Al Trilayers

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Probing the Dielectric Properties of Ultrathin Al/Al2O3/Al Trilayers Fabricated Using in situ Sputtering and Atomic Layer Deposition Jagaran Acharya, Jamie Samantha Wilt, Bo Liu, and Judy Z. Wu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b16506 • Publication Date (Web): 02 Jan 2018 Downloaded from http://pubs.acs.org on January 3, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a 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 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

ACS Applied Materials & Interfaces

Probing the Dielectric Properties of Ultrathin Al/Al2O3/Al Trilayers Fabricated Using in situ Sputtering and Atomic Layer Deposition Jagaran Acharya,* Jamie Wilt, Bo Liu and Judy Wu* 1

Department of Physics and Astronomy, University of Kansas, Lawrence, Kansas, 66045, USA Corresponding Authors Jagaran Acharya, email: [email protected] Judy Wu, email: [email protected]

Keywords: Atomic layer deposition, ultrathin film, dielectric properties, interfacial layer, capacitors

ABSTRACT Dielectric properties of ultrathin Al2O3 (1.1-4.4 nm) in metal-insulator-metal (M-I-M) Al/ Al2O3/Al trilayers fabricated in situ using an integrated sputtering and atomic layer deposition (ALD) system were investigated. An M-I interfacial layer (IL) formed during the pre-ALD sample transfer even under high vacuum has a profound effect on the dielectric properties of the Al2O3 with a significantly reduced dielectric constant (εr) of 0.5-3.3 as compared to the bulk εr ~9.2. Moreover, the observed soft-type electric breakdown suggests defects in both the M-I interface and the Al2O3 film. By controlling the pre-ALD exposure to reduce the IL to a negligible level, a high

εr up to 8.9 was obtained on the ALD Al2O3 films with thicknesses from 3.3-4.4 nm, corresponding to an effective oxide thickness (EOT) of ~1.4-1.9 nm respectively, which are comparable to the EOTs found in high-K dielectrics like HfO2 at 3-4 nm in thickness, and further suggest that the ultrathin ALD Al2O3 produced in optimal conditions may provide a low-cost alternative gate dielectric for CMOS. While εr decreases at a smaller Al2O3 thickness, the hardtype dielectric breakdown at 32 MV/cm and in situ scanning tunneling spectroscopy revealed band gap ~2.63 eV comparable to that of an epitaxial Al2O3 film. This suggests that the IL is unlikely a dominant reason for the reduced εr at the Al2O3 thickness of 1.1-2.2 nm, rather a consequence of the electron tunneling as confirmed in the transport measurement. This result demonstrates the critical importance in controlling the IL to achieving high-performance ultrathin dielectric in MIM structures.

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

INTRODUCTION With the advancement and miniaturization of microelectronics following Moore’s law in the last few decades, the need for pinhole and defect-free dielectrics with thickness approaching sub-nm to a few nm (ultrathin) has become critical to the future of microelectronics including insulating tunnel barriers in metal–insulator–metal (MIM) tunnel junctions1-6 and gate dielectrics in complementary metal-oxide-semiconductor (CMOS) technology.7-10 The dielectric layer in these devices made using ex situ processes have a dielectric constant (εr) significantly lower than the bulk single crystals values when the dielectric layer thickness is on the order of tens of nanometers or lower.8-9, 11-18 This trend indicates the presence of a metal-insulator-interfacial layer (M-I IL), which when connected in series with the dielectric film can significantly reduce

εr and the reduction becomes more severe at smaller thicknesses. Indeed, a monotonic decrease of εr with decreasing dielectric film thickness has been observed when the dielectric film thickness falls into the ultrathin regime.11, 16, 19 8-9, 17 An in situ deposition processes for MIM structures can reduce the formation of an IL by reducing exposure of sample surface to atmosphere during MIM fabrication. The development of various in situ deposition techniques such as magnetron sputtering,20-22 molecular beam epitaxy,1, 6 and atomic layer deposition (ALD)7, 10, 22-25 have been the focus of research recently for various applications including superconductor Josephson junctions for quantum information technology, magnetic tunnel junctions for data storage, and high K-dielectrics for CMOS technology. Among others, ALD is particular interesting due to several unique advantages.26-27 First, ALD is a chemical vapor deposition process that relies on well-defined chemical reactions, which occur only at the sample surface. This chemical process minimizes formation of the intermediate compounds in vapor form and is important to reduce defects and impurities in the


Page 2 of 23

Page 3 of 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


grown film. Second, ALD growth is self-limiting, enabling atomic control of the film thickness. Finally, ALD coating is highly conformal,28-29 which is important to obtain pinhole-free ultrathin dielectric films. Despite the exciting progress made in the ALD of dielectric thin films, such as Al2O3, HfO2, ZrO2, MgO, Al-doped ZnO,7-10, 17-18, 20, 22, 25, 27, 29 the dielectric properties of ultrathin insulating films in MIM structures are not optimal; most probably due to the influence of the IL at the M-I interface.27, 30 Taking Al2O3 as an example, the initiation of ALD Al2O3 growth on metal typically requires a monolayer of hydroxyl groups formed controllably on a metal surface, which is by no means trivial. Metals are typically classified into two categories: noble (Au, Pt, Ir, and Ru) and reactive (Al, Nb, Fe, and Co). In the former case, the first 30-50 ALD cycles serve as the so-called incubation period since ALD growth cannot be initiated until the surface hydroxylation is complete.20, 30 Unfortunately, this incubation period typically yields a defective M-I IL.11, 20, 31 On the other hand, reactive metals are sensitive to air and other chemicals during the ALD process. A defective surface layer, which is several nanometers in thickness,11, 20, 31 can form upon exposure of the reactive metal surface to air or even low vacuum.20, 22 This sensitivity means a defective IL at the M-I interface may form before the ultrathin ALD dielectric film growth. This IL can in turn cause defects in the ALD dielectric film. The IL is detrimental to the dielectric properties, especially as the dielectric layer approaches the ultrathin range of a few nanometers. In fact, the best-reported dielectric constant,

εr, is in the range of 7-8.5 for ALD Al2O3 films with thickness exceeding 40 nm, together with a significant leakage current density (Jleak) in the range of 10-10-10-8 A/cm2.11-15,

32-33

The εr

decreases monotonically with further decrease in ALD Al2O3 film thickness.11, 16, 19 For example, Groner et al reported εr ~7.7 for a 60 nm thick ALD Al2O3 film, which decreases to ~5.9 and 4



at 12 nm and 3 nm thicknesses, respectively.11 The leakage current increases to a high value of J ~10-7 A/cm2 as the Al2O3 dielectric thickness approach ultrathin thickness range approximately 6.5 nm

32

and 3 nm.11 A similar trend has also been observed in high-K dielectric films. For

example, dielectric constants in the range of 8-20 were found for ultrathin HfO2 films of thicknesses 1-4.5 nm, which is considerably lower than the bulk value of 25.8-9, 17 These results indicate that the IL can strongly degrade the dielectric properties of the MIM trilayer structure, preventing high-quality from being achieved in ultrathin dielectric films. In order to address this critical issue, this work explores the effect of controlling the preALD exposure of the metal surface34-35 to minimize the IL and its detrimental impact on ultrathin ALD Al2O3 dielectric thicknesses (1.1 – 4.4 nm) on an Al electrode using our integrated in situ sputtering/ALD system.25 The dielectric properties of these ultrathin ALD Al2O3 films were investigated using ex situ capacitance-voltage (C-V) characterization in combination with an in situ scanning tunneling spectroscopy (STS) measurement. This study demonstrates that the dielectric properties of the ALD Al2O3 films strongly depend on the IL. Remarkably, an εr up to 8.9, within 3% of the bulk value, was achieved by our ultrathin ALD Al2O3 dielectric films of thicknesses ~ 3.3-4.4 nm when the IL effect is negligible. We show that a decrease of εr at smaller thicknesses is primarily caused by electron tunneling, which is supported by the band gap of ~2.63 eV comparable to that of an epitaxial Al2O3 film and hard-type dielectric breakdown at 1.1 nm thickness.

RESULTS AND DISCUSSION In situ ultra-high vacuum deposition and mechanism for ALD Al2O3 growth Figure 1(a) shows a schematic diagram of in-house integrated in situ ultra-high vacuum sputtering and ALD Al2O3 fabrication chamber which allows for the deposition of ultrathin 4 ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 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


dielectrics in MIM trilayers with minimal IL growth. The IL is impossible to minimize with ex situ MIM deposition, as the metal electrodes will be exposed to air before growth of dielectric film. Thus to minimize the formation of the IL, it is critical to deposit all layers of the MIM trilayer in situ under UHV. This system is also equipped with in situ STS measurement to characterize the dielectric property of Al/Al2O3 film without breaking vacuum. Figure 1(b) shows the step-by-step growth of our ALD Al2O3 dielectric films in the optimal case with no IL. First, H2O precursor is pulsed into chamber to create a hydroxylated surface on the Al. A purge step follows after the H2O reacts with surface. Next TMA precursor is pulsed into chamber and purged after CH4 is produced as a byproduct. Last, H2O is pulsed to produce single layer of Al2O3. The deposited Al2O3 thickness is controlled by varying the number of ALD cycles. Each ALD cycle produce a single conformal monolayer which is about 0.11nm thick.20 The thickness of the ALD Al2O3 film per ALD cycle was calibrated using ellipsometry to be 1.1-1.2 Å/cycle,20, 25

which is consistent with the previous reports by other groups.11, 27, 36 Figure 2 (left) shows the

schematic diagram of MIM trilayer capacitors defined using a shadow mask and (right) shows a cross-sectional view of the device in which no significant IL is assumed.35 During C-V measurement the tungsten probe scratches through the top few nanometers of the ALD Al2O3 layer to make an ohmic contact with the bottom electrode. Thus, the measurement is made between top and bottom electrode to define Al/Al2O3/Al trilayer capacitors with area of about 0.08, 0.06, and 0.04 mm2.



Figure 1. Schematic diagrams of (a) in-house integrated in situ ALD-UHV sputtering for fabrication of Al/Al2O3/Al trilayers along with in situ STS characterization (b) growth mechanism for ALD Al2O3 with optimal ALD parameters without an interfacial layer


Page 6 of 23

Page 7 of 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


Figure 2: (Left) Schematic diagram (viewed from above) of the bottom electrode(yellow) with a shadow mask used to define the 400, 300 and 200 µm wide bridges and the top electrode (brown) with a 200 µm wide bridge to define the corresponding MIM capacitors with 400x200, 300x200 and 200x200 µm2 cross-sectional areas. (Right) shows a cross sectional view of the MIM tri-layer structure Ex situ C-V measurement and dielectric properties Figure 3(a) shows the measured capacitance for Al/Al2O3/Al trilayer capacitors of area 0.08 mm2 as function of the electric field E=V/t. Here t is the thickness of the ALD Al2O3 layer and V is the applied voltage. We consider samples with 10-40 ALD Al2O3 cycles, or about 1.1-4.4nm in thickness. E was restricted to ~2 MV/cm for all samples to avoid dielectric breakdown. At larger thicknesses of about 1.65-4.4 nm, the capacitance is independent of the applied voltage before dielectric breakdown. This constant capacitance indicates that high-quality Al2O3 dielectric growth was achieved. At the smallest thickness of 1.1 nm, the capacitance has a moderate E dependence. Specifically, capacitance decreases with increasing E, which suggests an increase in the leakage current due to electron tunneling.37



The variation of specific capacitance (Co) with junction area of 0.08, 0.06 and 0.04 mm2 MIM trilayer samples for ALD Al2O3 thickness in the range 1.1- 4.4 nm with a negligible IL is shown in Figure 3(b). The Co is expected to be a constant for ideal capacitors. A 10% variation of the specific capacitance was observed for 4.4 (black), 3.3 (red) and 2.2 nm (blue) thick Al2O3 chips. This variation is comparable to the capacitor area variation of the shadow mask. This result suggests that the ALD Al2O3 films are highly uniform in this thickness range. However, a higher variation of ~20-30% was observed on the 1.65 (dark cyan) and 1.1 (magenta) nm ALD Al2O3 chips, which is not surprising since the effect of defects and pinholes will be amplified at such a small thickness. From Figure 3(a), it can be observed that the capacitance vs. thickness is not proportional to 1/t, as expected in ideal capacitors. To shed light on the mechanisms behind this decreasing trend, the dielectric constant was calculated from the measured C values using the equation C=ε0εrA/t, where ε0 is the permittivity of free space and A is area of the capacitor. Figure 3(c) compares the calculated εr of the ALD Al2O3 films made with the optimal34 (thickness in the range of 1.1-4.4 nm) and non-optimal35 (at two different thicknesses of 1.1 nm and 4.4 nm) ALD conditions. The

εr values for the former are significantly higher than in the latter, suggesting that the dielectric properties of the thin ALD Al2O3 films strongly depends on the presence of the IL. 57 With optimal ALD conditions the IL is negligible, and εr is around 8.9 for 3.3–4.4 nm thick ALD Al2O3, corresponding to an effective oxide thickness (EOT=tHiK·3.9/εHik) ~1.4-1.9 nm, respectively, where tHiK and εHik are the thickness and dielectric constant of high-K materials. These EOTs are comparable to the EOTs of a high-K dielectric such as 3-4.5 nm thick HfO2 8-9 and this suggests that the optimal ultrathin ALD Al2O3 may provide a low-cost alternative gate dielectric for CMOS. Furthermore, this dielectric constant of 8.9 for 3.3-4.4 nm thick ALD 8 ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 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


Al2O3 films is only about 3.3% lower than the bulk Al2O3 value of ~9.216 and more than double the best previously reported value (εr ~4.0) for 3 nm thick ALD Al2O3.11 In contrast, with nonoptimal ALD conditions at which an IL is present, εr is considerably lower with a value in the range of 2.5-3.3 for 4.4 nm thick ALD Al2O3 as shown in Figure 3(c). The difference in εr illustrates the significant effect that an M-I IL has on the dielectric property of thin dielectric films. Specifically, a defective IL is in series with the ALD Al2O3 layer. This IL degrades the dielectric properties of the composite IL/ALD Al2O3 film. In addition to introducing a poorquality IL capacitor, the ALD Al2O3 film grown on top of an IL is defective, which can further decrease the dielectric constant. Even at a substantially larger thickness of ~60 nm, the εr was reported to be just 7.6 for ex situ deposited ALD Al2O3 film on n-type Si and Mo coated Sisubstrate.11 An IL of ~1.1 nm was reported to form on n-type Si substrate and a similar IL is suspected to form on Mo-coated Si-substrate.11 Since the IL is difficult to avoid in ex situ fabricated MIM trilayers, it is reasonable to expect lower εr in the samples with more significant IL growth. The εr ~ 7.2 is obtained for 2.2 nm ALD Al2O3 film without IL which corresponds to EOT~1.2 nm comparable to the EOT of a 4.5 nm of HfO2 high K dielectric film.17 This result demonstrates the feasibility of incorporating ultrathin ALD Al2O3 with other high-K material as a gate dielectric in CMOS. However, significant decrease in εr in the range of 0.4-0.9 was observed for the 2.2 nm ALD Al2O3 film with non-optimal ALD conditions when an IL was present. In fact, decreases in εr with the film thickness have been observed for the high-K dielectric HfO28-9, 17 with thicknesses in the range of 1-4.5nm and also for ZrO2 films26,46 in a comparable thickness range of 4-6.5 nm. This result illustrates the critical importance of the elimination of the IL in obtaining high-quality ultrathin dielectric films. The significantly reduced εr values observed for the ALD Al2O3 thin films with poor dielectric quality fabricated 9 ACS Paragon Plus Environment


Page 10 of 23

either in situ with a non-optimized condition or ex situ is due to the presence of the IL. When the IL is negligible, εr remains constant for the ALD Al2O3 films with 3.3 and 4.4 nm thickness. However, when thickness is further reduced below 3 nm, a monotonic decrease of εr with the dielectric film thickness is observed in Figure 3(c) especially in the thickness range of 1.1-2.2 nm. To rule out the possibility that a thin IL with a substantial thickness that could play a more significant role in this thickness range, a fitting of the measured capacitance ( ) as function of the dielectric thickness was carried out with an assumption that an IL capacitor ( ) is connected in series with an ideal capacitor for ALD Al2O3 films ( ) described by the equation (1).

=

+

Equation (1)

The specific measured capacitance can then be calculated using equation (2).

=

ε

Equation (2)

ε

ε

where and are thicknesses of the IL and ALD Al2O3 films in nm, and ε and ε are the dielectric constants for the interfacial layer and the ALD Al2O3 dielectric films, respectively. (in ) and ε are unknown parameters and several values of their ratio were used in the fitting shown in Figure 3(d). The ratio

ε

ε

from 0.01 to 0.3

= 0.01 curve (black) provides the best

fit to the specific measured capacitance of the 4.4, 3.3 and 2.2 nm thick ALD Al2O3 films. Assuming ε = 1 or 2 for the defective IL, the IL thickness would be = 0.1 Å and 0.2 Å respectively. This means that the IL is indeed negligible and the decrease in the measured specific capacitance of the ALD Al2O3 films of 1.1-2.2 nm is primarily due to the electron tunneling. In this case, the circuit could be modeled as a resistor connected in parallel with the Al/ALD Al2O3/Al capacitor schematically shown in the inset of Figure 3(d). Our results show 10 ACS Paragon Plus Environment

Page 11 of 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


that 30 Ω resistor connected in parallel with a 3.7 nF capacitor (comparable to that of the 1.65 nm ALD Al2O3 film) results in a decrease in capacitance to 2 nF. This result suggests that the decrease in the measured specific capacitance of the 1.1-2.2 nm thick ALD Al2O3 dielectrics is due to tunneling of electrons. In order to confirm this argument, we have carried out an ac impedance measurement on the MIM samples. Figure S1 summarizes the impedance results measured on the MIM samples with 4.4 nm (40 cycle) and 1.65 nm (15 cycle) ALD Al2O3 in thickness. In the former the circuit is expected to consist of only a capacitor, while in the latter, a combined circuit of a resistor connected in parallel with the Al/ALD Al2O3/Al capacitor is anticipated considering the electron tunneling as shown in inset for Figure 3(d). This expectation has been confirmed experimentally in the impedance measurement as shown in Figures S1 (a) and (b) respectively the Nyquist plot and phase variation of the 4.4nm (black) and 1.65 nm (blue) thick ALD Al2O3 MIM samples. Quantitatively, the values of capacitance at 1 KHz for the 4.4 nm and 1.65 nm samples are 1.4 and 2.3 nF respectively, which are in good agreement with that obtained from the C-V measurement in Figure 3 (a). The resistance of 52 ohm for the 1.65 nm sample is consistent with that used in inset in Figure 3(d).



Page 12 of 23

Figure 3: Variation of (a) Capacitance vs. electric field (C-E) for ALD Al2O3 fabricated with optimal ALD conditions for junction areas of 0.08 mm2 (b) Specific capacitance vs. junction area for 10-40 cycles of ALD Al2O3 fabricated with optimal ALD conditions (c) Variation of the dielectric constant for both optimal and non-optimal ALD condition, (d) Modeling of specific capacitance for ALD Al2O3 using interfacial layer capacitance. The inset shows a decrease in the measured capacitance for a standard 3.7 nF capacitor connected in parallel with a 30 Ω resistor

We now focus our discussion on the Jleak characteristic of three different dielectric films with an applied electric field considering ALD Al2O3 dielectric films with thicknesses 4.4, 2.2 and 1.1 nm as shown in Figure 4 (a). The 4.4 nm (black) thick ALD Al2O3 film with a negligible IL has a low Jleak~10-9A/cm2 at zero bias, which increases to ~10-7A/cm2 with an increase in E to 2 MV/cm. This value of Jleak at zero bias is two orders lower than that of the 6.5nm comparable to the Jleak of the 12nm

11

32

and

for ex-situ fabricated ALD Al2O3 films, which is an

indicator of the higher quality of in situ fabricated tunnel barriers with a negligible IL. For 2.2 nm thick ALD Al2O3 film (red) Jleak~10-6A/cm2 at zero bias increases to 10-4 A/cm2 with the corresponding increase in E to 2 MV/cm. A similar increasing trend in Jleak is observed for 1.1 nm thick Al2O3 films (black) with Jleak~10-3 A/cm2 at zero bias, which significantly increases to ~1 A/cm2 at 2 MV/cm.57 This increase in Jleak is mainly due to quantum tunneling of electrons that start to dominate at dielectric thickness below 2nm as expected for ultrathin dielectric films.11, 37 Thus, with a decrease in the thickness of ALD Al2O3 dielectric films MIMTJs are more susceptible to higher leakage current even at smaller E. This trend has nothing to do with the quality of the tunnel barrier as explained in discussion Figure 3. This argument is supported with an in situ STS study to be discussed in detail later. Due to the higher quality in the tunnel


Page 13 of 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


barrier through reduction of the IL and in situ ALD fabrication, we can push the limit of our C-V measurement to ultrathin ALD Al2O3 thickness, which is not only an extension of previous work done by many research groups11, 32-33 but also an important application in CMOS and MIMTJs. C-V measurements were attempted for 5 cycles ALD Al2O3, however a significant increase in leakage current led to a significantly reduced capacitance which was below the sensitivity limits of our apparatus. The breakdown dielectric behavior is studied using leakage current density vs. applied electric field (Jleak-E) by increasing E until the dielectric film shows a sudden increase in Jleak. This is an important parameter to evaluate the strength of the ALD Al2O3 dielectric film. Figure 4(b) shows Jleak ~10-7 A/cm2 (black) for the 4.4 nm thick ALD Al2O3 film with E~2 MV/cm which significantly increases to 10-5 A/cm2 (blue) due to the corresponding increase in leakage current when E approaches 5 MV/cm, this is known as soft dielectric breakdown.38-39 The dielectric film recovers its insulating property after the removal of E. Figures 4(c) and (d) show hard dielectric breakdown electric field (EHBR) which makes the dielectric film lose its insulating property and become conductive. The 4.4 nm thick ALD Al2O3 film (black) shows EHBR ~6.2 MV/cm comparable to ~7-8 MV/cm reported for a thicker film ~100 nm11, 32 indicating the better quality of these ultrathin dielectric ALD Al2O3 films fabricated using our in situ sputtering and ALD system.25 The enhancement in EHBR ~10 MV/cm is observed when the film thickness is reduced to 2.2 nm (blue). The EHBR further increases to ~32 MV/cm for 1.1 nm ALD Al2O3 film, which is consistent with the enhancement of EHBR ~30 MV/cm observed for 1.2nm ALD Al2O3 film.37, 40 The significant increase in EHBR.is related to an increase in leakage current due to the quantum tunneling of electrons through the ultrathin dielectric that prevents ALD Al2O3 films from hard dielectric breakdown at low E.37, 40 Thus, these observed high EHBR values in ultrathin ALD



Al2O3 films show the potential for reducing the thickness of gate dielectrics used in CMOS and improving performance in MIMTJs. Taking advantage of the unique completely in situ fabrication and characterization system, we further studied quality of tunnel junction using in situ STS measurement.

Figure 4: Leakage current density vs. electric field (J-E) fabricated using optimal ALD condition (a) for 40, 20 and 10 cycles ALD Al2O3 at low electric field (b) 40 cycles before and after soft


Page 14 of 23

Page 15 of 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


dielectric breakdown (c) 40 and 20 ALD Al2O3 cycles after hard dielectric breakdown and (d) 10 ALD Al2O3 cycles after hard dielectric breakdown

In situ STS measurement and dielectric properties Ex situ tunnel junction and capacitance measurements are limited to insulators which are greater than 1nm in thickness due to the requirement of low leakage current.7, 11, 25, 37, 40 In situ STS is not limited by such constraints. In fact, a high leakage current is required for STS to measure insulators, thus limiting the maximum measurable insulator thickness, for high-k dielectrics, to around 2-3 nm.41 Therefore, we used STS to quantify the ultrathin insulators in tandem with the ex situ tunnel junction and capacitance measurements. By varying the bias voltage applied between the sample and an atomically-sharp metallic tip and recording the tunneling current along with it’s derivative, STS can probe the local density of states of the insulator.42 To make electrical contact for the bias voltage, a molybdenum washer was mechanically clamped to the Si/Au substrate as shown in Figure 5(a). The M-I structure was then grown on top and the sample transferred to the STS chamber in situ under high vacuum to avoid additional alumina growth. Prior STS study of the ALD Al2O3 tunnel barrier on Al has shown that the tunneling barrier height, defined as the conduction band minimum(CBM), is constant with Al2O3 thickness in the range of 1-10 ALD cycles with a value of about 1.5 eV.35 This barrier height constancy with thickness is an indicator that the quality of our ALD Al2O3 tunnel barrier does not change with thickness. The interfacial layer (IL) formed primarily due to exposure of Al to trace O2 and/or H2O during the pre-ALD heating step was systematically reduced by controlling the heating parameters to obtain an increase in ALD Al2O3 barrier height from about 1.0 eV to about 1.5 eV



(with a negligible IL) along with a transition from soft-type to hard-type dielectric breakdown.3435

Therefore, the ALD Al2O3 band gap, which is another good indicator for the quality of the

insulator, must be constant as well. A representative STS dI/dV spectrum is shown in Figure 5(b). The band gap is defined as the difference between the CBM and the valence band maximum (VBM). On this log dI/dV scale the Valence and conduction bands should be roughly linear.43 The band gap region of the STS dI/dV curve is nearly flat, indicating a low leakage current through the insulator.21 We fit linear lines (bisquare method) to the valence band and the conduction band. The VBM and CBM were calculated to be about -1.0 eV and 1.6 eV respectively and were defined as the linear fit line’s intersection with the nearly flat band gap region. We found that the ALD Al2O3 band gap was 2.63 eV +/- 0.30 eV. This band gap value is actually comparable to the ultrathin (1.3 nm) α-Al2O3 band gap of 2– 4 eV.21, 44 Together with the C-V measurements, this excellent band gap indicates that high quality ultrathin ALD Al2O3 growth has been achieved when the IL is minimized. To expand upon the dielectric breakdown characteristics observed in Figure 4, the STS bias voltage was repeatedly ramped up and down with the tip height fixed over one location until dielectric breakdown. I-V and corresponding dI/dV spectra are shown in Figure 5(c) and 5(d) respectively for 10 cycle ALD Al2O3 films with a minimized IL. Our ultrathin ALD Al2O3 tunnel barrier broke down in a hard-type breakdown behavior where the tunneling current suddenly increased by a factor of ~1000. This breakdown event occurred near the end of the 13th spectra and can be clearly seen in the corresponding IV curve (Figure 5(c)). Subsequent spectra had tunneling current which was above the saturation current of the STM (~100 nA). This type of breakdown behavior is consistent with epitaxial Al2O3, indicating that no significant defective IL is present in our ALD Al2O3 tunnel barriers.45


Page 16 of 23

Page 17 of 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


Figure 5: In Situ Scanning tunneling spectroscopy for the ALD Al2O3 tunnel barrier (a) A schematic for the sample mounting scheme (b) A representative of dI/dV spectra with blue and red lines are bisquare method linear fits to the valence band and the conduction band respectively (c) The breakdown characteristics with I-V and (d) corresponding dI/dV spectra where the bias voltage was sequentially ramped up and down with a fixed tip-sample distance. The legend denotes which spectrum in the sequence.

CONCLUSIONS



In summary, we have investigated the effect of the M-I interface on the dielectric properties of in situ fabricated Al/Al2O3/Al trilayers with ultrathin (1.1 nm-4.4 nm) ALD Al2O3 dielectric films. Several important observations have been made. First, an IL can still form at the Al/Al2O3 interface in high vacuum at a non-optimal ALD growth condition and it has a profound effect on the dielectric properties of the ALD Al2O3 dielectric films. Specifically, the IL at the Al/Al2O3 interface is highly defective AlOx on which the ALD Al2O3 is defective, resulting in a low dielectric constant of ~3.3 for the 4.4 nm thick ALD Al2O3 films and soft-type dielectric breakdown ascribed to the mobile defects in the IL/ALD Al2O3. On the other hand, the IL can be systematically controlled by preventing pre-ALD exposure of trace amounts of H2O, oxygen and other chemical species in vacuum. For the 3.3-4.4 nm thick ALD Al2O3 dielectric films with a negligible IL, we obtained high εr ~8.9 approaching that of the bulk Al2O3 (9.2) corresponding to EOT ~1.4-1.9 nm respectively and low leakage current density ~10-9 A/cm2 at zero bias. This high-quality dielectric property is further confirmed by the hard-type dielectric breakdown with a breakdown electric field up to ~32 MV/cm on the 1.1 nm thick ALD Al2O3. These results illustrate the critical importance in controlling the M-I interface in order to obtain high quality dielectric films. It also provides the feasibility to reduce the thickness of gate dielectrics for CMOS and MIMTJs, which are important to a large variety of microelectronic applications.

MATERIALS AND METHODS To create the MIM trilayer structure for the C-V measurements, a bilayer of Nb (20 nm)/Al (7 nm) was DC magnetron sputtered onto a Si/SiO2 substrate using a shadow mask with a Nb and Al deposition rate of 1.7 nm/s and 0.5 nm/s respectively. After sputtering, the samples were transferred in situ under high vacuum to the ALD chamber and dynamically heated for 15 min, which was the optimal condition identified in our previous work to minimize IL formation.35 Following this dynamic heating, 10, 20 and 40 cycles ALD Al2O3 were deposited with a 5 sccm 18 ACS Paragon Plus Environment

Page 18 of 23

Page 19 of 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


N2 carrier gas for the dielectric tunnel barrier deposition. The ALD growth of Al2O3 occurs with alternating pulses of H2O and trimethylaluminum (TMA) via a ligand exchange at the heated sample surface.34 The two precursors are separated by a purge of an inert carrier gas N2

27

to

insure the formation a single monolayer of amorphous Al2O3 which is ~1.1 Å in thicknesses.20, 25 After short exposure in air for 10 min, a 100 nm thick Al top contact was defined through another shadow mask. On each MIM Al/Al2O3/Al trilayer sample, three devices of 0.08, 0.06 and 0.04 mm2 areas, respectively, were obtained for C-V measurements using tungsten probes (Lakeshore) in a probe station. An Agilent semiconductor analyzer was employed for the C-V measurement and the leakage current vs. voltage (I-V) characteristic. During the C-V measurement, transient displacement current was measured in response to a linear voltage ramp performed by superimposing a small sinusoidal oscillating signal of 30 mV (1 KHz) on the voltage sweep applied across the capacitor. During the I-V measurement, only steady-state or true conduction current through the MIM trilayer was recorded. For STS characterization, the Al/Al2O3 bilayers (without top Al electrode) were transferred in situ after the ALD fabrication of the Al2O3 of ~ 1 nm in thickness, to the STM chamber at the pressure of about 2×10-10 Torr. A mechanically cleaved Pt-Ir tip was used for all STS studies at room temperature. The constant height I-V and dI/dV spectroscopy were taken simultaneously using a lock-in amplifier with a voltage modulation of 100 mV at 1 kHz. The CBM, tunnel barrier height and VBM were estimated by the intersection of two bisquare-method linear fits to ln(dI/dV) similar to the method reported in.43 One line fits the band gap regime, and the other the CBM and VBM. This ln(dI/dV) linear fit method was chosen over I-V or (dI/dV)/(I/V) fit methods for it’s insensitivity to high noise in the STS spectra.5, 46 To examine the dielectric



breakdown of the Al2O3 tunnel barrier, the STS tip was held fixed at each scanned location and the bias was sequentially ramped up and down 20 times. ASSOCIATED CONTENT Supporting Information The following files are available free of charge. 1. AC impedance measurement for Ultrathin Al/Al2O3/Al Trilayers Fabricated Using in situ Sputtering and Atomic Layer Deposition. (File type PDF) AUTHOR INFORMATION Corresponding Authors * [email protected] * [email protected] ORCID Jagaran Acharya: 0000-0003-1129-0974 Judy Wu: 0000-0001-7040-4420 Author Contributions J.A. and J.Z.W. designed the experiment. J.A. prepared the samples for MIM and performed the dielectric properties characterization and most of the analysis. B. L. contributed with some MIM device fabrication and measurement. J.S.W. helped with STS sample fabrication and measurement. All authors contributed for the discussion of results. J.A. and J.Z.W. led the effort in development of the manuscript. Funding Sources The authors acknowledge support in part by US Army Research Office contract No. AROW911NF-16-1-0029, and National Science Foundation contracts Nos. NSF-DMR-1337737, and NSF-DMR-1508494. Notes The Authors declare no competing financial interest

ACKNOWLEGEMENT 20 ACS Paragon Plus Environment

Page 20 of 23

Page 21 of 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


The authors acknowledge support in part by US Army Research Office contract No. AROW911NF-16-1-0029, and National Science Foundation contracts Nos. NSF-DMR-1337737, and NSF-DMR-1508494. REFERENCES (1) Yuasa, S.; Fukushima, A.; Kubota, H.; Suzuki, Y.; Ando, K. Giant tunneling magnetoresistance up to 410% at room temperature in fully epitaxial Co/MgO/Co magnetic tunnel junctions with bcc Co (001) electrodes. Applied Physics Letters 2006, 89 (4), 042505042505.3. (2) Kaiser, C. High Quality Nb, Al-AlO_1tnx, Nb Josephson Junctions: Technological Development and Macroscopic Quantum Experiments, KIT Scientific Publishing: 2011; Vol. 4. (3) Abelson, L. A.; Kerber, G. L. Superconductor integrated circuit fabrication technology. Proceedings of the IEEE 2004, 92 (10), 1517-1533. (4) Long, S.; Lian, X.; Cagli, C.; Cartoixà, X.; Rurali, R.; Miranda, E.; Jiménez, D.; Perniola, L.; Liu, M.; Suñé, J. Quantum-size effects in hafnium-oxide resistive switching. Applied Physics Letters 2013, 102 (18), 183505, DOI: doi:http://dx.doi.org/10.1063/1.4802265. (5) Dorneles, L.; Schaefer, D.; Carara, M.; Schelp, L. The use of Simmons’ equation to quantify the insulating barrier parameters in Al/AlO x/Al tunnel junctions. Applied physics letters 2003, 82 (17), 2832-2834. (6) Yuasa, S.; Nagahama, T.; Fukushima, A.; Suzuki, Y.; Ando, K. Giant room-temperature magnetoresistance in single-crystal Fe/MgO/Fe magnetic tunnel junctions. Nature materials 2004, 3 (12), 868-871. (7) Robertson, J.; Wallace, R. M. High-K materials and metal gates for CMOS applications. Materials Science and Engineering: R: Reports 2015, 88 (Supplement C), 1-41, DOI: https://doi.org/10.1016/j.mser.2014.11.001. (8) Lee, K.; Lee, Y.; Chang, P.; Huang, M.; Chang, Y.; Hong, M.; Kwo, J. Achieving 1 nm capacitive effective thickness in atomic layer deposited Hf O 2 on In 0.53 Ga 0.47 As. Applied Physics Letters 2008, 92 (25), 252908. (9) Kim, S.; Woo, S.; Kim, H.; Kim, I.; Lee, K.; Jeong, W.; Park, T.; Jeon, H. Atomic Layer Deposition of HfO2 Thin Films on Ultrathin SiO2 Formed by Remote Plasma Oxidation. Journal of Korean Physical Society 2008, 52, 1103. (10) Wilk, G. D.; Wallace, R. M.; Anthony, J. High-κ gate dielectrics: Current status and materials properties considerations. Journal of applied physics 2001, 89 (10), 5243-5275. (11) Groner, M.; Elam, J.; Fabreguette, F.; George, S. M. Electrical characterization of thin Al 2 O 3 films grown by atomic layer deposition on silicon and various metal substrates. Thin Solid Films 2002, 413 (1), 186-197. (12) Fan, J.-F.; Sugioka, K.; Toyoda, K. Low-temperature growth of thin films of Al2O3 by sequential surface chemical reaction of trimethylaluminum and H2O2. Japanese journal of applied physics 1991, 30 (6B), L1139. (13) Kukli, K.; Ritala, M.; Leskelä, M.; Jokinen, J. Atomic layer epitaxy growth of aluminum oxide thin films from a novel Al (CH 3) 2 Cl precursor and H 2 O. Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 1997, 15 (4), 2214-2218.



(14) Ludeke, R.; Cuberes, M.; Cartier, E. Local transport and trapping issues in Al 2 O 3 gate oxide structures. Applied Physics Letters 2000, 76 (20), 2886-2888. (15) Casey, H. C. Devices for Integrated Circuits: Silicon and III-V Compound Semiconductors, Wiley: 1999. (16) Birey, H. Thickness dependence of the dielectric constant and resistance of Al2O3 films. Journal of Applied Physics 1977, 48 (12), 5209-5212. (17) Kim, H.; McIntyre, P. C. Atomic layer deposition of ultrathin metal-oxide films for nanoscale device applications. Journal of the Korean Physical Society 2006, 48 (1), 5-17. (18) Martin, D.; Grube, M.; Weinreich, W.; Müller, J.; Weber, W. M.; Schröder, U.; Riechert, H.; Mikolajick, T. Mesoscopic analysis of leakage current suppression in ZrO2/Al2O3/ZrO2 nano-laminates. Journal of Applied Physics 2013, 113 (19), 194103. (19) Natori, K.; Otani, D.; Sano, N. Thickness dependence of the effective dielectric constant in a thin film capacitor. Applied physics letters 1998, 73 (5), 632-634. (20) Elliot, A. J.; Malek, G.; Wille, L.; Lu, R.; Han, S.; Wu, J. Z.; Talvacchio, J.; Lewis, R. M. Probing the Nucleation of in Atomic Layer Deposition on Aluminum for Ultrathin Tunneling Barriers in Josephson Junctions. Applied Superconductivity, IEEE Transactions on 2013, 23 (3), 1101405-1101405. (21) Dietrich, C.; Boyen, H.-G.; Koslowski, B. Characterization of ultrathin insulating Al2O3 films grown on Nb (110)/sapphire (0001) by tunneling spectroscopy and microscopy. Journal of applied physics 2003, 94, 1478-1484. (22) Lu, R.; Elliot, A. J.; Wille, L.; Mao, B.; Han, S.; Wu, J. Z.; Talvacchio, J.; Schulze, H. M.; Lewis, R. M.; Ewing, D. J. Fabrication of Josephson Junctions Using In Situ Magnetron Sputtering and Atomic Layer Deposition. Applied Superconductivity, IEEE Transactions on 2013, 23 (3), 1100705-1100705. (23) Houssa, M. High k Gate Dielectrics, CRC Press: 2003. (24) Chen, S.; Lai, C.; Chin, A.; Hsieh, J.; Liu, J. High-density MIM capacitors using Al 2 O 3 and AlTiO x dielectrics. IEEE Electron Device Letters 2002, 23 (4), 185-187. (25) Elliot, A. J.; Malek, G. A.; Lu, R.; Han, S.; Yu, H.; Zhao, S.; Wu, J. Z. Integrating atomic layer deposition and ultra-high vacuum physical vapor deposition for in situ fabrication of tunnel junctions. Rev Sci Instrum 2014, 85 (7), 073904, DOI: doi:http://dx.doi.org/10.1063/1.4890286. (26) Zhu, C.; Cho, B. J.; Li, M. F. Atomic Layer Deposited High‐κ Films and Their Role in Metal‐Insulator‐Metal Capacitors for Si RF/Analog Integrated Circuit Applications. Chemical Vapor Deposition 2006, 12 (2‐3), 165-171. (27) George, S. M. Atomic Layer Deposition: An Overview. Chemical Reviews 2009, 110 (1), 111-131, DOI: 10.1021/cr900056b. (28) Banerjee, P.; Perez, I.; Henn-Lecordier, L.; Lee, S. B.; Rubloff, G. W. Nanotubular metal– insulator–metal capacitor arrays for energy storage. Nature Nanotechnology 2009, 4 (5), 292-296. (29) Malek, G. A.; Brown, E.; Klankowski, S. A.; Liu, J.; Elliot, A. J.; Lu, R.; Li, J.; Wu, J. Atomic layer deposition of Al-doped ZnO/Al2O3 double layers on vertically aligned carbon nanofiber arrays. ACS applied materials & interfaces 2014, 6 (9), 6865-6871. (30) Elliott, S. D. Atomic-scale simulation of ALD chemistry. Semiconductor Science and Technology 2012, 27 (7), 074008. (31) Kukli, K.; Ritala, M.; Pilvi, T.; Aaltonen, T.; Aarik, J.; Lautala, M.; Leskelä, M. Atomic layer deposition rate, phase composition and performance of HfO2 films on noble metal and alkoxylated silicon substrates. Materials Science and Engineering B 2005, 118, 112-116.


Page 22 of 23

Page 23 of 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


(32) Yang, W. S.; Kim, Y. K.; Yang, S.-Y.; Choi, J. H.; Park, H. S.; Lee, S. I.; Yoo, J.-B. Effect of SiO 2 intermediate layer on Al 2 O 3/SiO 2/n+-poly Si interface deposited using atomic layer deposition (ALD) for deep submicron device applications. Surface and coatings technology 2000, 131 (1), 79-83. (33) Higashi, G.; Fleming, C. Sequential surface chemical reaction limited growth of high quality Al2O3 dielectrics. Applied Physics Letters 1989, 55 (19), 1963-1965. (34) Wilt, J.; Sakidja, R.; Goul, R.; Wu, J. Z. Effect of an Interfacial Layer on Electron Tunneling through Atomically Thin Al2O3 Tunnel Barriers. ACS Applied Materials & Interfaces 2017, 9 (42), 37468-37475, DOI: 10.1021/acsami.7b12170. (35) Wilt, J.; Gong, Y.; Gong, M.; Su, F.; Xu, H.; Sakidja, R.; Elliot, A.; Lu, R.; Zhao, S.; Han, S. Atomically Thin Al 2 O 3 Films for Tunnel Junctions. Physical Review Applied 2017, 7 (6), 064022. (36) Barbos, C.; Blanc-Pelissier, D.; Fave, A.; Botella, C.; Regreny, P.; Grenet, G.; Blanquet, E.; Crisci, A.; Lemiti, M. Al 2 O 3 thin films deposited by thermal atomic layer deposition: Characterization for photovoltaic applications. Thin Solid Films 2016, 617, 108-113. (37) Lin, H.; Ye, P.; Wilk, G. Leakage current and breakdown electric-field studies on ultrathin atomic-layer-deposited Al 2 O 3 on GaAs. Applied physics letters 2005, 87 (18), 182904. (38) Weir, B.; Silverman, P.; Monroe, D.; Krisch, K.; Alam, M.; Alers, G.; Sorsch, T.; Timp, G.; Baumann, F.; Liu, C. In Ultra-thin gate dielectrics: They break down, but do they fail?, Electron Devices Meeting, 1997. IEDM'97. Technical Digest., International, IEEE: 1997; pp 73-76. (39) Miranda, E.; Suñé, J.; Rodríguez, R.; Nafria, M.; Aymerich, X.; Fonseca, L.; Campabadal, F. Soft breakdown conduction in ultrathin (3-5 nm) gate dielectrics. IEEE Transactions on Electron Devices 2000, 47 (1), 82-89. (40) Wu, Y.; Lin, H.; Ye, P.; Wilk, G. Current transport and maximum dielectric strength of atomic-layer-deposited ultrathin Al 2 O 3 on GaAs. Applied physics letters 2007, 90 (7), 072105. (41) Bonnell, D. A. Scanning tunneling microscopy and spectroscopy of oxide surfaces. Progress in surface science 1998, 57 (3), 187-252. (42) Zandvliet, H. J.; van Houselt, A. Scanning tunneling spectroscopy. Annual Review of Analytical Chemistry 2009, 2, 37-55. (43) Ugeda, M. M.; Bradley, A. J.; Shi, S.-F.; Felipe, H.; Zhang, Y.; Qiu, D. Y.; Ruan, W.; Mo, S.-K.; Hussain, Z.; Shen, Z.-X. Giant bandgap renormalization and excitonic effects in a monolayer transition metal dichalcogenide semiconductor. Nature materials 2014, 13 (12), 1091-1095. (44) Lai, X.; Chusuei, C. C.; Luo, K.; Guo, Q.; Goodman, D. Imaging ultrathin Al 2 O 3 films with scanning tunneling microscopy. Chemical Physics Letters 2000, 330 (3), 226-230. (45) Magtoto, N.; Niu, C.; Ekstrom, B.; Addepalli, S.; Kelber, J. Dielectric breakdown of ultrathin aluminum oxide films induced by scanning tunneling microscopy. Applied Physics Letters 2000, 77 (14), 2228-2230. (46) Feenstra, R. M.; Gaan, S.; Meyer, G.; Rieder, K. Low-temperature tunneling spectroscopy of Ge (111) c (2× 8) surfaces. Physical Review B 2005, 71 (12), 125316.


Al Trilayers

Recommend Documents