION/OFF ratio of 8.6  108 were comparable to those of the control device. This improvement could be explained by scavenging and passivation effects of the TaOx film on the ZnON channel layer. Densityof-state (DOS)-based modeling and simulation was performed to obtain greater insight with regard to increasing the performance of the ZnON TFTs with a TaOx encapsulation layer. A smaller number of subgap states near the conduction band (CB) minimum and a higher net carrier density for the TaOx capped 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 30

device increased the Fermi-energy level toward the CB edge under thermal equilibrium conditions, leading to efficient band conduction and fast carrier transport under the on-state condition.

T. Kim and M. J. Kim contributed equally to this work.

2 ACS Paragon Plus Environment

Page 3 of 30 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. INTRODUCTION Amorphous oxide semiconductor (AOS) materials such as zinc oxide (ZnO) and indium gallium zinc oxide (IGZO) have attracted considerable attention as an alternative to current silicon-based semiconductors for use in liquid crystal (LC) and organic light-emitting diode (OLED) screen applications due to their intriguing properties including a high mobility, good switching modulation ratio, transparency, and excellent uniformity.1 Rapid advances with regard to IGZO materials and processes since its discovery in 2004 have led to the commercialization of IGZO thin-film transistors (TFTs) for ultra-lowpower LC mobile phones and large-size high-resolution OLED televisions.2-4 However, the notorious instability of AOS TFTs under gate bias and illumination stress conditions is still a big concern where oxygen vacancy defects (VO) in AOS materials are responsible for a huge negative shift in threshold voltage (VTH).5-9 Recently, zinc oxynitride (ZnON) materials were proposed as a promising n-type semiconductor due to its low electron effective mass and high carrier mobility.10-12 Optoelectronic properties such as its optical band-gap (Eg), effective carrier mobility, and density of the ZnON thin films can be tailored by controlling the anion ratio of wurtzite ZnO (Eg ~3.2 eV) and cubic Zn3N2 (Eg ~1.0 eV).11 In ZnO, persistent photoconductivity (PPC) after prolonged light exposure is attributed to the photo-ionization of neutral VO to metastable VO2+.13 The substitution of oxygen anions with nitrogen anions in ZnO caused the band-gap to narrow gradually, which could be explained by the superposition of the nitrogen 2p band above the oxygen 2p band at the valence band (VB) edge.11 Thus, neutral VO defects at ~1.0 eV above the VB edge of ZnO could be buried in ZnON, which strongly mitigated the PPC effect and negative bias illumination stress (NBIS)-induced VTH instability in ZnO and AOS TFTs.11,14 Owing to these attractive benefits, the effects of anion composition, annealing temperature on the structural, chemical, and electrical properties of ZnON thin films have been intensively studied, rendering remarkably high field-effect mobilities (> 50 cm2/Vs) and excellent ION/OFF ratios (>107) in the ZnON TFTs.15-17 Despite these merits, the chemical 3 ACS Paragon Plus Environment


Page 4 of 30

instability of ZnON materials is still an open question to be resolved prior to their facile implementation within commercial electronics where nitrogen-related defects and out-diffusion due to rather weak Zn-N chemical bonding invoke the deterioration of electrical properties and long-term stability.18 In this study, a new scavenge route was suggested to improve the chemical and electrical stabilities of n-channel ZnON thin films, consisting of the deposition of metallic tantalum films onto sputtered ZnON followed by subsequent thermal annealing. ZnON TFTs fabricated by this novel encapsulation method exhibited a high field-effect mobility of 89.4 cm2/Vs, low subthreshold gate swing of 0.33 V/decade, good ION/OFF ratio of 8.6  108, and long-term storage stability.

2. EXPERIMENTAL ZnON TFTs with an inverted staggered bottom-gate and top-contact structure were fabricated on Si substrates. A 100 nm-thick SiO2 gate insulator layer was grown via thermal oxidation onto a heavilydoped p-type Si wafer. The heavily-doped Si substrate acted as the gate electrode. A 20 nm-thick ZnON channel layer was deposited via a reactive sputtering process where the Ar/O2/N2 reactive gas ratio, DC power, and chamber pressure during channel preparation were 5/2.5/30, 100 W, and 5 mTorr, respectively. An ITO film was used as the source/drain (S/D) electrode and was deposited onto the ZnON/SiO2/Si substrate via magnetron sputtering. The DC power and chamber pressure under an Ar atmosphere were 50 W and 5 mTorr, respectively. The ZnON channel island and ITO S/D electrode were patterned through a shadow mask during sputter deposition. The width (W) and length (L) values of the ZnON TFTs were 1000 μm and 300 μm, respectively. The devices were then contact-annealed in an electric furnace for 1 h at 250 C under an air atmosphere (hereafter referred to as the control device). A 10 nm-thick Ta film was selectively deposited via DC sputtering through a shadow mask onto the ZnON film between the source and drain electrode with dimensions of W/L = 2300 m / 150 m (Figure 1). A


Page 5 of 30 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


final post-deposition annealing (PDA) step was performed at 200 C for 1 hour under O2 ambient, which will be referred to as Ta-capped ZnON TFTs.

Figure 1. Schematic diagram of Ta capped ZnON TFTs fabricated onto a highly doped p-type silicon wafer. The structural properties of the ZnON films were evaluated via grazing incidence X-ray diffraction (GIXRD, X’Pert PRO, PANalytical, Egham, Surrey, UK) using Cu K radiation and high-resolution transmission electron microscopy (HRTEM, Tecnai F20, FEI, Hillsboro, Oregon, USA). The chemical states of the ZnON films were examined via X-ray photoelectron spectroscopy (XPS, SIGMA PROBE, Thermo VG, Waltham, Massachusetts, USA). For XPS analysis, ZnON films were inserted into the XPS chamber within 2 hours after PDA processing at 250 C. The back surface region (~5 nm) of the ZnON films was plasma etched in situ within the XPS chamber to remove surface contamination. The N and O 1s XP spectra were deconvoluted into three asymmetric Lorentzian-Gaussian peaks with Shirley background. Curve-fitting for all spectra were performed with fixed full width at half-maximum and constant Gaussian to Lorentzian ratio parameters while allowing variability with regard to the peak 5 ACS Paragon Plus Environment


Page 6 of 30

binding energies and areas. The electrical characteristics of the TFTs were measured using a Keithley 2636 source parameter at room temperature under air ambient, and a two-dimensional device simulator ATLAS (Silvaco, Silvaco International, Santa Clara, CA, USA) was used to evaluate the density-of-state (DOS) distribution under the conduction band minimum (EC) and was used to calculate the energy band diagram of the ZnON and Ta-capped ZnON TFTs.

3. RESULTS AND DISCUSSION Figure 2 shows the GIXRD patterns of the control and Ta-capped ZnON films on SiO2/Si substrates. With regard to the control ZnON film, thermal annealing was performed at 250 C under an air atmosphere. The control ZnON film exhibited no noticeable peaks except for 51.4 (due to the Si substrate), suggesting its amorphous or nanocrystal nature.14 The Ta-capped ZnON samples were subjected to a two-step annealing process consisting of first annealing at 250 C under an air atmosphere prior to the deposition of Ta film and a second PDA at a different temperature of 100, 200, 300, and 600 C under an O2 atmosphere after the deposition of Ta film to investigate Ta-induced microstructure evolution of the underlying ZnON films. The Ta-capped ZnON samples after the second PDA at 100 and 200 C exhibited discernible peaks at 38.2, which could be attributed to the (202) reflection of the metallic -Ta (JCPDS card NO. 25-1280) phase. This metallic -Ta phase disappeared for the Ta-capped ZnON sample after the second PDA at 300 C, indicating the metallic Ta film was converted to an amorphous TaOx film via thermal oxidation.19 Interestingly, a new peak near 35.1 for this sample was observed. It is well-known that crystalline Zn3N2 possesses an anti-bixbyite structure with a lattice constant of 9.788 Å, belonging to the body-centered cubic (bcc) system (space group Ia3).20 In the case of a high purity Zn3N2 crystal, a (321) peak is expected to appear at approximately 34.3, which was quite close to that observed (35.1). Considering that the anion O2- ions substituted with sub-lattice sites in the Zn3N2 crystal, this peak could be assigned to (321) of the cubic system. A smaller lattice constant of 9.540 Å calculated from the 6 ACS Paragon Plus Environment

Page 7 of 30 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


observed (321) reflection position could be understood by the fact that the O2- ion exhibited a smaller radius than the N3- ion. This suggested that the control ZnON sample consisted of aggregated Osubstituted Zn3N2 nanocrystallites. By contrast, the ZnON sample after a high PDA temperature of 600 C exhibited a well-defined polycrystalline wurtzite ZnO structure (JCPDS card NO. 36-1451), regardless of the existence of a Ta capping layer (Figure 2). The Gibbs free energies of formation (ΔGf) for Zn3N2 and ZnO were calculated to be 156.4 kJ/mole and -262.3 kJ/mole,21 respectively at the elevated temperature of 600 C. The negative ΔGf value of ZnO means that ZnO is a thermodynamically stable phase. In contrast, the Zn3N2 is a highly unstable state, considering its positive ΔGf value. Thus, nitrogen will be dissociated from Zn-N lattice and subsequently completely diffused out from the ZnON channel layer through the thin TaOx layer at high annealing temperature of 600 C, leading to the conversion from ZnON to ZnO. Indeed, this conversion from Zn3N2 (or ZnON) to ZnO at high temperatures (> 500 °C) has been frequently reported in the literature.22, 23 Owing to this reason, characterization of ZnON films without/with a Ta capping layer focused on a relative low temperature PDA of 200 C. When the annealing process was performed at the low temperature ( 200 C), this kind of conversion can be kinetically prevented by its low available thermal energy. Because the contribution of an amorphous region possibly existing in the ZnON films (in the case of PDA temperatures less than 250 C) was not reflected in the XRD analysis,14, 24 HRTEM analysis was performed to reveal the entire microstructure of the ZnON films.



Figure 2. GIXRD spectra of the ZnON/SiO2/Si stack with and without a Ta layer. A Ta layer film was annealed at various temperatures under atmospheric O2.


Page 8 of 30

Page 9 of 30 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 3. Cross-sectional TEM images of the (a) control ZnON film and (b) Ta-capped ZnON film. Fast Fourier Transform (FFT) patterns of the selected area were shown in the right figures: A (C) and B (D) were obtained from the center and bottom region of control ZnON film (Ta-capped ZnON film), respectively.

Figure 3 shows HRTEM images of the control and Ta-capped ZnON samples annealed at 200 C for 1 hour under O2 ambient. The periodic lattice image in Figure 3a clearly indicates that the control ZnON film is in the crystalline phase. Evolution of the selective area diffraction patterns (SADPs) yielded variations with regard to the degree of crystallization along the depth direction of the control ZnON films. ZnON films near the SiO2/Si substrate exhibited rather diffused hallow SADPs, suggesting a lack of 9 ACS Paragon Plus Environment


Page 10 of 30

highly ordered cations and anions. However, the SADPs of the ZnON films exhibited discernible Bragg spots with the growing direction, indicating that crystalline quality improved with an increase in deposition time. Interestingly, the depth-dependent evolution of SADPs disappeared for Ta-capped ZnON films as can be seen in Figure 3b. The Ta-capped ZnON films exhibited better uniformity in terms of the degree of crystallization along the depth direction compared to the control ZnON films. The SADPs were identified as (400), (222), and (044) reflections of crystalline Zn3N2 with an anti-bixbyite structure, corresponding to an inter-planar spacing of 2.42 Å, 2.79 Å, and 1.71 Å, respectively. These values were smaller than those of conventional cubic Zn3N2 crystals by about 0.1% because Zn-N possesses a larger bond length than Zn-O under an identical coordination number of six (bond lengths of Zn-N and Zn-O were 2.11 Å and 1.91 Å, respectively).25, 26 Therefore, the Ta-capped ZnON film can be regarded as an aggregate of O-substituted Zn2N3 nanocrystallites. In our previous research, a chemical reaction with a transition Ta layer enabled low-temperature crystallization of the amorphous IGZO film.19 The reduction in onset crystallization temperature of IGZO from > 600 C (without catalytic layer) to 300 C was attributed to the donation of electrons from Ta into IGZO and subsequent facile rearrangement due to the change in the bonding characteristics of the IGZO. Similar Ta-induced crystallization can also occur in ZnON system. Indeed, the metallic Zn and VO was found to exist near the TaOx/ZnON interface after 2nd PDA at 200 C as shown in XP spectra of Zn 2p (Figure S1) and O 1s (Figure S2) (also see Table S1), indicating that the Zn-O bond is dissociated by the oxidation of Ta layer during PDA step. In the case of nitrogen, it was difficult to analyze its chemical states near the TaOx/ZnON interface because of the overlapping between Ta 4p3/2 and N 1s (see Figure S3). Once the nuclei were formed near the TaOx/ZnON interface, the crystalline region grew and propagated in the depth direction. It constitutes the reason for the better uniformity and crystallinity of Ta-capped ZnON compared to the control ZnON films. Considering carrier transport under the TFT on-state mainly occurred in the ZnON region near the ZnON/SiO2 interface, it was anticipated that the Ta-capped ZnON films would exhibit promising carrier 10 ACS Paragon Plus Environment

Page 11 of 30 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


mobilities due to an enhanced ordering of ZnON lattices. The disparity of the structural analysis result between XRD and TEM can arise from the different scattering power of two methods. The discernible peak for the control and Ta-capped ZnON films annealed at 200 C could not be observed in XRD results because i) the thickness of ZnON films is so thin (~ 20 nm) and ii) the ZnON films had either a poorly crystallized structure (in case of control ZnON film) or the rather random orientation with nanoscale grain size (Ta-capped ZnO film). On the other hand, the scattering power of an electron is approximately 104 times larger than that of an X-ray,27 which is the reason for the better resolution of TEM compared to XRD.

Figure 4. XPS depth profile of the (a) control ZnON film and (b) Ta-capped ZnON film. Figure 4 shows the depth profiles of zinc, oxygen, and nitrogen in the control and Ta-capped ZnON films annealed at 200 C for 1 hour under O2 ambient. In the control ZnON film, constituent elements such as zinc, oxygen, and nitrogen were uniformly distributed along the depth direction except for the surface area. It could also be seen that the anion compositions of oxygen and nitrogen were approximately 11 ACS Paragon Plus Environment


Page 12 of 30

30 and 10 at.%, respectively. By contrast, the Ta-capped ZnON film exhibited inter-diffusion of tantalum and zinc atoms whereas the distribution of oxygen and nitrogen anions was similar to that of the control ZnON film. Oxygen composition of ~10 at.% in Ta film indicated that the metallic tantalum film was partially oxidized to TaOx during the 2nd PDA at 200 C under an O2 atmosphere. Observation of a metallic Ta-related peak in the XRD analysis suggested that the partial oxidation occurred effectively near a grain boundary of Ta film not the entire Ta region due to greater diffusivity of oxygen species along twodimensional defects. The rapid concentration gradient of oxygen near the TaOx film suggested a supply of oxygen molecules from atmospheric O2. The chemical states of the different ZnON films were examined via XPS. Figure 5a, b shows the N 1s XP spectra of the control and Ta-capped ZnON films. Photoelectron binding energies were calibrated to the C 1s peak for C-C bonds at 284.5 eV. The N 1s peak was deconvoluted into three peaks at 395.2, 396.0, and 397.4 eV. The N 1s peaks centered at 395.2 and 396.0 were assigned to the Zn-N bonding energies for stoichiometric Zn3N2 and non-stoichiometric ZnxNy, respectively. The peak at 397.4 eV originated from N-N chemical bonds.12-20, 24, 28-31 The increase (decrease) in intensity of stoichiometric Zn3N2 (non-stoichiometric ZnxNy) for the Ta-capped ZnON films suggested that the electrical transport properties could be substantially improved for transistors with this channel layer because point defects such as VN in non-stoichiometric ZnxNy could act as potential electron trapping centers and disconnect an electron conduction pathway.11, 24, 32, 33 The O 1s XPS spectra for both ZnON films were also compared and depicted in Figure 5c, d. The asymmetric O 1s peaks were deconvoluted into three bases centered at 530.5, 531.8, and 532.8 eV. The peaks could be attributed to oxygen bonded to fully coordinated Zn ions (Zn-O lattice), under-coordinated Zn ions (oxygen vacancy, VO), and impurity related oxygen such as hydroxyl groups (OH), respectively.29,

34

Compared to the

control ZnON film, Ta-capped ZnON films exhibited less OH fractions as summarized in Table 1. It was noted that oxygen substituted at nitrogen sites of stoichiometric Zn3N2 was a shallow electron donor.35 Thus, the free electron density (Ne) for the Ta-capped ZnO film was expected to be higher than that of the 12 ACS Paragon Plus Environment

Page 13 of 30 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


control ZnON film, which will be discussed later. It was obvious that the metal capping and subsequent annealing process was an effective method to reduce anion-related defect centers such as VN as well as OH-related impurities within the ZnON films. It is also noted that Ta ions significantly diffused into the ZnON film during 2nd PDA as shown in Figure 4b. The Ta ions have the different valence states such as +1, +2, +4, and +5 (see Figure S4). These Ta ions with different valence states can capture the free electron carriers depending on EF level, which can degrade the gate swing capability of the fabricated transistor.

Figure 5. N 1s XP spectra of the (a) control ZnON film and (b) Ta-capped ZnON film. O 1s XP spectra of the (c) control ZnON film and (d) Ta-capped ZnON film.

Table 1. Relative Chemical Composition of the Control and Ta-capped ZnON films Obtained via XPS Analyses. (Each Fraction was Calculated as the Area Ratio of the Gaussian Distribution Centered on Each Binding Energy to the Total Area.) 13 ACS Paragon Plus Environment


N 1s

O 1s

Page 14 of 30

Chemical states

Control ZnON

Ta-Capped ZnON

Non-stoichiometric ZnxNy

46.8%

38.7%

Stoichiometric Zn3N2

42.7%

52.1%

N-N bond

10.4%

9.2%

Zn-O bond

88.1%

89.5%

VO

5.7%

7.8%

Impurity (OH)

6.2%

2.7%

The representative transfer characteristics for the ZnON TFTs with a control and Ta-capped channel can be seen in Figure 6. The control ZnON TFT yielded a high mobility (FE) of 36.2 cm2/Vs, a subthreshold gate swing (SS) of 0.28 V/decade, a threshold voltage (VTH) of 1.28V, and an ION/OFF ratio of 2.9  108. The mobility for the control ZnON device with oxygen-rich anions was comparable to that for state-ofthe-art ZnON TFTs reported in the literature, which can be related to the low electron effective mass of the ZnON system. By contrast, a significant performance enhancement was observed for the Ta-capped ZnON TFTs. The FE of ~89.4 cm2/Vs, SS of 0.33 V/decade, VTH of -0.45V, and ION/OFF ratio of 8.6  108, respectively, as summarized in Table 2. The statistical data for the electrical parameters were obtained from 20 control and Ta-capped ZnON TFTs, as shown in the box plot of Figure S5. Improvements in the transport and IDS modulation capability of the Ta capped ZnON TFTs were reflected in their superior output characteristics in Figures 6c, d. It was interesting to discuss the physical rationale with regard to the beneficial effects of TaOx encapsulation in terms of mobility enhancement. Given a thermal annealing condition of 250 C, the Gibbs free energies of formation (Gf) for ZnO and Ta2O5 were -348.1 and 1911.2 kJ/mole, respectively.36, 37 A more negative Gf for Ta2O5 meant that Ta exhibited a stronger oxidation power than ZnO. Thus, PDA of the Ta/ZnON stack at high temperature would cause oxidation 14 ACS Paragon Plus Environment

Page 15 of 30 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


of the Ta film and simultaneous reduction of the ZnON film to metallic Zn. However, if the PDA temperature was moderate (< 250 C), the high activation energy barrier (> 1 eV) against the dissociation of Zn-O bonds would prevent strong oxidation of the TaOx film involving the conversion of ZnON to a metallic Zn film. Instead, weakly bonded oxygen and/or nitrogen species such as OH and N-N bonds could be selectively eliminated near the back surface of the ZnON films.19, 38, 39 This tentative model was consistent with the fact that the N and O 1s XP spectra for ZnON with a TaOx encapsulation film exhibited a smaller intensity for non-stoichiometric ZnxNy, N-N bonds, and OH-related impurity peaks. Because those defects acted as scattering centers in ZnON, reductions in weakly bonded species could result in an enhanced carrier mobility. The feasibility of the other transition metals such as Ti and Ni as an effective capping layer was also examined. Our finding was that the usage of TiOx and NiOx was also effective to boost the field-effect mobility of the resulting ZnON transistors as shown in Figure S6. The comparable mobility of 75 ~ 81 cm2/Vs was obtained for these capped devices, indicating that the capping approach can be extendable to the other transition metals. However, it has to be mentioned that the SS values for the Ti or Ni-capped devices were more degraded to 0.50 ~ 0.53 V/decades, which is larger than that (0.33 V/decade) for the Ta-capped devices. Although the exact reason is beyond the scope of this paper, the lower oxidation power of Ni and Ti may be one of the possible origin compared to that of Ta. The Gibbs free energies of formation (ΔGf) of NiO, TiO2 and Ta2O5 are -211.54 kJ/mole, -889.41 kJ/mole and 1911.2 kJ/mole, respectively, at the given annealing temperature.21 The capability of facilitating the lattice ordering and creating the VO should be strongly related to the oxidation power of the capping materials, which is the highest for the Ta capping. It should be also noted that the surface effect for the passivated device invokes the deterioration of transfer characteristics in terms of SS and VTH values when the ultrathin channel is used (for example, < 5 nm).40, 41 In that case, the TaOx film may passivate the adverse surface states, which also results in the improvement of field-effect mobility for the capped ZnON transistors. For comparison, the TFTs with 15 ACS Paragon Plus Environment


Page 16 of 30

more thin ZnON channel thickness of 10 nm were also characterized. The significant deterioration in SS and VTH value for the device with 10-nm-thick ZnON film was not observed (Figure S7), indicating that the back surface defect is not a significant limiting factor for these devices with the channel thickness  10 nm. It can be also shown that all capped transistors with 10, 20 and 30-nm-thick ZnON channel layers exhibited the enhanced mobility and on-current values compared to those without the TaOx capping layer, which consistent with our interpretation (Table S3).

Figure 6. Transfer characteristics of the (a) control ZnON and (b) Ta-capped ZnON TFTs. Output characteristics of the (c) control ZnON and (d) Ta-capped ZnON TFTs.

Table 2. Transfer Characteristics of the (a) Control ZnON TFT and (b) Ta-capped ZnON TFTs. 16 ACS Paragon Plus Environment

Page 17 of 30 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 2

Samples

FE (cm /Vs)

SS (V/decade)

VTH (V)

ION/OFF

(a) Control ZnON TFT

36.2 ± 1.8

0.28 ± 0.04

1.28 ± 0.89

2.9 × 10

(b) Ta-capped ZnON TFT

89.4 ± 9.2

0.33 ± 0.14

-0.45 ± 1.17

8.6 × 10

8 8

Figure 7. Transfer characteristics of the (a) control ZnON and (b) Ta-capped ZnON TFTs. The black and red lines denote the transfer characteristics before and after storage for 1 month under an air atmosphere, respectively. Next, the environmental stability of TFTs with a control and Ta-capped ZnON channel was investigated. Figure 7 depicts variations in the transfer characteristics of the ZnON TFTs under an air atmosphere over a period of 30 days. The control ZnON TFTs suffered from a severe stretch-out under the subthreshold drain current region with an increased IOFF value. Nitrogen anions were found to be easily dissociated due to a weak bonding energy with the metal ions, which led to nitrogen out-diffusion.18 The undesirable nitrogen dissociation phenomena could be strongly suppressed through TaOx encapsulation of the ZnON channel layer, as shown in Figure 7b. A multiple trap and release (MTR) mechanism could be used to describe carrier transport in the amorphous and nanocrystalline disordered semiconductor. With regard to nanocrystalline ZnON, most of 17 ACS Paragon Plus Environment


Page 18 of 30

the delocalized electron carriers were likely to be captured in the nanoscale domain/grain boundaries where a tiny portion of the captured electrons could be activated from localized trap states to the conduction band (CB). Thus, an increasing free carrier concentration (nfree) due to chemical doping or the application of a positive gate field made the electrons fill the lower trap states effectively, which led to a population enhancement of nfree in CB and mobility boosting. In this MTR mechanism, the apparent fieldeffect mobility (FE) could be calculated as follows:33 𝜇𝐹𝐸 = 𝜇𝑏𝑎𝑛𝑑 ∗

(

𝑛𝑓𝑟𝑒𝑒 𝑛𝑡𝑜𝑡

)= 𝜇

𝑏𝑎𝑛𝑑

∗

(

𝑛𝑓𝑟𝑒𝑒 𝑛𝑓𝑟𝑒𝑒 + 𝑛𝑡𝑟𝑎𝑝

),

(1)

where ntot is the total trap density and ntrap is the captured carrier density.

Figure 8. (a) Simulated transfer characteristics of control ZnON (blue line) and Ta-capped ZnON TFTs (black line) at VDS = 0.1. (b) Extracted DOS parameters for the control ZnON (blue line) and Ta-capped ZnON TFTs (black line). Table 3. Key Parameters of the Control ZnON and Ta-capped ZnON TFTs used during Simulation.

NTA [cm− 3·eV−1]

Control ZnON TFT

Ta-capped ZnON TFT

4.0  1019

1.0  1019


Page 19 of 30 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


kTTA [eV]

0.025

0.025

NDA [cm− 3·eV−1]

3.0  1017

1.0  1017

kTDA [eV]

0.3

0.3

NSD [cm− 3·eV−1]

1.0  1017

4.0  1017

KTSD [eV]

0.1

0.1

ESD [eV]

1.21

1.21

Ne [cm− 3]

1.5  1017

2  1018

μband [cm2 V−1·s−1]

45

45

Figure 8a shows the TCAD simulation results of the control ZnON and Ta-capped ZnON TFTs. Three kinds of states consisting of an acceptor-like tail state (gTA), acceptor-like deep state (gDA), and shallow donor-like state (gSD) were considered to fit the I-V characteristics of both devices.33 The total densityof-state (DOS) distribution below the conduction band minimum (EC) was modeled as the following linear summation of gTA, gDA, and gSD:

𝑔𝑇𝐴 (𝐸) + 𝑔𝐷𝐴 (𝐸) + 𝑔𝑆𝐷 (𝐸)

(

= 𝑁𝑇𝐴𝑒𝑥𝑝 ―

)

𝐸𝑐 ― 𝐸 𝑘𝑇𝑇𝐴

( ( ))

+ 𝑁𝑇𝐷𝑒𝑥𝑝 ―

𝐸𝑐 ― 𝐸 2 𝑘𝑇𝐷𝐴

( (

+ 𝑁𝑆𝐷𝑒𝑥𝑝 ―

𝐸𝑉 ― 𝐸𝑆𝐷 ― 𝐸 2 𝑘𝑇𝑆𝐷

) ).

(2).

An exponential tail distribution is described by the conduction band edge intercept densities (NTA and NTD) and characteristic decay energy (kTTA and kTDA). For Gaussian distributions, it can be described by the total density of states (NSD), characteristic decay energy (TSD), and peak energy distribution (ESD).

From the TCAD simulation, the extracted key parameters were summarized in Table 3 and each component was depicted in Figure 8b. Interestingly, the NTA and NDA values for the Ta-capped ZnON TFT were reduced from 4.0  1019 cm-3eV-1 and 3.0  1017 cm-3eV-1 (control ZnON TFT) to 1.0  1019



Page 20 of 30

cm-3eV-1 and 1.0  1017 cm-3eV-1, respectively. By contrast, the NSD value of the shallow donor-like state (gSD) for the Ta-capped ZnON TFT increased from 1.0  1017 cm-3eV-1 to 4.0  1017 cm-3eV-1. Because the overall DOS distribution below EC was dominated by the acceptor-like tail state, the filling of gate field-induced carriers into the localized trap states below EC occurred effectively for the Ta-capped ZnON device due to a much smaller gTA distribution. Thus, a higher carrier mobility was expected for the Tacapped ZnON device, which was consistent with the experimental IV result. For a better understanding, the schematic energy diagram of for the Ta-capped ZnON TFT was calculated and can be seen in Figure 9. The channel region of ZnON along the channel length direction was divided into three regions, two Ta-uncapped regions and one Ta-capped region, because the half region of the ZnON active island was selectively capped with the Ta film, as depicted in Fig. 1. The Ne values for the Ta-uncapped and capped ZnON films were estimated to be 1.5  1017 cm-3 and 2  1018 cm-3, respectively, from independent C-V analysis, which was reflected into the TCAD simulation. It should also be noted that the high Ne value for the Ta-capped ZnON film could be attributed to a high quality lattice Zn2N3 structure (O-substituted anti-bixbyite), which was consistent with the XPS results. It could be clearly seen that the smaller acceptor-like trap density and larger Ne value for the Ta-capped ZnON channel region result in an elevation of the Fermi energy level toward EC under thermal equilibrium conditions (see Figures 10a, b). A contour of the current density for both devices calculated from the TCAD simulation can be seen in Figures 10c, d. Under the linear region, the current density profile of the control ZnON TFTs was uniform along the channel length direction in which the largest current density occurred near the dielectric/channel interface. By contrast, the Ta-capped device exhibited bulk accumulation-like transport in the ZnON channel layer under the Ta-capped region; facile elevation of the quasi-Fermi level as a result of a lower trap state distribution and large Ne value widened the effective channel thickness considerably. Therefore, the total injected current density from the source electrode into the channel region was enhanced, leading to an improved apparent field-effect mobility of the Ta-capped 20 ACS Paragon Plus Environment

Page 21 of 30 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


ZnON TFTs. Finally, it would be interesting why the Ta-capped ZnON TFTs had the negatively shifted VTH compared to the uncapped control device. There are several factors affecting the value of VTH value of ZnON TFTs. Generally, the creation of shallow donor such as VO causes the negative displacement of VTH. Thus, the negative shift of VTH for the Ta-capped device is mainly attributed to the increase in VO concentration, which was already mentioned in XPS analysis. In contrast, the gap states including the tailing states and deep states can capture the free electron carriers depending the position of quasi-Fermi level, which will result in the positive displacement of VTH and the increase in SS value. First, the acceptor-like tailing states below CB edge would be related to the disordering of Zn cation. The less nonstoichiometry defect density, better crystallinity and uniformity of the Ta-capped ZnON film are responsible for the reduction (3.6  1016 cm-3eV-1 at EC – EF = 0.3 eV) in the acceptor-like deep states, compared to that (1.1  1017 cm-3eV-1 at EC – EF = 0.3 eV) for the control ZnON device (see Figure 8b). However, this rationale is not consistent with the increased SS value for the Ta-capped device. One of the plausible mechanisms is the in-diffusion of Ta ions into the ZnON film during the 2nd PDA at 200 C (see Figure 4b). The multiple chemical valence states of Ta cation such as Ta1+, Ta2+, Ta4+, Ta5+ etc. can introduce the creation of gap-states in the ZnON semiconductor. The creation of shallow donor-like VO cannot be ruled out as an origin of increased SS value for the Ta-capped devices. The huge increase (4.0  1017 cm-3eV-1 at EC – EF = 0.3 eV) in shallow donor-like trap states in TCAD result for the Ta-capped device is tentatively attributed to both the incorporation of diffused Ta in ZnON film and the creation of donor-like VO defects whereas the donor-like trap state for the control device was ~1.0  1017 cm-3eV-1 at EF – EC = -0.3 eV. The total gap state density including the donor-like and acceptor-like deep states were calculated to be 2.1  1017 cm-3eV-1 and 4.3  1017 cm-3 eV-1 at EC – EF = 0.3 eV for the control and Tacapped devices, respectively. We noted that the SS values in the subthreshold drain current characteristics for the ZnON TFTs were extracted at the VGS value, which corresponds to the band bending condition of EC – EF = 0.3 eV. The higher SS value for the Ta-capped device is corroborated with its larger total gap 21 ACS Paragon Plus Environment


Page 22 of 30

state number near EC – EF = 0.3 eV because the SS value determined from the transfer characteristics is a good indicator of the total trap density including the donor-like and acceptor-like trap states. Furthermore, the increase (~1.8  1018 cm-3) in the Ne value for the Ta-capped ZnON film compared to that for the uncapped ZnON film outnumbered the increase (~7.9  1017 cm-3) in the total trap density, which indicates that the VTH positive shift for the Ta-capped ZnON TFTs can be mainly attributed to the creation of beneficial shallow donor center such as VO.

Figure 9. A schematic energy-band diagram dictated by trap-limited conduction (TLC) modulation for the Ta-capped ZnON TFTs.


Page 23 of 30 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 10. (a), (b) Simulated energy-band diagrams for the control ZnON and Ta-capped ZnON TFTs. (c), (d) Current density according to the position of the ZnON TFT and Ta-capped ZnON TFT at VDS = 0.1 V and VGS = 4 V. The higher the value from purple to red, the larger the value in each graph.

4. CONCLUSION In summary, the use of TaOx films as a scavenging and encapsulation layer has been presented; the film could be produced via deposition from a Ta target using DC sputtering followed by a modest annealing process. The volume fraction of nanoscale ZnON crystallites, which were dispersed within an amorphous matrix, was diminished along the depth direction for the control ZnON film. The TaOx/ZnON stack resulted in uniform crystallization of the overall ZnON films as well as an enhanced stoichiometric Zn3N2 portion with less defect centers and a higher free electron density. The transistor fabricated using a TaOx 23 ACS Paragon Plus Environment


Page 24 of 30

passivation layer exhibited an excellent mobility of 89.4 cm2/Vs and a good ION/OFF ratio of 8.6  108, which could be explained by the scavenging effect of non-stoichiometric anion-related defects and encapsulation capability during and/or via TaOx formation. This improvement was confirmed through TCAD simulation. Bulk accumulation-like transport in the ZnON channel layer under the Ta-capped region greatly contributed to boosting the drain current, which occurred due to facile elevation of the quasi-Fermi level as a result of a lower acceptor-like trap state distribution and large Ne value. The promising effects of Ta capped ZnON TFTs were also confirmed by their superior aging characteristics compared to the control ZnON TFTs. Therefore, the novel encapsulation method using protective TaOx formation could be a useful route for producing high-performance and reliable ZnON TFTs.

ASSOCIATED CONTENT

Supporting Information XP spectra of the control ZnON and Ta-capped ZnON film; Statistical data of the control and Ta-capped ZnON TFTs; Transfer characteristics of the Ti and Ni-capped ZnON TFTs; Transfer characteristics of the control and Ta-capped TFTs with different ZnON channel thickness.

AUTHOR INFORMATION

Corresponding Author *E-mail: [email protected]


Page 25 of 30 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


ACKNOWLEDGMENT This research was supported by the MOTIE (Ministry of Trade, Industry & Energy (10051403) and a KDRC (Korea Display Research Consortium) support program for the development of future device technologies for the display industry. This work was also supported by the industrial strategic technology development program funded by MKE/KEIT under grants 10079974.



Page 26 of 30

REFERENCES (1) Nomura, K.; Ohta, H.; Takagi, A.; Kamiya, T.; Hirano, M.; Hosono, H. Room-Temperature Fabrication of Transparent Flexible Thin-Film Transistors using Amorphous Oxide Semiconductors. Nature 2004, 432, 488–492. (2) Kwon, J. Y.; Jeong, J. K. Recent Progress in High Performance and Reliable n-Type Transition Metal Oxide-Based Thin Film Transistors. Semicond. Sci. Technol. 2015, 30, 024002. (3) Kamiya, T.; Nomura, K.; Hosono, H. Present Status of Amorphous In–Ga–Zn–O Thin-Film Transistors. Sci. Technol. Adv. Mater. 2010, 11, 044305. (4) Park, J.-S.; Kim, H.; Kim, I.-D. Overview of Electroceramic Materials for Oxide Semiconductor Thin Film Transistors. J. Electroceram. 2014, 32, 117−140. (5) Lee, K.-H.; Jung, J.S.; Son, K.S.; Park, J.S.; Kim, T.S.; Choi, R.; Jeong, J.K.; Kwon, J.-Y.; Koo, B.; Lee, S. The effect of Moisture on the Photon-enhanced Negative Bias Thermal Instability in Ga–In–Zn– O Thin Film Transistors. Appl. Phys. Lett. 2009, 95, 232106. (6) Noh, H. K.; Chang, K. J.; Ryu, B.; Lee, W. J. Electronic Structure of Oxygen-vacancy Defects in Amorphous In-Ga-Zn-O Semiconductors. Phys. Rev. B. 2011, 84, 115205. (7) Jeong, J. K. Photo-bias Instability of Metal Oxide Thin Film Transistors for Advanced Active Matrix Displays. J. Mater. Res. 2013, 28, 2071-2084. (8) Yang, B. S.; Park, S.; Oh, S.; Kim, Y. J.; Jeong, J. K.; Hwang, C. S.; Kim, H. J. Improvement of the Photo-bias Stability of the Zn-Sn-O Field Effect Transistors by an Ozone Treatment. J. Mater. Chem. 2012, 22, 10994-10998. (9) Oh, H.; Yoon, S. M.; Ryu, M. K.; Hwang, C. S.; Yang, S.; Park, S. H. K. Photon-accelerated Negative Bias Instability Involving Subgap States Creation in Amorphous In-Ga-Zn-O Thin Film Transistor. Appl. Phys. Lett. 2010, 97, 183502. (10) Ye, Y.; Lim, R.; White, J. M. High Mobility Amorphous Zinc Oxynitride Semiconductor Material 26 ACS Paragon Plus Environment

Page 27 of 30 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


for Thin Film Transistors. J. Appl. Physics. 2009, 106, 074512. (11) Kim, H.-S.; Jeon, S. H.; Park, J. S.; Kim, T. S.; Son, K. S.; Seon, J.-B.; Seo, S.-J.; Kim, S.-J.; Lee, E.; Chung, J. G.; Lee, H.; Han, S.; Ryu, M.; Lee, S. Y.; Kim, K. Anion Control as a Strategy to Achieve High-mobility and High-stability Oxide Thin-film Transistors. Sci. Rep. 2013, 3, 1459. (12) Gao, H.; Zhang, X.; Zhao, Y.; Yan, B. The Correlation of Material Properties and Deposition Condition of ZnON Thin Films. AIP Adv. 2017, 7, 025111. (13) Lany, S.; Zunger, A. Anion Vacancies as a Source of Persistent Photoconductivity in II-VI and Chalcopyrite Semiconductors. Phys. Rev. B 2005, 72, 035215. (14) Lee, E.; Benayad, A.; Shin, T.; Lee, H.; Ko, D. S.; Kim, T. S.; Son, K. S.; Ryu, M.; Jeon, S.; Park, G. S. Nanocrystalline ZnON; High Mobility and Low Band Gap Semiconductor Material for High Performance Switch Transistor and Image Sensor Application. Sci. Rep. 2014, 4, 4948. (15) Ok, K. C.; Jeong, H. J.; Kim, H. S.; Park, J. S. Highly Stable ZnON Thin-Film Transistors With High Field-Effect Mobility Exceeding 50 cm2/Vs. IEEE Electron Device Lett. 2015, 36, 38-40. (16) Jang, J. T.; Park, J.; Ahn, B. D.; Kim, D. M.; Choi, S. J.; Kim, H. S.; Kim, D. H. Study on the Photoresponse of Amorphous In-Ga-Zn-O and Zinc Oxynitride Semiconductor Devices by the Extraction of Sub-Gap-State Distribution and Device Simulation. ACS Appl. Mater. Interfaces 2015, 7, 1557015577. (17) Kuan, C. I.; Lin, H. C.; Li, P. W.; Huang, T. Y. High-Performance Submicrometer ZnON Thin-Film Transistors with Record Field-Effect Mobility. IEEE Electron Device Lett. 2016, 37, 303-305. (18) Lee, E.; Kim, T.; Benayad, A.; Kim, H.; Jeon, S.; Park, G. S. Ar Plasma Treated ZnON Transistor for Future Thin Film Electronics. Appl. Phys. Lett. 2015, 107. 122105. (19) Shin, Y.; Kim, S. T.; Kim, K.; Kim, M. Y.; Oh, S.; Jeong, J. K. The Mobility Enhancement of Indium Gallium Zinc Oxide Transistors via Low-temperature Crystallization using a Tantalum Catalytic Layer. Sci. Rep. 2017, 7. 10885. 27 ACS Paragon Plus Environment


Page 28 of 30

(20) Kuriyama, K.; Takahashi, Y.; Sunohara, F. Optical band gap of Zn3N2 films Phys. Rev. B. 1993, 48, 2781-2782. (21) Barin, I. Thermochemical data of pure substances; 3rd Ed. VCH: New York, 1995. (22) Wang, C.; Ji, Z.; Liu, K.; Xiang, Y.; Ye, Z. p-Type ZnO thin films prepared by oxidation of Zn3N2 thin films deposited by DC magnetron sputtering. Journal of Crystal Growth, 2003, 259, 3. (23) Zou, C. W.; Chen, R. Q.; Gao, W. The microstructures and the electrical and optical properties of ZnO:N films prepared by thermal oxidation of Zn3N2 precursor. Solid State Communications, 2009, 149, 45. (24) Park, J.; Kim, Y. S.; Ok, K. C.; Park, Y. C.; Kim, H. Y.; Park, J. S.; Kim, H. S. A Study on the Electron Transport Properties of ZnON Semiconductors with Respect to the Relative Anion Content. Sci. Rep. 2016, 6,24787. (25) Partin, D.E.; Williams, D.J.; O’Keeffe, M. The Crystal Structures of Mg3N2 and Zn3N2. J. Solid State Chem. 1997, 132, 56–59. (26) Wu, H.-Y.; Cheng, X.-L.; Hu, C.-H.; Zhou, P. The Structure and Thermodynamic Properties of Zinc Oxide with Wurtzite and Rocksalt Structure under High Pressures. Physica B., 2010, 405, 606-612. (27) Bruce, D. W.; O’Hare, D.; Walton, R. I. Structure from Diffraction Methods, John Wiley & Sons, Ltd: United Kingdom, 2014; pp 216–219. (28) Hwang, A. Y.; Kim, S. T.; Ji, H.; Shin, Y.; Jeong, J. K. Metal-Induced Crystallization of Amorphous Zinc Tin Oxide Semiconductors for High Mobility Thin-Film Transistors. Appl. Phys. Lett. 2016, 108, 152111. (29) Tabet, N.; Faiz, M.; Al-Oteibi, A. XPS Study of Nitrogen implanted ZnO Thin Films Obtained by DC-magnetron Reactive Plasma. J. Electron Spectrosc. Relat. Phenom. 2008, 163, 15−18. (30) Petravic, M.; Deenapanray, P.; Coleman, V.; Jagadish, C.; Kim, K. J.; Kim, B.; Koike, K.; Sasa, S.; Inoue, M.; Yano, M. Chemical States of Nitrogen in ZnO Studied by Near Edge X-ray Absorption Fine 28 ACS Paragon Plus Environment

Page 29 of 30 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


Structure and Core-level Photoemission Spectroscopies. Surf. Sci. 2006, 600, 81−85. (31) Li, X. H.; Xu, H. Y.; Zhang, X. T.; Liu, Y. C.; Sun, J. W.; Lu, Y. M. Local Chemical States and Thermal Stabilities of Nitrogen Dopants in ZnO Film Studied by Temperature-dependent X-ray Photoelectron Spectroscopy. Appl. Phys. Lett. 2009, 95, 191903. (32) Long, R.; Dai, Y.; Yu, L.; Huang, B.; Han, S. Atomic Geometry and Electronic Structure of Defects in Zn3N2. Thin Solid Films. 2008, 516, 1297−1301. (33) Lee, S.; Nathan, A.; Ye, Y.; Guo, Y.; Robertson, J. Localized Tail States and Electron Mobility in Amorphous ZnON Thin Film Transistors. Sci. Rep. 2015, 5, 13467. (34) Ok, K. C.; Jeong, H. J.; Lee, H. M.; Park, J.; Park, J. S. Comparative Studies on the Physical and Electronic Properties of Reactively Sputtered ZnO and ZnON Semiconductors. Ceram. Int. 2015, 41, 13281−13284. (35) Cao, X.; Sato, A.; Ninomiya, Y.; Yamada, N. Oxygen-Doped Zinc Nitride as a High-Mobility Nitride-Based Semiconductor. J. Phys. Chem. C 2015, 119, 5327– 5333. (36) Lee, C. B.; Kang, B. S.; Benayad, A.; Lee, M. J.; Ahn, S. E.; Kim, K. H.; Stefanovich, G.; Park, Y.; Yoo, I. K. Effects of Metal Electrodes on the Resistive Memory Switching Property of NiO Thin Films. Appl. Phys. Lett. 2008, 93, 042115. (37) Jin, Z. W.; Fukumura, T.; Kawasaki, M.; Ando, K.; Saito, H.; Sekiguchi, T.; Yoo, Y. Z.; Murakami, M.; Matsumoto, Y.; Hasegawa, T.; Koinuma, H. High Throughput Fabrication of Transition-metal-doped Epitaxial ZnO Thin Films: A Series of Oxide-diluted Magnetic Semiconductors and Their Properties. Appl. Phys. Lett. 2001, 78, 3824-3826. (38) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies; CRC Press: Boca Raton, FL, 2007.



Page 30 of 30

(39) Zan, H. W.; Yeh, C. C.; Meng, H. F.; Tsai, C. C.; Chen, L. H. Achieving High Field-Effect Mobility in Amorphous Indium-Gallium-Zinc Oxide by Capping a Strong Reduction Layer. Adv. Mater. 2012, 24, 3509-3514. (40) Chiang, T. H.; Yeh, B. S.; Wager, J. F.; Amorphous IGZO Thin-Film Transistors with Ultrathin Channel Layers. IEEE Transactions on Electron Devices, 2015, 62, 11. (41) Zhang, B.; Li, H.; Zhang, X.; Luo, Y.; Wang, Q.; Song, A. Performance regeneration of InGaZnO transistors with ultra-thin channels. Appl. Phys. Lett., 2015, 106, 093506.

Table of Contents


Boosting Carrier Mobility in Zinc Oxynitride Thin-Film Transistors via

Recommend Documents