Evidence of Negative Capacitance in Piezoelectric ZnO Thin Films

Subscriber access provided by NEW YORK MED COLL

Article

Evidence of negative capacitance in piezoelectric ZnO thin films sputtered on interdigital electrodes Marco Laurenti, Alessio Verna, and Alessandro Chiolerio ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b05336 • Publication Date (Web): 22 Oct 2015 Downloaded from http://pubs.acs.org on October 23, 2015

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

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

Page 1 of 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

Evidence of negative capacitance in piezoelectric ZnO thin films sputtered on interdigital electrodes Marco Laurenti,§ Alessio Verna,§,† and Alessandro Chiolerio §,* §

Center for Space Human Robotics, Istituto Italiano di Tecnologia, C.so Trento, 21, 10129

Torino, Italy †

Department of Applied Science and Technology, Politecnico di Torino, C.so Duca degli

Abruzzi 24, 10129 Torino, Italy KEYWORDS: zinc oxide, negative capacitance, electrical properties, thin films, sputtering

ABSTRACT. The scaling paradigm known as Moore’s Law, with the shrinking of transistors and their doubling on a chip every 2 years, is going to reach a painful end. Another less known paradigm, the so called Koomey’s Law, stating that the computing efficiency doubles every 1.57 years, poses other important challenges, since the efficiency of rechargeable energy sources is substantially constant and any other evolution is based on device architecture only. How can we still increase the computational power / reduce the power consumption of our electronic environments? A first answer to this question comes from the quest for new functionalities. Within this aim, negative capacitance (NC) is becoming one of the most intriguing and studied phenomena since it can be exploited for reducing the aforementioned limiting effects in the downscaling of electronic devices. Here we report the evidence of negative capacitance in 80 nm-thick ZnO thin films sputtered on Au interdigital electrodes (IDEs). Highly (002)-oriented

ACS Paragon Plus Environment

1


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

ZnO thin films, with a fine-grained surface nanostructure and the desired chemical composition, are deposited at room temperature on different IDEs structures. DC electrical measurements highlighted the semiconducting nature of ZnO (current density in the order of 10-3 A/cm2). When turned into the AC regime (from 20 Hz up to 2 MHz) the presence of NC values is observed in the low-frequency range (20 – 120 Hz). The loss of metal/semiconductor interface charge states under forward bias conditions, together with the presence of oxygen vacancies and piezoelectric/electrostriction effects, are believed to be at the basis of the observed negative behavior, suggesting that ZnO-thin-film-based FETs can be a powerful instrument to go beyond the Boltzmann limit and the downscaling of integrated circuit elements required for the fabrication of portable and miniaturized electronic devices, especially for electric household appliances working in the low 50 Hz utility frequency.

1. INTRODUCTION The development of low-power consumption devices with size-reduced integrated circuits (ICs) is becoming one of the main challenges for both materials scientists and engineers, because of the increasing demand for portable, miniaturized electronics applications. Nevertheless, the ongoing scaling down of Field Effect Transistors (FETs) is prevented by some limitations. These mainly consist in the possibility of reducing the subthreshold slope (SS), i.e., the gate voltage required to increase of one decade the drain current flowing through the transistor channel, below the Boltzmann limit of 60 meV/dec together with the resulting increase of the power dissipation energy in high-density ICs.1, 2 Negative capacitance (NC) phenomena are reported to be promising effects for going below to the aforementioned Boltzmann limit. Different works reported about the observation of NC effects in Schottky diodes3 and p-n junctions.4 In the first


2

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


case, such kind of phenomenon is generally associated to the loss of interface charge occupying interfacial states below the Fermi level (see Figure 1(a)).5 In the case of p-n junctions, the observed change in sign of the capacitance is related to the depopulation of minority carriers under high forward bias conditions (see Figure 1(b)).6 The presence of a negative capacitance has been also reported for organic polymers.7, 8, 9, 10 In this case, the observed negative behavior has been predicted to be ascribable to charge trapping in a dispersive media when the carrier mobility has a negative imaginary component.11 A key role in NC effects is also played by oxygen vacancies.12 These are often present in metal oxides like zinc oxide (ZnO), titanium oxide, and tantalum oxide, and are believed to be responsible for the appearance of memristive phenomena.13, 14 From 2008, different theoretical works predicted the possibility of reducing the SS below the Boltzmann limit by using a ferroelectric insulator as gate oxide (Figure 1(c)).15, 16, 17, 18

It is shown that the interaction among electric dipoles of the ferroelectric gate oxide should

provide, in principle, a positive feedback effect capable of implementing a step-up voltage transformer, i.e., a voltage amplification that increases the channel potential more than that applied externally. This can reduce the SS without further increasing the external gate voltage. The physical principle which is at the basis of NC in ferroelectric materials is based on the presence of a thermodynamic unstable phase (i.e., a order-disorder phase transition). Because of its intrinsic instability, the experimental observation of such kind of phenomena is not trivial and has been only recently reported.19 In particular, Khan et al. experimentally observed the presence of a NC effect in a thin, epitaxial ferroelectric PZT film.20 According to the predicted theoretical results of Datta and coworkers,15 it is emphasized that, in principle, also other microscopic mechanisms are possible to be capable of giving rise to the appearance of NC phenomena and consequently providing the inner voltage amplification on the channel potential discussed above.


3


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

Page 4 of 30

Figure 1. Sketch of some device concepts showing the presence of NC phenomena. (a) Metal/Semiconductor interface: under suitable bias conditions, hot electrons injected at the M/S interface can pull out electrons from interfacial states. This is at the basis of the NC effects observed in such devices. (b) p-n junction: under high forward bias conditions the level of injected minority carriers increases exponentially. This causes the capacitance to turn into negative values. (c) Ferroelectric oxide: the alignment of electric dipoles inside the ferroelectric gives rise to an electrical polarization capable of increasing the channel potential. (d) Piezoelectric oxide: the inner piezopotential provides additional electrons to the 2D electron gas (2DEG), enhancing the channel potential. This explains the negative capacitance effect observed in piezoelectric materials.

Piezoelectricity is a physical mechanism often correlated to the presence of ferroelectricity. It represents a linear coupling between the mechanical and electrical states of a certain material. Some works recently reported on the exploitation of piezoelectric phenomena for inducing NC and thus reducing the SS in FETs. A simplified sketch of this system is shown in Figure 1(d). In particular, Jana and coworkers showed that a reduction in the Boltzmann limit is possible by using an electric field-induced electrostriction of a piezoelectric gate oxide barrier of an


4

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


AlN/GaN heterostructure.21 Then et al. reported about the evidence of NC by the addition of a piezoelectric AlInN layer to the gate stack of a GaN MOS-HEMT.22 To go beyond the comprehension of this physical mechanism, Wong et al. developed a thermodynamic model to quantitatively analyze the NC effect when the piezoelectric effect takes place.23 They considered ferroelectric, piezoelectric, and electrostiction effects as possible physical mechanisms able to induce NC. In their model it is shown that, despite both piezoelectricity and electrostriction could in principle provide NC effects, these are considered to be not strong enough to achieve the negative regime, since the required electric fields are orders of magnitude higher than the breakdown voltages of the most common piezoelectric materials. They also considered electrostriction as a second-order effect which could provide the required thermodynamic instability proper of NC. This could be possible when a negative electrostriction coefficient is present. They observed that since most of the common piezoelectric materials are characterized by positive coefficients, electrostriction prevents the compressive positive feedback effect from occurring. Moreover, considering the reported reduction of piezoelectric effects on NC, these are supposed to be completely masked by ferroelectricity, when present. However, several materials show piezoelectricity and no ferroelectricity, thus a possible influence of their piezoelectric behavior on the NC can be evidenced and not masked from the stronger ferroelectric one. Among these, ZnO is one of the most investigated. It is a II-VI wide band gap semiconductor (3.37 eV), showing an exciton binding energy of around 60 meV at room temperature and being characterized by a theoretical negative electrostriction coefficient of −3.6 10−22 m2 ·V−2.24 The interest in studying ZnO-based nanomaterials is especially related to the easiness in preparing ZnO micro- and nanostructures,25, 26, 27 with lots of different techniques.25, 28, 29 The resulting ZnO structures generally show the presence of intrinsic point defects, like zinc interstitials and


5


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

Page 6 of 30

oxygen vacancies, that originate the intrinsic n-type electrical behavior of undoped ZnO. Generally, ZnO nanostructures show the hexagonal wurtzite-like crystal structure, belonging to the non centrosymmetric C46v space group. The lack of a center of symmetry in wurtzite ZnO involves the tetrahedrally coordinated Zn2+ cations and O2- anions. In the absence of mechanical forces acting on ZnO, the center of the positive and negative charges overlap, and the resulting dipole moment is zero. However, when a mechanical stress is applied on the tetrahedron, the center of the cations and anions will no longer coincide, giving rise to a non-zero dipole moment. The resulting macroscopic potential drop along the strain direction in the crystal is usually called piezopotential, and has been successfully exploited for the fabrication of piezoelectric sensors and nanogenerators.25, 27 In this work we report on the preparation and characterization of ZnO thin films sputtered on IDEs having different geometrical parameters. The morphology, crystal structure, and chemical composition of ZnO was first investigated by means of Field-Emission Scanning Electron Microscope, X-Ray Diffraction, and X-Ray Photoelectron Spectroscopy. A fine-grained 80 nmthick ZnO film with a columnar, c-axis oriented nanostructure was obtained, having the desired chemical composition and stoichiometry. The electrical characteristics of ZnO thin films was first investigated by Hall effect measurements, in order to determine the nature of majority carriers, resistivity, and the carriers’ mobility and concentration. The resulting ZnO thin film was deposited onto Au IDEs patterned by optical photolithography and wet-chemical etching processes. DC electrical analyses pointed out the presence of the expected semiconducting behavior of ZnO. The AC electrical properties of ZnO deposited onto the different IDEs structures were also investigated in the 20 Hz – 2 MHz frequency range. In all the cases, a


6

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


remarkable negative capacitance effect mostly pronounced at low frequencies (20 Hz – 120 Hz) was observed. 2. EXPERIMENTAL SECTION 2.1. Preparation of interdigital electrodes (IDEs) A Ti (10 nm)/Au (60 nm) bilayer metal structure was first deposited by e-beam evaporation on commercial thermal oxidized silicon wafers (1 µm oxide thickness). The electrodes geometries summarized in Table 1 were then defined on the wafers by standard UV photolithography (using Fujifilm HPR 504 photoresist). Short distance electrodes (10 and 30 µm) were defined by using a high resolution photolithographic chromium mask, while for longer distance IDEs (100 µm), a polyester mask with lower resolution was used. After photoresist development, a wet etching process with KI : I2 : H2O (4 g : 1 g : 40 ml) solution was used to remove the Au layer. Then wafers were rinsed with DI water and titanium was finally etched with HF : H2O2 : H2O (1:1:20) solution.

2.2. Deposition of Zinc Oxide thin films ZnO thin film was deposited by means of radio-frequency (RF) magnetron sputtering (KS-300 Confocal Dual machine, Kenosistec) starting from a 3” ceramic ZnO target (Kurt J. Lesker, purity 99.999%), with a target-to-substrate distance of about 8 cm. Suitable vacuum conditions with a base pressure of 4×10-5 Pa were obtained with a rotary and a turbo molecular pump. A RF signal at a working frequency of 13.56 MHz was employed to create the plasma. The deposition process was carried out at room temperature, with a RF power density of 0.66 W·cm-2, in a


7


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

Page 8 of 30

mixed atmosphere of Ar (95%) and O2 (5%), and with a pressure of 0.66 Pa. To prevent any incorporation of contaminants in the deposited film, the target was cleaned with a 15 min sputtering process in a pure Ar atmosphere before starting the deposition. Table 1. Geometrical parameters of the different IDEs configurations used for the assessment of the electrical properties of ZnO thin films. IDE configuration

A B C D

Interelectrode distance [µm] 10 30 100 100

Effective length [cm]

Effective area [cm2]

0.482 0.304 0.190 0.482

0.241·10-3 0.456·10-3 0.950·10-3 2.410·10-3

2.3. Characterization methods The morphology and average thickness of ZnO thin films was evaluated by Field Emission Scanning Electron Microscope (Carl Zeiss Dual-Beam Auriga instrument). X-Ray Diffraction measurements were performed by a Panalytical X’Pert Pro Diffractometer, equipped with a Cu Kα radiation as X-ray source (λ = 1.54059 Å). The chemical composition of ZnO thin film was analyzed by means of X-ray Photoelectron Spectroscopy (XPS) measurements, using a PHI 5000 VersaProbe (Physical Electronics) system. The system is equipped with a monochromatic Al Kα X-ray source and with an Ar+ gun in order to remove any contaminant or pollutant from the surface of the analyzed material. The piezoelectric behavior of ZnO thin films was evaluated by using a Piezo Evaluation System (PES, TFAnalyzer 2000HS, Aixacct). Hall effect measurements were performed at room temperature using an MMR K2500-3RSLP instrument. The electrical properties were tested on ZnO thin films deposited on Au IDEs having a different finger distance of 10, 30, and 100 µm. Both the DC and AC electrical regime were investigated at room


8

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


temperature by using a 2-point micro-contact setup, using a Keithley 2635A multimeter (DC) and an Agilent E4980A precision LCR meter (from 20 Hz up to 2 MHz ). 3. RESULTS AND DISCUSSION 3.1. Morphological characterization of ZnO thin film deposited on IDEs The morphology and average thickness of ZnO thin film were first evaluated by means of Field Emission Scanning Electron Microscopy (FESEM). Figure 2(a) shows the surface morphology, which mainly consists of closely-packed, nanometer-sized crystal grains, rounded in shape, with an average diameter ranging between 20 and 30 nm. The cross-section FESEM analysis, reported as inset of Figure 2(a), allowed to estimate the average thickness of the ZnO thin film (80±2 nm) and pointed out the presence of a columnar growth. This aspect is usually reported in the case of highly c-axis oriented, sputtered ZnO thin films,25, 30 thus it can be considered as a first evidence of the presence of the desired crystal orientation in the investigated ZnO sample. The observed morphological characteristics directly come from the particular deposition conditions used for this work. By considering the sputtering time and the ZnO film thickness, it is possible to estimate the deposition rate, which is around 4.5 Å·min-1. The limited deposition rate used in this work, as well as the reduced energy of the impinging ions participating in the process for growing the ZnO layer, are expected to considerably limit the surface mobility of the sputtered atoms adsorbed on the substrate, which can thus aggregate into a high number of small-sized crystal grains. A high amount of grain boundaries is also expected from the observed morphology.


9


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

Page 10 of 30

Figure 2. (a) FESEM images showing the fine-grained surface morphology of ZnO, with the related cross-section; (b) sketch of the three different IDEs configurations using for electrical characterizations of ZnO thin films, where white represents the metalized part and green the bare substrate; (c) and (d) FESEM images acquired at different magnifications showing the ZnO layer deposited on the IDEs with a finger distance of 10 µm.

This aspect must be carefully taken into account in the evaluation of the electrical properties of ZnO thin films, since grain boundaries represent crystal defects for charge carriers moving inside the sample.30 ZnO thin films were also deposited on Au IDEs (sketch of the different geometries shown in Figure 2(b) having different interelectrode distances, i.e., 10, 30, and 100 µm. As an example, Figure 2(c) provides an overview of the ZnO thin film deposited on the 10 µmdistanced IDEs (see Figure S1 for the 30 µm-distanced and 100 µm-distanced geometries). To better investigate the morphological characteristics of ZnO deposited on this kind of interdigital structure FESEM analyses were also carried out at higher magnifications (Figure 2(d)). It can be


10

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


observed that ZnO entirely covers the electrode surface as well as the region between fingers, without any presence of cracks.

3.2. X-Ray Diffraction of ZnO thin film The crystal structure and orientation of ZnO thin films were evaluated by means of X-Ray Diffraction measurements. Figure 3(a) shows the XRD pattern acquired from the ZnO thin film deposited on a Si wafer. Only reflections coming from the (002) crystal plane are detected for ZnO, with a 2θ peak position placed at 34.34°, in good agreement with the data reported from JCPDS-ICDD (card n. 89-1397) for ZnO. The presence of the single (002) diffraction peak is representative of the hexagonal wurtzite-like ZnO crystal phase characterizing the analyzed sample. Since only the wurtzite (002) peak is detected, most of the ZnO crystal grains composing the films structure are supposed to be oriented along the c-axis direction. This aspect well agrees with the cross-section FESEM image reported as inset of Figure 2(a), where columnar crystal grains oriented perpendicularly with respect to the substrate are clearly visible. The shape of the diffraction peak can be considered as representative of the crystal quality of the analyzed material.31, 32 ZnO thin films are generally affected by the presence of several native crystal point defects, like oxygen vacancies and zinc interstitials, as well as hydrogen impurities inducing donor level acceptors.33 These result in spatial distortions of the crystal lattice as well as in inner stresses, involving the shift and broadening of the diffraction peak position and shape, respectively.31, 32, 33 The estimation of the FWHM related to the (002) diffraction peak can be obtained by a Gaussian fit. In the present work, the estimated FWHM is 0.26°, leading to the conclusion that the presence of native defects in the analyzed ZnO thin film is quite limited.


11


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

Page 12 of 30

Hence the slow deposition rate employed during the thin-film growth is here justified. Further, residual stresses are found to be negligible, because of the reduced ZnO thickness.

Figure 3. (a) XRD pattern of ZnO thin film deposited on Si wafer. The asterisk shows the diffraction contribution of the Si substrate. (b) Wide-scan XPS pattern acquired on ZnO thin film. High-resolution XPS spectra related to (c) Zn2p core-electrons, and (d) O1s core-electrons.

3.3. X-Ray Photoelectron Spectroscopy of ZnO thin films One of the factors influencing the physical properties of ZnO is the chemical composition and stoichiometry. In the present work both these aspects were investigated by XPS analyses. Figure 3(b) shows the wide-scan XPS pattern detected after cleaning the surface of the ZnO


12

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


sample with a Ar+ plasma process, to remove any adsorbed contaminants or pollutants. In the ideal case of a defect-free ZnO thin film, Zn and O atoms should be either present in a 50% atomic percentage, giving a stoichiometry ratio equal to one. However, the presence of a metallic phase due to an excess of Zn atoms is generally reported for sputter-deposited ZnO layers,33 which also show an oxygen deficiency. Both these aspects are generally considered as the main reason for the intrinsic n-type electrical behavior of ZnO.34, 35 XPS analyses confirm that the analyzed sample is not in the ideal stoichiometry condition. An excess of Zn atoms is pointed out from the quantitative compositional analysis. Indeed, the atomic percentage estimated for zinc and oxygen is 56.1% and 43.9%, respectively. The stoichiometric ratio of the analyzed sample can be estimated by considering the HighResolution (HR) Zn2p and O1s XPS spectra reported in Figure 3(c) and 3(d), respectively. The HR Zn2p pattern is described by a single peak which originates from two different contributions, that is from Zn2p core electrons involved in Zn−O and Zn−Zn bond configurations. However, the two corresponding peaks are very close to each other and cannot be resolved.36 As a consequence, there is no way of precisely estimating the percentage of Zn−O bonds from this spectrum. The situation is different when considering the HR O1s pattern. In this case different peaks can be clearly distinguished and their corresponding binding energy (BE) can be properly resolved by fitting the HR O1s curve, as shown in Figure 3(d). Since the BE is characteristic of each chemical bond, these last can be properly identified by comparing the obtained BE values with those reported in the National Institute of Standards and Technology database. In the present case it is found that oxygen atoms can be bonded to Zn or hydrogen, with characteristics binding energies of 529.8 and 531.3 eV, respectively. This second peak is also associated to O2- ions in the oxygen deficient


13


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

Page 14 of 30

regions within the ZnO thin film, and thus representative of oxygen vacancies.37 The ZnO stoichiometry can thus be estimated simply as the ratio between the area under the Zn2p and O-Zn HR peaks. In the present case it is greater than 1 (~ 1.29), because of the presence of the metallic Zn phase. Finally, the presence of oxygen-hydrogen chemical bonds witnesses that hydrogen impurities are present in the analyzed ZnO sample. This aspect, together with the excess of Zn atoms and oxygen deficiency, is expected to influence the electrical behavior of the investigated ZnO thin films. Indeed, as previously observed, hydrogen impurities, together with oxygen vacancies and zinc interstitials, affect the electrical conductivity of ZnO, and are responsible for its n-type intrinsic behavior. 3.4. Piezoelectric and electronic properties The presence of piezoelectricity and of electrostriction in the devices considered in this work was qualitatively investigated by measuring the mechanical displacement of the ZnO thin film under the application of an external voltage. Figure 4 shows the mechanical deformation exhibited by the ZnO thin film in such conditions. The behavior is not perfectly linear as expected, because of several contributions which bias the physical measure: electrostriction, substrate bending. In particular, electrostriction strongly affects the deformation of the device, and further increases the recorded displacement value. Moreover, it is also considered as the main factor responsible for the non-complete linearity of the characteristic, which indeed shows a remarkable decrease in the mechanical displacement as the bias voltage is increased up to its maximum value of 300 V. It is worth noting that a proper estimation of the ZnO piezoelectric constant cannot be obtained in the present case, since it would be highly overestimated by the strong mechanical displacement induced by electrostriction and from the bending of the substrate, which cannot be neglected. Even if no quantitative results can


14

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


be obtained from the reported data, we want to underline that our purpose is to show that in presence of an external bias voltage, a mechanical deformation starts to appear, which is representative of the co-existence of both piezoelectric phenomena as well as electrostriction effects. Hall effect measurements were performed at room temperature in order to investigate the resistivity, type of majority carrier and the corresponding carriers’ mobility and concentration. The resistivity is found to be 3.9×10-3 ohm·cm. The n-type nature of the analyzed ZnO thin film is confirmed since electrons are found as majority carriers. Moreover a remarkable high electron mobility of 62.7 cm2V-1s-1 and a carrier concentration of 2.54×1019 cm-3 are estimated. A such enhanced mobility, together with the low resistivity, are already reported for ZnO thin films with a reduced thickness (generally lower than 100 nm) comparable to our samples.37, 38, 39, 40 In the present case, XPS analyses pointed out the presence of oxygen vacancies and hydrogen impurities, which act as free carrier donors.37 These can be considered as the main factors responsible for our observed high carrier concentration. Nevertheless, a lower mobility value would be expected in the presence of a so high carrier concentration, mainly because of the increased amount of scattering phenomena among free carriers. Moreover, FESEM analyses highlighted the presence of grain boundaries which should act as defects and thus further decrease the electron mobility.38 Despite all these considerations, XRD characterization evidenced the good crystal quality of our ZnO thin films, as previously discussed. Only reflections coming from the (002) crystal plane are detected, witnessing that each of the single columns visible from inset of Figure 2(a) can be treated as single-crystals. The presence of densely-packed, single crystal, columnar grains is reported to promote the presence of direct conduction pathways


15


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

Page 16 of 30

for charge carriers.40 Then, it is expected that those carriers moving inside each single crystal column are less affected from scattering phenomena, resulting in the enhanced electron mobility observed in the present case.

Figure 4. Displacement vs. applied voltage recorded for samples A and C, witnessing the presence of both piezoelectricity and electrostriction effects.

The electrical behavior of ZnO thin films deposited on IDEs having different geometrical characteristics is first evaluated by DC analyses, between -200 V and +200 V. The resulting J-E curves for each of the considered electrodes configurations are shown in Figure 5, highlighting the overall semiconducting behavior of ZnO. Some small differences can be appreciated when the IDE geometry is changed. In the case of both 100 µm and 30 µm finger distance, the current suddenly increases up to 5 mA·cm-2 at low positive bias voltages (i.e., low electric fields), while when the finger distance is decreased up to 10 µm, the current density still remains limited (around 1 mA·cm-2).


16

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


Figure 5. Current density vs. electric field curves for ZnO thin films sputtered on different IDEs geometries.

The AC electrical properties of ZnO thin films deposited on the different IDEs structures were first investigated by the application of an alternating signal voltage (signal amplitude 100 mV) with a frequency sweep from 20 Hz up to 2 MHz.


17


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 18 of 30

Figure 6. Nyquist plots measured on ZnO thin films deposited on IDEs having different geometrical parameters. The color bar indicates the frequency range 20 Hz – 2 MHz.

The Nyquist plots are shown in Figure 6 for all the adopted IDEs geometries. It is found that the capacitance starts from negative values in the very low frequency range (20 Hz – 120 Hz), but suddenly goes to positive ones as the frequency is increased (see insets of Figure 6). In particular, NC values of -24.8 nF, and -210.0 nF were detected at a frequency of 20 Hz for samples B and D, respectively, while samples A and C showed negative values of -2.0 nF and -7.48 nF at 120 Hz.


18

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


Figure 7. Phase plots of sputtered ZnO thin films deposited on different IDE geometries. Signal frequency 1 MHz. The color bar indicates the DC voltage range.

The frequency behavior of the investigated material was then analyzed by measuring the AC response at fixed frequency values of 10 kHz, 100 kHz and 1 MHz. In all these cases, a ±40 V DC bias voltage was applied on the samples. The resulting phase plots acquired at 1 MHz are shown in Figure 7, while the ones measured at 10 kHz and 100 kHz are shown in Figure S2 and S3 of the Supporting Information. In all the devices negative values of ZnO reactance are clearly visible: defining it positive when the associated phase is capacitive and negative when inductive, counterintuitively we observed the second behavior this time at rather high frequencies. We observe that the measurements, having also a DC component, bring the material to a strongly nonlinear condition, where also its piezoelectric effect becomes important. Phase plots resulted in closed-loop curves at frequencies of 100 kHz and 1 MHz.


19


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

Page 20 of 30

The presence of NC in semiconductors is not surprising. Different works already reported in the literature about the presence of NC phenomena in semiconductor-based devices like p-n junctions and/or Schottky diodes. In these cases, it is believed that interface and bulk charge states play a fundamental role in the induction of such intriguing physical phenomena, which can be of great interest in view of downscaling the IC dimensions for portable and miniaturized devices, as also for low-frequency applications like electric household appliances. This is consistent with the effects we report in the low frequency regime. On the other hand, the evidence of NC effects in ferroelectric materials has been only recently pointed out. In 2008 a theoretical model predicting the existence of inductive behaviors in ferroelectrics has been developed by Datta et al..15 This theoretical result has been experimentally confirmed only in 2014, from Appleby et al.19 The importance of the work developed by Datta and coworkers is not only related to the class of ferroelectric materials. Indeed, in their model it is also stated that every microscopic mechanism, able to provide the required positive feedback for voltage amplification to occur, can be considered as a potential candidate for arising NC. For example, in the case of ferroelectric materials this positive feedback comes from electric dipoles interactions, which result in the presence of a non-zero inner polarization of the ferroelectric gate oxide. In our present work, we believe that the observed NC effects can be ascribed to a combination of different physical phenomena coexisting in ZnO, where charge trapping acts at low frequencies, while piezoelectricity act at frequencies as high as 1 MHz according to our measurements. The presence of the metal/oxide junction at the interface between the Au electrodes and the ZnO layer causes the appearance of interface charge states. Under suitable bias conditions, a loss of interface charges from occupied states below the Fermi level appears. The presence of oxygen vacancies in the ZnO layer, witnessed by XPS analyses, strongly promotes the presence of NC


20

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


phenomena too. To this purpose, the relaxation material theory can provide a valid explanation of NC effects at the hands of oxygen vacancies.41, 42 When an electric bias is applied between a couple of electrodes, electrical dipoles oriented along the electric field direction are created. At the metal/oxide interface oxygen vacancies, i.e. holes, can easily recombine with the dipole electrons near the junction, causing a decrease in the charge dipole itself. Moreover, the presence of a preferential (002) crystal orientation in the investigated material can be associated to the existence of piezoelectric phenomena in ZnO. Even though the resulting effects of piezoelectricity on the appearance of NC cannot be quantitatively estimated at this level, we believe that these cannot be totally excluded and are considered to play a key role in the promotion of the observed negative behavior. Moreover, also electrostriction phenomena can contribute to the observed NC. The appearance of such behavior is reported in Figure 4, and cannot be distinguished from piezoelectricity.

4. CONCLUSIONS In conclusion, we report on the appearance of negative capacitance phenomena in 80 nm-thick sputtered ZnO thin films, deposited by the RF magnetron sputtering technique on Au interdigital electrodes having different geometrical parameters. The morphology, crystal structure, and chemical composition analyses pointed out the presence of a fine-grained, nanometer-sized surface morphology in ZnO thin films, which are characterized by a preferential (002) crystal orientation as well as the right stoichiometry. Hall measurements evidenced the n-type behavior of ZnO, showing a low resistivity of 3.9×10-3 ohm·cm, with mobility and carrier concentration values of 62.7 cm2V-1s-1 and 2.54×1019 cm-3, respectively. In the DC regime we observed a


21


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

Page 22 of 30

typical semiconducting behavior, dependent on the geometry of the IDEs underlying ZnO. In the AC regime, the appearance of frequency-dependent negative values of the capacitance started to be visible, with very low values in the low frequency region (20 Hz – 120 Hz). Loss of interface charge states, together with the presence of oxygen vacancies as well as piezoelectric and electrostriction effects are believed to be at the basis of the observed negative phenomena, suggesting that ZnO-based FETs can be a powerful instrument to go beyond the Boltzmann limit and open new possibilities in the down-scaling of elemental integrated circuits required for the fabrication of portable and miniaturized electronic devices, as well for low-frequency electric household applications. Supporting Information Available: FESEM images of ZnO thin films sputtered on 30 and 100 µm-distanced IDEs. Phase plots of ZnO thin films sputtered on different IDE geometries (signal frequency: 10 and 100 kHz). This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *

Phone: +39 011 5091 903. E-Mail: [email protected]

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. The authors declare no competing financial interest. ACKNOWLEDGMENT


22

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


The authors gratefully acknowledge Dr. Marco Fontana and Dr. Micaela Castellino for performing FESEM and XPS measurements, respectively.

ABBREVIATIONS NC = negative capacitance IDEs = interdigital electrodes DC = direct current AC = alternating current FETs = field effect transistors ICs = integrated circuits SS = subthreshold slope ZnO = zinc oxide PZT = lead zirconate titanate 2DEG = two-dimensional electron gas MOS-HEMT = metal oxide semiconductor - high electron mobility transistor FESEM = field emission scanning electron microscopy XRD = X-ray diffraction XPS = X-ray photoelectron spectroscopy PES = piezo evaluation system RF = radio-frequency JCPDS-ICDD = Joint Committee on Powder Diffraction Standards-International Centre for Diffraction Data FWHM = full width at half maximum HR = high resolution BE = binding energy

REFERENCES


23


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 24 of 30

(1) Moore, G.E. Cramming More Components onto Integrated Circuits. Electronics (N. Y., NY, U. S.) 1965, S 0018-9219(98)00753-1, 114−117. (2) Koomey, J. G.; Berard, S.; Sanchez, M.; Wong, H. Implications of Historical Trends in the Electrical Efficiency of Computing. IEEE Ann. Hist. Comput. 2010, 33, 46−54. (3) Korucu D.; Turut, A.; Altındal, Ş. The Origin of Negative Capacitance in Au/n-GaAs Schottky Barrier Diodes (SBDs) Prepared by Photolithography Technique in the Wide Frequency Range. Curr. Appl. Phys. 2013, 13, 1101−1108. (4) Mora-Seró, I.; Bisquert, J.; Fabregat-Santiago, F.; Garcia-Belmonte, G. Implications of the Negative Capacitance Observed at Forward Bias in Nanocomposite and Polycrystalline Solar Cells. Nano Lett. 2006, 6, 640−650. (5) Wu, X.; Yang, E. S.; Evans, H. L. Negative Capacitance at Metal‐Semiconductor Interfaces. J. Appl. Phys. (Melville, NY, U. S.) 1990, 68, 2845−2848. (6) van den Biesen, J. J. H. Modelling the Inductive Behaviour of Short-Base p-n Junction Diodes at High Forward Bias. Solid-State Electron. 1990, 33, 1471-1476. (7) Ehrenfreund, E.; Lungenschmied, C.: Dennler, G.; Neugebauer, H.; Sariciftci, N. S. Negative Capacitance in Organic Semiconductor Devices: Bipolar Injection and Charge Recombination Mechanism. Appl. Phys. Lett. 2007, 91, 012112-1−012112-3. (8) Altındal, Ş.; Uslu, H. The Origin of Anomalous Peak and Negative Capacitance in the Forward Bias Capacitance-Voltage Characteristics of Au/PVA/n-Si Structures. J. Appl. Phys. (Melville, NY, U. S.) 2011, 109, 074503-1−074503-7. (9) Chiolerio, A.; Bocchini, S.; Porro, S. Inkjet Printed Negative Supercapacitors: Synthesis of Polyaniline-Based Inks, Doping Agent Effect, and Advanced Electronic Devices Applications. Adv. Funct. Mater. 2014, 24, 3375−3383.


24

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


(10)

Bocchini, S.; Chiolerio, A.; Porro, S.; Accardo, D.; Garino, N.; Bejtka, K.;

Perrone, D.; Pirri, C. F. Synthesis of Polyaniline-Based Inks, Doping Thereof and Test Device Printing towards Electronic Applications. J. Mater. Chem. C 2013, 1, 5101−5109. (11)

Kwok, H. L. Modeling Negative Capacitance Effect in Organic Polymers. Solid-

State Electron. 2003, 47, 1089−1093. (12)

Hu, P.; Lu, J. Q.; Wu, S. X.; Lv, Q. B.; Li, S. W. Coexistence of Memristive

Behaviors and Negative Capacitance Effects in Single-Crystal TiO2 Thin-Film-Based Devices. IEEE Electron Device Lett. 2012, 33, 890–892. (13)

Joshua Yang, J.; Strukov, D. B.; Stewart, D. R. Memristive Devices for

Computing. Nat. Nanotechnol. 2013, 8, 13−24. (14)

Roppolo, I.; Bejtka, K.; Asvarov, A.;

Pirri, C. F.; Chiolerio, A. Resistive

Hysteresis in Nanocomposites Containing ZnO Nanoparticles: Mechanism Unveiled. paper submitted. (15)

Salahuddin, S.; Datta, S. Use of Negative Capacitance to Provide Voltage

Amplification for Low Power Nanoscale Devices. Nano Lett. 2008, 8, 405–410. (16)

Krowne, C. M.; Kirchoefer, S. W.; Chang, W.; Pond, J. M.; Alldredge, L. M. B.

Examination of the Possibility of Negative Capacitance Using Ferroelectric Materials in Solid State Electronic Devices. Nano Lett. 2011, 11, 988–992. (17)

Krowne, C. M. Evaluation of the Differential Capacitance for Ferroelectric

Materials Using either Charge Based or Energy Based Expressions. J. Adv. Dielectr. 2014, 4, 1450024-1–1450024-10.


25


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

(18)

Page 26 of 30

Masuduzzaman, M.; Ashraful Alam, M. Effective Nanometer Airgap of NEMS

Devices Using Negative Capacitance of Ferroelectric Materials. Nano Lett. 2014, 14, 3160−3165. (19)

Appleby, D. J. R.; Ponon, N. K.; Kwa, Kelvin S. K.; Zou, B.; Petrov, P. K.;

Wang, T.; Alford, N. M.; O’Neill, A. Experimental Observation of Negative Capacitance in Ferroelectrics at Room Temperature. Nano Lett. 2014, 14, 3864−3868. (20)

Khan, A.I.; Chatterjee, K.; Wang, B.; Drapcho, S.; You, L.; Serrao, C.; Bakaul, S.

R.; Ramesh, R.; Salahuddin, S. Negative Capacitance in a Ferroelectric Capacitor. Nat. Mater. 2015, 14, 182–186. (21)

Jana, R. K.; Snider, G. L.; Jena, D. On the Possibility of Sub 60 mV/decade

Subthreshold Switching in Piezoelectric Gate Barrier Transistors. Phys. Status Solidi C 2013, 10, 1469–1472. (22)

Then, H. W.; Dasgupta, S.; Radosavljevic, M.; Chow, L.; Chu-Kung, B.; Dewey,

G.; Gardner, S.; Gao, X.; Kavalieros, J.; Mukherjee, N.; Metz, M.; Oliver, M.; Pillarisetty, R.; Rao, V.; Sung, S. H.; Yang, G.; Chau, R. Experimental Observation and Physics of “Negative” Capacitance and Steeper than 40 mV/decade Subthreshold Swing in Al0.83In0.17N/AlN/GaN MOS-HEMT on SiC Substrate. IEEE Int. Electron Devices Meet. 2013, 117, 28.3.1-28.3.4. (23)

Wong, J. C.; Salahuddin, S. Can Piezoelectricity Lead to Negative Capacitance?.

IEEE Int. Electron Devices Meet. 2014, 13.5.1-13.5.4. (24)

Kornev, I.; Willatzen, M.; Lassen, B.; Lew Yan Voon, L. C. Electrostriction

Coefficients of GaN, AlN, MgO and ZnO in the Wurtzite Structure from First‐Principles. AIP Conf. Proc. 2010, 71, 1199.


26

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


(25)

Laurenti, M.; Stassi, S.; Lorenzoni, M.; Fontana, M.; Canavese, G.; Cauda, V.;

Pirri, C. F. Evaluation of the Piezoelectric Properties and Voltage Generation of Flexible Zinc Oxide Thin Films. Nanotechnology 2015, 26, 215704-1–2157-9. (26)

Laurenti, M.; Garino, N.; Porro, S.; Fontana, M.; Gerbaldi, C. Zinc Oxide

Nanostructures by Chemical Vapour Deposition as Anodes for Li-Ion Batteries. J. Alloys Compd. 2015, 640, 321–326. (27)

Laurenti, M.; Canavese, G.; Sacco, A.; Fontana, M.; Bejtka, K.; Castellino, M.;

Pirri, C.F.; Cauda, V. Nanobranched ZnO Structure: p-Type Doping Induces Piezoelectric Voltage Generation and Ferroelectric-Photovoltaic Effect. Adv. Mater. (Weinheim, Ger.) 2015, 27, 4218–4223. (28)

Laurenti, M.; Cauda, V.; Gazia, R.; Fontana, M.; Rivera, V.F.; Bianco, S.;

Canavese, G. Wettability Control on ZnO Nanowires Driven by Seed Layer Properties. Eur. J. Inorg. Chem. 2013, 2013, 2520–2527. (29)

Laurenti, M.; Verna, A.; Fontana, M.; Quaglio, M.; Porro, S. Selective Growth of

ZnO Nanowires on Substrates Patterned by Photolithography and Inkjet Printing. Appl. Phys. A: Mater. Sci. Process. 2014, 117, 901–907. (30)

Gardeniers, J. G. E.; Rittersma, Z. M.; Burger, G. J. Preferred Orientation and

Piezoelectricity in Sputtered ZnO Films. J. Appl. Phys. (Melville, NY, U. S.) 1998, 83, 7844–7854. (31)

Fultz, B.; Howe, J. M. In Transmission Electron Microscopy and Diffractometry

of Materials; Fultz, B., Howe, J. M., Eds.; Springer-Verlag: Berlin Heidelberg, 2013; Chapter 1, pp 1–57.


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

(32)

Page 28 of 30

Yogamalar, R.; Srinivasan, R.; Vinu, A.; Ariga, K.; Chandra Bose, A. X-Ray

Peak Broadening Analysis in ZnO Nanoparticles. Solid State Commun. 2009, 149, 1919– 1923. (33)

Kohan, A. F.; Ceder, G.; Morgan, D.; Van de Walle, C. G. First-Principles Study

of Native Point Defects in ZnO. Phys. Rev. B: Condens. Matter Mater. Phys. 2000. 61, 15019–15027. (34)

Quaranta, F.; Valentini, A.; Rizzi, F. R.; Casamassima, G. Dual‐Ion‐Beam Sputter

Deposition of ZnO Films. J. Appl. Phys. (Melville, NY, U. S.) 1993, 74, 244–248. (35)

Saw, K. G.; Ibrahim, K.; Lim, Y. T.; Chai, M. K. Self-Compensation in ZnO Thin

Films: an Insight from X-Ray Photoelectron Spectroscopy, Raman Spectroscopy and Time-of-Flight Secondary Ion Mass Spectroscopy Analyses. Thin Solid Films 2007, 515, 2879–2884. (36)

Klein, J. C.; Hercules, D. M. Surface Characterization of Model Urushibara

Catalysts. J. Catal. 1983, 82, 424–441. (37)

Bang, S.; Lee, S.; Park, J.; Park, S.; Jeong, W.; Jeon, H. Investigation of the

Effects of Interface Carrier Concentration on ZnO Thin Film Transistors Fabricated by Atomic Layer Deposition. J. Phys. D: Appl. Phys. 2009, 42, 235102.1–235102.6. (38)

Kaidashev, E. M.; Lorenz, M.; von Wenckstern, H.; Rahm, A.; Semmelhack, H. -

C.; Han, K. -H.; Benndorf, G.; Bundesmann, C.; Hochmuth, H.; Grundmann, M. High Electron Mobility of Epitaxial ZnO Thin Films on c-Plane Sapphire Grown by Multistep Pulsed-Laser Deposition. Appl. Phys. Lett. 2003, 82, 3901–3903. (39)

Jayah, N.A.; Yahaya, H.; Mahmood, M.R.; Terasako, T.; Yasui, K.; Hashim, A.

M. High Electron Mobility and Low Carrier Concentration of Hydrothermally Grown


28

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


ZnO Thin Films on Seeded a-Plane Sapphire at Low Temperature. Nanoscale Res. Lett. 2015, 10:7, 1–10. (40)

Baxter, J. B.; Schmuttenmaer, C. A. Conductivity of ZnO Nanowires,

Nanoparticles, and Thin Films Using Time-Resolved Terahertz Spectroscopy. J. Phys. Chem. B 2006, 110, 25229–25239. (41)

Gökçen, M.; Altuntaş, H.; Altındal, Ş.; Özçelik, S. Frequency and Voltage

Dependence of Negative Capacitance in Au/SiO2/n-GaAs Structures. Mater. Sci. Semicond. Process. 2012, 15, 41–46. (42)

Jones, B. K.;

Santana,

J.; McPherson, M. Negative Capacitance Effects in

Semiconductor Diodes. Solid State Commun. 1998, 107, 47–50.


29


Page 30 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 Paragon Plus Environment

30

Evidence of Negative Capacitance in Piezoelectric ZnO Thin Films

Recommend Documents