Using ab initio simulation, we study a ferroelectric layer of a few nanometers made of hafnia (HfO2) under the influence of Si doping with TiN electrodes. We evaluate the orthorhombic phase of Pca21 symmetry, its ferroelectric switching and the incidence of doping with silicon. We show that the ferroelectric switching can involve a 90° characteristic angle with corresponding activation energy which is lowered by a factor three due to Si doping at 3% at. A full MFM (Metal-Ferroelectric-Metal) model is derived in order to simulate finite-size effects. This model is compatible with a reversal of a polar HfO2:Si with a (111) preferential orientation. Validity and usefulness of such a model are discussed for ferroelectric devices optimization.

The recent discovery of ferroelectric behavior in doped hafnia-based dielectrics, attributed to a non-centrosymmetric orthorhombic phase, has potential for use in attractive applications such as negative differential capacitance field-effect-transistors (NCFET) and ferroelectric random access memory devices (FeRAM). Alloying with similar oxides like ZrO2, doping with specific elements such as Si, novel processing methods, encapsulation and annealing schemes are also some of the techniques that are being explored to target structural modifications and stabilization of the non-centrosymmetric phase. In this study, we utilized synchrotron-based x-ray diffraction in the grazing incidence in plane geometry (GIIXRD) to determine the crystalline phases in hafnia-zirconia (HZO) compositional alloys deposited by atomic layer deposition (ALD). Here we compare and contrast the structural phases and ferroelectric properties of mechanically confined HZO films in metal-insulator-metal (MIM) and metal-insulator-semiconductor (MIS) structures. Both MIM and MIS structures reveals a host of reflections due to non-monoclinic phases in the d-spacing region between 1.75Å to 4Å. The non-monoclinic phases are believed to consist of tetragonal and orthorhombic phases. Compared to the MIS structures a suppression of the monoclinic phase in MIM structures with 50% zirconia or less was observed. The correlation of the electrical properties with the structural analysis obtained by GIIXRD highlights the importance of understanding the effects of the underlying substrate (metal vs. Si) for different target applications.

Resistive Random Access Memory (RRAM) is a novel form of non-volatile memory that is expected to play a major role in future computing and memory solutions. It has been shown that the resistance of RRAM devices can be precisely tuned by modulating switching voltages, by limiting peak current, and by adjusting the switching pulse duration. This enables the realization of novel applications such as memristive neuromorphic computing and neural network computing. The RRAM devices described in this work utilize an inert tungsten bottom electrode, hafnium oxide based active switching layer, a titanium oxygen exchange layer, and an inert titanium nitride top electrode. Linear sweep and controlled pulse (down to 10 ns) based electrical characterization of RRAM devices was performed in a 1 transistor 1 RRAM (1T1R) configuration to determine endurance, reliability, retention and threshold voltage parameters. We demonstrated endurance values above 108 cycles with an average on/off ratio of 15 and pulse voltages for set/reset operation of ±1.5V. The on-chip 1T1R structures show an excellent controllability with respect to the low and high resistive state by manipulating the peak current from 75 up to 350µA we were able to achieve 10 discrete resistive states. Our results demonstrate that the set operation (which shifts the RRAM device from the high to the low resistance state) is only dependent on the voltage of the switching pulse and the peak current limit. The reset operation, however, occurs in an analog fashion and appears to be dependent on the total energy of the applied switching pulse. Pulse energy was modulated by varying the peak voltage which resulted in a larger relative change of the RRAM device resistance.

Transition metal nitrides containing metal ions in high oxidation states are a significant goal for the discovery of new families of semiconducting materials. Most metal nitride compounds prepared at high temperature and high pressure from the elements have metallic bonding. However amorphous or nanocrystalline compounds can be prepared via metal-organic chemistry routes giving rise to precursors with a high nitrogen:metal ratio. Using X-ray diffraction in parallel with high pressure laser heating in the diamond anvil cell this work highlights the possibility of retaining the composition and structure of a metastable nanocrystalline precursor under high pressure-temperature conditions. Specifically, a nanocrystalline Hf3N4 with a tetragonal defect-fluorite structure can be crystallized under high-P,T conditions. Increasing the pressure and temperature of crystallization leads to the formation of a fully recoverable orthorhombic (defect cottunite-structured) polymorph. This approach identifies a novel class of pathways to the synthesis of new crystalline nitrogen-rich transition metal nitrides.

Grazing Incidence Small-Angle X-ray Scattering (GISAXS) is a versatile technique for the analysis of nano and micro thin films surfaces. The scattering data depend strongly on the form and distribution of the scattering objects. In the present work GISAXS is used to study hafnium dioxide (HfO2) thin films deposited by magnetron sputtering using different deposition processes and post-deposition annealing conditions. Two distinct types of 15 nm thick samples were produced using different sputtering targets and different gas mixtures. The GISAXS results show that the ellipsoids that compose the thin films present a reduction in their size for both samples sets. For the sputtered Hf metal target samples, the ellipsoid diameter value shifted from 9 nm (as-deposited) to 6 nm following a 800 °C thermal treatment. For the sputtered HfO2 target samples the diameter value shifts from 19 nm (as-deposited) to 3 nm after a 800 °C anneal in oxygen. The size distribution, for both sets of samples, follows a Gaussian distribution function.

The structures and energies of stoichiometric and oxygen-deficient monoclinic HfO2 were calculated using density functional theory. The electronic interactions in HfO2 were calculated using the LDA+U and GGA+U formalisms, where on-site Coulomb corrections were applied to the 5d electrons of hafnium (Ud ) and the 2p electrons of oxygen (Up ). Properties calculated using these techniques are compared to results obtained from LDA, GGA, hybrid functionals, and experiment. Ultimately, we show that LDA+Ud +Up and GGA+Ud +Up calculations of HfO2’s electronic and structural properties achieve a level of accuracy on par with much more computationally demanding hybrid functional techniques, such as PBE0 and HSE06.

Physical properties of filaments in Cu/HfO2/Pt conducting-bridge memory (CB-RAM) were investigated basing on direct observation by conducting atomic force microscopy (C-AFM) and energy dispersive X-ray spectroscopy (EDS), R-T characteristics until liquid nitrogen temperature, and I-V characteristics both in air and in vacuum. As a result, physical picture of filaments in Cu/HfO2/Pt structures was revealed. Filaments consist of Cu containing large residual resistance and the cross-sectional area of the filament, S fila, was roughly proportional to set voltage, V set, even when current compliance was kept constant. Interestingly, resistivities of filaments are same among all the filaments in different samples and are invariant even after repetitive switching that changes resistance of the filaments. Cu/HfO2/Pt obeyed the universal relation that reset current, I reset, is proportional to the inverse of resistance in a low resistance state, 1/R LRS, which is known to be applicable to oxygen-migration-based resistive switching memories such as Pt/NiO/Pt. Considering the invariance of resistivity of the filament, this suggests the fact that I reset is decided dominantly by S fila. In addition, it was suggested that moisture is necessary for dissolution and migration of Cu to form filaments.

Field observations from the Trinity Peninsula Group at View Point on the Antarctic Peninsula indicate that thick, southward-younging and overturned clastic sedimentary rocks, comprising unusually coarse conglomeratic lenses within a succession of fine-grained sandstone–mudstone couplets, are the deposits of debris and turbidity flows on or at the foot of a submarine slope. Three detrital zircons from the sandstone–mudstone couplets date deposition at 302 ± 3 Ma, at or shortly after the Carboniferous–Permian boundary. Conglomerates predominantly consist of quartzite and granite and contain boulders exceeding 500 mm in diameter. Zircons from granitoid clasts and a silicic volcanic clast yield U–Pb ages of 466 ± 3 Ma, 373 ± 5 Ma and 487 ± 4 Ma, respectively and have corresponding average εHft values between +0.3 and +7.6. A quartzite clast, conglomerate matrix and sandstone interbedded with the conglomerate units have broadly similar detrital zircon age distributions and Hf isotope compositions. The clast and detrital zircon ages match well with sources within Patagonia; however, the age of one granite clast and the εHf characteristics of some detrital zircons point to a lesser South Africa or Ellsworth Mountain-like contribution, and the quartzite and granite-dominated composition of the conglomerates is similar to upper Palaeozoic diamictites in the Ellsworth Mountains. Unlike detrital zircons, large conglomerate clasts limit possible transport distance, and suggest sedimentation took place on or near the edge of continental crust. Comparison with other upper Palaeozoic to Mesozoic sediments in the Antarctic Peninsula and Patagonia, including detrital zircon composition and the style of deformation, suggests deposition of the Trinity Peninsula Group in an upper plate basin on an active margin, rather than a subduction-related accretionary setting, with slow extension and rifting punctuated by short periods of compression.

A significant challenge for high throughput nucleic acid analysis and sequencing is to increase both throughput and sensitivity. Electrical detection methods are advantageous since they can be easily scaled to high density arrays, are highly sensitive, and do not require bulky optical equipment for readout. A focus of most nucleic acid based sensors is the detection of sequence-specific hybridization events between complementary strands of DNA or RNA. These hybridization events can be detected electrically, due to the intrinsic negative charge associated with the phosphate-rich nucleic acid backbone. Field effect transistors (FETs) and high electron mobility transistors (HEMTs) are ideal devices for detecting such hybridization events, due to their high sensitivity to changes in electrical field strength. A key concern for the construction of DNA-based FET and HEMT biosensors is the immobilization of probe oligonucleotides on the active region of the sensor. In previous work, our group has shown that single stranded DNA can be directly immobilized onto semiconductor materials without the need for complex surface chemistry or crosslinking strategies. In the present work, we have shown that the immobilization of single stranded DNA onto these materials is influenced by the terminal phosphate group of the DNA molecule, independent of backbone phosphates. This agrees with previous studies in which phosphates and phosphonates exhibited strong attachment to a variety of metal oxides. We have also shown that surface-immobilized DNA is available for hybridization and that hybridization is sequence specific. Phosphate-dependent immobilization was demonstrated for HfO2, AlGaN, and ZrO2 surfaces using optical detection of DNA-DNA hybridization, as well as x-ray photoelectron spectroscopy (XPS) analysis of DNA-modified surfaces.

For the first time, we discuss the compatibility of stress proximity technique (SPT) with dual stress liner (DSL) in high-κ/metal gate (HK/MG) technology. The short-channel mobility enhancement and the drive current improvement brought by SPT have been demonstrated at 32nm technology node. With maintained short channel control and threshold voltage roll-off characteristics, SPT has achieved 7% drive current improvement for both nFET and pFET from the optimization of SPT with DSL.

Hafnium(IV) oxide (HfO2) has replaced silicon oxide as a gate dielectric material in leading edge CMOS technology, providing significant improvement in gate performance for field effect transistors (FETs). We are currently exploring this high-k dielectric for its use in nucleic acid-based FET biosensors. Due to its intrinsic negative charge, label-free detection of DNA can be achieved in the gate region of high-sensitivity FET devices. Previous work has shown that phosphates and phosphonates coordinate specifically onto metal oxide substrates including aluminum and titanium oxides. This property can therefore be exploited for direct immobilization of biomolecules such as nucleic acids. Our work demonstrates that 5’ phosphate-terminated single stranded DNA (ssDNA) can be directly immobilized onto HfO2 surfaces, without the need for additional chemical modification or crosslinking. Non-phosphorylated ssDNA does not form stable surface interactions with HfO2, indicating that immobilization is dependent upon the 5’ terminal phosphate. Further work has shown that surface immobilized ssDNA can be hybridized to complementary target DNA and that sequence-based hybridization specificity is preserved. These results suggest that the direct DNA-HfO2 immobilization strategy can enable nucleic acid-based biosensing assays on HfO2 terminated surfaces. This work will further enable high sensitivity electrical detection of biological targets utilizing transistor-based technologies.

Fine bulk samples of delta-phase Hf hydride with various hydrogen contents (CH) ranging from 1.62 to 1.72 in the atomic ratio (H/Hf) were prepared, and their thermal and mechanical properties were characterized. In the temperature range from room temperature to around 650 K, the heat capacity and thermal diffusivity of the samples were measured and the thermal conductivity was evacuated. The elastic modulus was calculated from the measured sound velocity. The Vickers hardness was measured at room temperature. Effects of CH and/or temperature on the properties of Hf hydrides were discussed. At room temperature, the thermal conductivity values of the Hf hydrides were 23 Wm−1K−1. The Young's and shear moduli and the Vickers hardness of Hf hydride decreased with increasing CH.

HfO2 is a promising dielectric for alternative high-k gate dielectrics and DRAM capacitor applications because of its high dielectric constant, excellent thermal stability and compatibility with Si processing. Particularly atomic layer deposition (ALD) is suitable as a deposition tool for gate dielectrics as well as capacitor dielectrics due to its excellent conformity and precise thickness control. In order to achieve ideal ALD performance, it is critical to have appropriate precursors. In order to achieve ideal ALD performance, it is critical to use appropriate enabling precursors. So far, Tetrakisethylmethylamidohafnium (TEMAHf) has been considered as one of the most promising precursors due to its good physical properties. However, this precursor has relatively low thermal stability, which has often become a drawback to ALD process for HfO2. To circumvent the decomposition issues of TEMAHf during ALD deposition, often low process temperatures have been used that lead to less dense films and high level of carbon contamination. In order to overcome these challenges, the novel hafnium formamidinate (Hf-FAMD) precursor is developed as an alternate Hf source at Dow Electronic Materials. Following our success with formamidinate platform in previously reported novel lanthanum formamidinate (La-FAMD) source, this new precursor also exhibits high thermal stability and high reactivity towards water and ozone. In this work, we report, for the first time, comparative studies with TEMAHf and novel Hf-FAMD source, e.g. Hf-FAMD exhibits acceptable vapor pressure (> 0.1 Torr at 100 °C) similar to that of TEMAHf, and higher thermal stability than TEMAHf, thus leading to high quality ALD films. We also present the crystal structure of La-FAMD, elucidated by X-Ray Crystallography, and physical properties of novel Hf-FAMD relevant to ALD.

The scaling of contemporary metal-insulator-metal (MIM) capacitors requires oxides of higher dielectric constant (>10), such as hafnium oxide (∼18) and titanium oxide (∼40). Intensive research of these oxides and oxide stacks is needed to develop them into high quality electronic materials for their application as capacitors in high temperature environments. High-k dielectrics such as HfO2 and HfO2/TiO2/HfO2 have been grown by thermal oxidation to fabricate MIM capacitors on SiO2/Si substrates and on sapphire substrates also. The thermally grown Al/HfO2/TiO2/HfO2/Pt/Ti/SiO2/Si MIM capacitor is reported here for the first time. The MIM capacitor using HfO2/TiO2/HfO2 dielectric film shows a similar frequency dependence using HfO2 dielectric on a SiO2/Si substrate, whilst its voltage linearity coefficients, leakage current and temperature coefficient are higher than the capacitor employing HfO2 dielectric. The MIM capacitor with HfO2 dielectric fabricated on sapphire substrate shows the strongest frequency dependence, voltage linearity coefficient and temperature dependence which is related to the surface roughness of substrate. The high capacitance density of these capacitors, ranging from 5.21 fF/µm2, meets the ITRS requirements for analog capacitor up to 2012. The MIM capacitor using 30nm HfO2 dielectric film illustrates highest capacitance density, 5.21 fF/µm2, a VCC of 236 ppm/V2, a temperature coefficient of 290 ppm/ºC, measured up to 300 ºC, and leakage current density which is 1.3 × 10−7 A/cm2 at 1V.

Gd2O3 and HfO2 films exhibit photovoltaic charging that depends on temperature. The dielectric properties of the surface and concomitant photovoltaic surface charging increases following oxygen exposure under ultrahigh vacuum conditions suggesting that the surface conductivity plays a role in the surface photovoltage charging.

The growth of HfO2 thin films on a HF-dipped p-Si(100) substrate at 200 °C by atomic-layer deposition (ALD) using Hf[N(C2H5)(CH3)]4 and H2O vapor as precursors is demonstrated. Uniform HfO2 thin films are obtained on a 4-in. silicon wafer, and the energy-band gap and band offset are determined by x-ray photoelectron spectroscopy analysis. The as-deposited HfO2 thin film is amorphous and able to crystallize at 500 ∼ 600 °C with only the monoclinic phase. As for the electrical performance of Au–Ti–HfO2–Si metal oxide semiconductor capacitors, a dielectric constant of ∼17.8 and an equivalent oxide thickness value of ∼1.39 nm are obtained from the 40-cycle ALD film after annealing at 500 °C. In addition, the breakdown field is in the range of 5 ∼ 5.5 MV/cm, and the fixed charge density is on the order of 1012 cm−2, depending on the annealing temperatures. The interface quality of HfO2 thin films on silicon is satisfactory with an interface-trap charge density of ∼3.7 × 1011 cm−2 eV−1.

Sudden jumps in nanoindentation load-displacement curves of bulk metallic glasses (BMGs) signify the onset of plastic deformation. These events are investigated on varying compositions of Zr- and Hf-rich BMGs. Load-versus-displacement graphs for spherical indentations are analyzed to determine displacement, load, intensity of deformation, energy per volume, energy loss, and pressure corresponding to these key locations. Attention is focused on pressure, energy loss, and energy per volume at initial plasticity in response to varying strain rates, indenter tip radii, preload, and material composition. Energy loss was found to correlate with preload. The Zr-rich metallic glass was found to plastically deform in response to smaller loads than Hf-rich specimens.

The nature of the interface between Si and Ge substrates and high-k dielectrics often controls the performance of MOSFET devices. Precleaning and/or chemical functionalization of the surfaces can dramatically affect the formation of an interfacial layer. We have used in-situ IR spectroscopy to probe the relevant interfaces during ALD growth for a variety of surface treatments, including H- and Cl-termination, and nitridation. This paper focuses on understanding of the mechanisms for interfacial SiO2 (or GeOx) formation during HfO2 growth using tetrakis-ethylmethylamidohafnium (TEMAH) as the metal precursor and water or ozone as the oxygen precursor. We find that impurities arising from incomplete ligand elimination during growth (e.g. OH for H2O processing and CO- and NO-containing species for O3 processing) are incorporated into the HfO2 film during growth. Upon annealing, most of these species react, but can also migrate to the interface. Nitridation of Si and Ge surfaces will in general prevent SiO2 or GeOx formation but can also affect the growth rate.

In this paper, various approaches to extend scalability of Hafnium-based dielectrics are reported.Among the three crystal phases of HfO2 (monoclinic, cubic and tetragonal), the tetragonal phase has been reported to have the highest dielectric constant.Tetragonal phase stabilization by crystallizing the thin HfO2 using a metal capping layer and by adding zirconium is demonstrated.The microstructure, morphology, optical properties and impurities of HfxZr1-xO2 dielectrics (for 0<x<1) are discussed.Subtle but important modification to high-k / Si interface characteristics resulting from addition of Zr into HfO2 is reported.To further boost the dielectric constant of hafnium-based dielectrics, incorporation of TiO2, which has been reported to have high dielectric constant, is explored.HfxZr1-xO2/TiO2 bilayer films were fabricated. 30 Å TiO2 films were deposited on a 5, 8, 12 or 15 Å HfxZr1-xO2 underlayer to determine the minimum thickness needed to maintain good thermal stability with Si substrate. CV and IV results indicated that 12-15 Å is the optimal thickness range for the HfxZr1-xO2 underlayer.A dielectric constant as high as 150 for TiO2 layer is extracted from TiO2 thickness series deposited on12 Å HfxZr1-xO2 underlayer.In addition to increasing the k-value of Hafnium-based dielectrics, it is important that the threshold voltage of these high-k devices is low.Here we report the use of thin Al2O3 capping layers to modulate PMOS threshold voltages.About 100 mV reduction in threshold voltage is achieved by capping HfO2 with a 5Å Al2O3 film.Finally, dielectric scaling by modifying the Si/high-k interfacial layer is attempted.Nitrogen incorporation into HfxZr1-xO2 is shown to be a simple and effective method to lower the capacitance equivalent thickness (CET) of Hafnium-based dielectrics.