Low-k Research Articles

Equivalent circuit fitting method for microwave characterisation of low-k dielectric thin films

Abstract A new equivalent circuit fitting analysis scheme is proposed to analyse the measured data of test structures originally developed to characterise high-κ dielectrics at frequencies up to 5 GHz (Zhengxiang et al 1998 IEEE Trans. Electron Devices 45 1811–6). It is compared to an extension of the analysis which can be used for high-κ dielectrics at up to 50 GHz (Rundqvist et al 2004 Integr. Ferroelectr. 60 1–19). The proposed scheme focuses on the accurate characterisation of low-κ dielectrics, which exhibit greater sensitivity to various parasitics. The scheme utilises the same concentric capacitor devices and measurement setup as the original method, maintaining the advantage of ease in fabrication of the original approach. The physical model employed in the analysis step of the original method, which tended to overestimate the dielectric permittivity, has been enhanced by incorporating fringing fields and a parasitic gap capacitance as circuit elements. The accuracy of the new approach is validated using experimental data, demonstrating its ability to more accurately determine the dielectric permittivity compared to the original method.

An ultra high-endurance memristor using back-end-of-line amorphous SiC

Integrating resistive memory or neuromorphic memristors into mainstream silicon technology can be substantially facilitated if the memories are built in the back-end-of-line (BEOL) and stacked directly above the logic circuitries. Here we report a promising memristor employing a plasma-enhanced chemical vapour deposition (PECVD) bilayer of amorphous SiC/Si as device layer and Cu as an active electrode. Its endurance exceeds one billion cycles with an ON/OFF ratio of ca. two orders of magnitude. Resistance drift is observed in the first 200 million cycles, after which the devices settle with a coefficient of variation of ca. 10% for both the low and high resistance states. Ohmic conduction in the low resistance state is attributed to the formation of Cu conductive filaments inside the bilayer structure, where the nanoscale grain boundaries in the Si layer provide the pre-defined pathway for Cu ion migration. Rupture of the conductive filament leads to current conduction dominated by reverse bias Schottky emission. Multistate switching is achieved by precisely controlling the pulse conditions for potential neuromorphic computing applications. The PECVD deposition method employed here has been frequently used to deposit typical BEOL SiOC low-k interlayer dielectrics. This makes it a unique memristor system with great potential for integration.

A novel double-gate trench SOI LDMOS with double-dielectric material by TCAD simulation study

In this paper, a novel double-gate trench silicon-on-insulator lateral double-diffused metal oxide semiconductor field-effect transistor (LDMOS) with double-dielectric material (DGDK-LDMOS) is proposed. DGDK-LDMOS has two dielectric materials: a reverse-L-shaped high-k (HK) thin film and an low-k (LK) buried oxide layer. The HK thin film optimizes the electric field distribution on the drift region surface, attracting electric flux, and the excellent withstand voltage of the LK buried oxide layer can significantly improve the breakdown voltage (BV) and reduce specific on-resistance (R on,sp) of the device. The modulation mechanism of LDMOS by HK thin film and LK buried oxide layer is analyzed. The results show that compared with conventional LDMOS, when the permittivity of HK thin film is 25 and the permittivity of LK buried oxide is 3, the BV of DGDK-LDMOS is increased by 89.6%, the R on,sp is decreased by 26.4%, and the figure of merit (FOM, FOM = BV2/R on,sp) is increased by 397.2% from 3.6 MW cm−2 to 17.9 MW cm−2. Meanwhile, the output characteristics, transfer characteristics, lattice temperature, AC characteristics and switching characteristics of DGDK-LDMOS are also discussed and compared.

Optimization of the Thermally Conductive Low-k Polymer Dielectrics Based on Multisource Free-Volume Effects.

Polymers/polymer matrix composites possessing low dielectric constants (low-k polymer dielectrics) contribute to the advance of electronics, for instance, microprocessor chips, mobile phone antennas, and data communication terminals. However, the intrinsic long-chain structural characteristic results in poor thermal conductivities, which draw heat accumulation and undermine the outstanding low-k performance of polymers. Herein, multisource free-volume effects that combine two novel kinds of extra free volume with the known in-cage free volume of polyhedral oligomeric silsesquioxanes (POSSs) are discussed to reduce the capacity for dielectric constant reduction. The multisource free-volume effects of POSSs are associated with the thermal conductive network formed by the hexagonal boron nitride (BN) in the polymer matrix. The results show a decent balance between low-k performance (dielectric constant is 2.08 at 1 MHz and 1.98 at 10 GHz) and thermal conductivity (0.555 W m-1 K-1, 4.91 times the matrix). The results provide a new idea to maximize the free-volume effects of POSSs to optimize dielectric properties together with other desired performances for the dielectrics.

Screening Aluminum-Based Compounds as Low-κ Dielectrics for High-Frequency Applications

Screening Aluminum-Based Compounds as Low-κ Dielectrics for High-Frequency Applications

Investigating the Effect of Plasma Parameters on the Dielectric Constant of Sicoh Thin Films Grown by PECVD Using Dimethyldimethoxysilane

The demand for high-performance, low-power, and high-density semiconductor devices has led to the reduction in size of semiconductor chips, resulting in resistance-capacitance (RC) delay. To mitigate the RC delay, low dielectric constant (low-k) insulators have been developed, with the focus on developing new materials with lower dielectric constant. In this report, we present our investigation into the effect of electron density and temperature on the reduction of dielectric constant in SiCOH thin films, utilizing dimethyldimethoxysilane (DMDMS) as a variable with respect to pressure. Our results reveal that both electron density and temperature decline with increasing pressure due to localized electron kinetics and increased inelastic collisions between electrons and neutral species in the chamber, respectively. Our findings also show that the dielectric constant and refractive index of SiCOH thin films decrease at low electron density and temperature. Quadrupole mass spectroscopy reveals that CH3 radicals dissociate less at higher pressures, while Fourier transform infrared spectroscopy confirms an increase in Si-CH3 bonds in SiCOH thin films with increasing pressure. Our findings provide insights into the relationship between electron density and temperature and the dielectric constant of SiCOH thin films, which can contribute to the development of low-k insulators for high-performance semiconductor devices for applications such as big data, mobility, and the Internet of Things (IoT).

The characterization of low-k thin films and their fracture analysis in a WLCSP device

The IC industry implements low dielectric constant (low-k) materials in copper/low-k interconnects to reduce the RC delay and crosstalk noises. As low-k dielectrics are weaker than traditional SiO2, measuring their mechanical properties is important for numerical simulations to evaluate the structural integrity of a package with low-k materials. In this study, we fabricated freestanding samples of low-k thin film with/without pre-cracks on both sides using lithography and exfoliation for uniaxial tension and coefficient of thermal expansion tests. The tension tests revealed that the average fracture toughness and Young's modulus of the low-k film are 1.41 MPa·m1/2 and ∼2.5 GPa, respectively. Using the measured mechanical properties, we performed a fracture analysis by a 2D finite element model for the low-k material in a wafer-level chip-size package under thermal loading. The computational results showed that the critical microcrack in the low-k layer occurs near the solder balls, where the KI increases rapidly with crack length and exceeds KIC. The results are consistent with the experimental findings in literature. The present methods can help evaluate the integrity of low-k material in a package and improve reliability.

Investigation into the effect of a PECVD-deposited SiOx chamber coating on the selective, radical-based NF3 etching of TaN with respect to BEOL low-k

This paper explores the use of a low temperature inductively coupled plasma discharge containing an NF3/Ar mixture for the isotropic, dry etching of TaN with selectivity to low-k dielectric; relying on radically based etch, without nonselective ion bombardment. With a clean chamber condition, no etch selectivity is found between TaN and the low-k dielectric, but when the chamber condition is manipulated by the addition of a plasma-deposited SiOx coating to the chamber wall, selectivity between TaN and the low-k dielectric is achieved. Deposition occurred on the low-k dielectric when the coating was applied to the chamber walls before etch, and while the TaN film etch rate did decrease with a coating applied, an etching regime was still observed. The coating was found to add significant atomic oxygen to the etch processes and decreased etching of the low-k dielectric. The deposition regime apparent for the low-k dielectric was made possible by the inability of the fluorine radicals to volatize silicon oxyfluoride compounds, causing deposition of a silicon oxyfluoride film on the low-k surface. The same etching inhibition was not observed on TaN, allowing selective etching.

1.2 kV 4H–SiC Super Junction UMOSFET with a Low-K dielectric pillar

A 1.2 kV 4H–SiC Super Junction (SJ) UMOSFET with a Low-K (LK) dielectric pillar below the trench gate (LK-SJ-UMOSFET) is proposed in this paper. The device features a P-type doping pillar (P-pillar) at the source-side and a deep LK trench beneath the gate. Firstly, the gate-to-drain capacitance (Cgd) is greatly decreased by the LK dielectric pillar. Secondly, both the LK dielectric and the P-pillar cause the assistant depletion of the N-type doping pillar (N-pillar), resulting in an increased N-pillar doping concentration and a decreased specific on-resistance (Ron,sp). The proposed LK-SJ-UMOSFET is demonstrated to have a low high frequency Figure-Of-Merit (HF-FOM) Ron×Cgd = 28.628 mΩ pF and Ron × Qgd = 29.274 mΩ nC. The HF-FOM of the proposed 4H–SiC LK-SJ-UMOSFET is lower than that of the other 4H–SiC UMOSFET in recent simulation studies and reports.

Role of dose optimization in Ru atomic layer deposition for low resistivity films

Ruthenium (Ru) is an alternative to copper (Cu) and cobalt (Co) interconnect layers in sub 20 nm features due to its low resistivity in scaled wires and low diffusion into porous low-K dielectrics (SiCOH). Two goals for a successful Ru atomic layer deposition (ALD) process are to enable films with resistivity values as close as possible to that of bulk Ru and to enable selective deposition to achieve bottom-up fill of vias. In this work, the effects of dose variation on resistivity and selectivity of the Ru ALD process using a dicarbonyl-bis(5-methyl-2, 4-hexanediketonato) Ru(II) precursor, Ru(IHD)2(CO)2 (“Carish”), and O2/He coreactant were investigated. Instead of varying the Carish precursor dose to optimize the growth rate per cycle, the precursor dose was optimized to reduce the film resistivity from 18.5 to 10.2 μΩ cm. By varying the O2/He coreactant dose, the substrate selectivity of the ALD process was successfully enhanced as evidenced by the increased nucleation delay on bis(N,N-dimethylamino)dimethylsilane passivated SiO2 over hydrofluoric acid-cleaned SiO2. These findings highlight the importance of dose optimization beyond the ALD saturation point in developing a selective and low resistivity Ru ALD process. Density functional theory calculations were performed to provide a mechanistic understanding of the underlying surface reactions of the Carish precursor and the roles of CH3 passivation and O2 coreactants.

Self-assembled monolayers on porous low-k dielectrics by decyltrimethoxysilane vapor treatment: A perspective from electrical characteristics and time-dependence-dielectric-breakdown reliability

Self-assembled monolayer (SAM) is an attractive technology for back-end-of–line interconnects of integrated circuits for advanced technological nodes. In this study, SAMs were formed on the porous low-k films by decyltrimethoxysilane (DTMOS) in vapor phase at different temperatures. The effects of the formation SAMs on the electrical characteristics and reliability of the porous low-k films were characterized. With DTMOS vapor treatment at 50°C∼100°C, SAMs were formed onto the porous low-k films, thus enhancing the dielectric breakdown field, failure time, and Cu barrier capacity. Additionally, DTMOS vapor treatment at higher temperatures resulted in larger enhancement and more thermal stability. The formation SAMs was demonstrated to effectively repair the damaged porous low-k film, thus reducing the dielectric constant; although, the bulk dielectric constant was higher than that of the pristine film. To sum up, this study provided a promising SAMs processing derived from DTMOS vapor phase, for ensuring better integrity for the porous low-k films in the advanced technological nodes.

Nanoindentation based method to determine the thermal expansion coefficients of low-k dielectrics

This paper demonstrates the use of nanoindentation at elevated temperatures to measure the coefficient of thermal expansion (CTE) of low-k dielectric films. This methodology was calibrated using curvature measurements to elucidate the impact of tip geometry and elastic modulus on the theoretical model to calculate the CTE. Further validation was achieved by applying this methodology on a variety of low-k dielectric films with different elastic modulus, but similar matrix density. The advantage of this methodology over other techniques is the capability to simultaneously measure the elastic modulus, the easy sample preparation, short measurement time and easy analysis.

Properties of ultrathin molybdenum films for interconnect applications

The structural and electrical properties of Mo thin films with thicknesses between 3 and 50 nm, deposited by physical vapor deposition, have been evaluated in order to assess the potential of Mo as an alternative to Cu or W for nano-interconnect applications. Mo films deposited on SiO2/Si (100) were polycrystalline with randomly oriented grains close to the interface and the progressive formation of a (110) texture above 5 nm film thickness. Adhesion between Mo and low-κ dielectrics was strong with adhesion energies above 5 J/m2. The films showed intrinsic tensile stress of ∼1 GPa, which decreased with increasing thickness. The Mo resistivity showed a weaker thickness dependence than Cu, which rendered Mo competitive with conventional TaN/Cu/TaN metallization below metal thicknesses of 8 nm. Semiclassical resistivity modeling found that the thin film resistivity was limited by grain boundary scattering with a reflection coefficient of R = 0.46 ± 0.03. Furthermore, the effects of Mo deposition on different dielectric and metallic substrates and of post-deposition annealing up to 950ºC were investigated. Crystallinity, impurity concentration, and resistivity were all affected by the substrate. Post-deposition annealing in H2 led to continuous grain growth, followed by recrystallization at 860ºC. Furthermore, annealing at 650°C led to compressive stress in the films. By contrast, annealing in H2/N2 led to the incorporation of N and ultimately to Mo nitride formation above 500ºC. The results indicate that Mo can be a promising candidate for advanced interconnect metallization schemes if surface oxidation as well as impurity incorporation can be kept under control.

Application of the Tikhonov Regularization Method in Problems of Ellipsometic Porometry of Low-K Dielectrics

In the development of promising ULIS scaling technologies, one of the key roles is played by porous dielectrics with a low permittivity used to isolate interconnects in a metallization system. Condensation of gaseous products in the pores of such films makes it possible to solve the most important problem that prevents the integration of such films, to carry out low-damage plasma etching. However, methods for studying porosity are also based on the study of the adsorption isotherm during condensation in film pores. Therefore, the study of adsorption in pores is one of the most important practical problems arising in the creation of dielectrics with a low permittivity and the study of low-damaging methods for their structuring. The method of ellipsometric porosimetry is an easy-to-implement and accurate approach for obtaining an adsorption isotherm; however, its further analysis and determination of the pore size distribution are reduced to solving an integral equation and is an ill-posed problem. In this paper, we propose to apply Tikhonov’s regularization method to solve it. The method is verified on model data and used to study a low-k dielectric sample with an initial thickness of 202 nm and a permittivity of 2.3 based on organosilicate glass.

The nanomechanical properties of non-crosslinked calcium aluminosilicate hydrate: The influences of tetrahedral Al and curing age

Calcium aluminosilicate hydrate (C-A-S-H) is the binding phase of both blended cement-based and alkali-activated materials. The intrinsic mechanical properties of non-cross-linked C-A-S-H are important while experimentally unvalidated. Here, the properties are for the first time measured using high-pressure X-ray diffraction. The incompressibility and bulk modulus K0 of C-A-S-Hs are correlated to their nanostructure and stability using nuclear magnetic resonance and X-ray absorption spectroscopies. Al coordination in stable C-A-S-H (Al/Si = 0.1) cured for 546 days is purely tetrahedral (AlIV), while in metastable C-A-S-H (Al/Si = 0.05) cured for only 182 days is both AlIV and pentahedral (AlV). The stable C-A-S-H is stiffer along the a,b,c-axis with higher K0 relative to C-S-H. Short-curing-induced metastable C-A-S-H (Al/Si = 0.05) shows expanded interlayer and softer c-axis, thus lower K0 than C-S-H and the stable C-A-S-H. Our results highlight the stiffening effect of AlIV incorporation and the negative influences of insufficient curing on the nanomechanical properties of non-cross-linked C-A-S-H at Ca/Si = 1.

(Dielectric Science & Technology Thomas Callinan Award) Materials and Processes As Enablers for Moore Moore and Beyond Moore Technologies

In this presentation, an overview will be presented of historic and recent developments achieved w.r.t. materials and processes that enabled continued performance scaling of CMOS and beyond CMOS applications. Starting from early high-k work. For decades, thermal oxidation of crystalline Silicon into SiO2 has been the gate oxide material in CMOS technology. Miniaturization for performance improvement, required the gate dielectric to decrease in thickness to support the required increase in capacitance (per unit area) and drive current (per device width). Early 2000, the thickness scaled below 1.5 nm, causing drastically increased leakage currents, high power consumption and reduced device reliability. Replacing the SiO2 gate dielectric with a high-κ (dielectric constant) material allows increased gate capacitance without the associated leakage effects. Activities involved the unit process step development of dielectric and metal deposition processes, advanced interface preparation, electrical and physical characterization and wet and dry etch process development. In a second section, emphasis will be on emerging nanomaterials. Nanotechnology is defined as the manipulation of matter with at least one dimension sized from below 100nm, thus all CMOS activities are functional nanotechnology. The latter concentrated initially on Graphene, but extended further towards Transition Metal Dicalchogenide Materials. Emphasis of the work will be on growth, functionalization, dielectric passivation and doping of these materials. Lastly, expanding on the nanomaterial know-how also progress with regard to low-k dielectrics, area selective deposition processes (for bottom-up lithography) and application of nanochemistry research for photoresist material screening, where chemical interactions at the nanoscale, related to the EUV lithography (radiation chemistry) are hampering the pattern scaling.Throughout the presentation, recognition will be given to people (PhD students) that contributed to this work and the presentation is also in honor of Prof Dolf Landheer and Professor Samares Kar with whom I organized my first ECS symposium.

Facile Strategy for Preparing a Rosin-Based Low-k Material: Molecular Design of Free Volume.

Low-k dielectrics are urgently needed in modern integrated circuits. The introduction of free volume instead of porous structures has become a powerful strategy to reduce the k value. According to this strategy, the biomass resource rosin-containing hydrogenated phenanthrene ring was introduced into benzocyclobutene (BCB) resin to reduce the k value; then a rosin-based BCB monomer was successfully synthesized. Meanwhile, the BCB monomer without a rosin skeleton was prepared. After converting the monomers into thermo-crosslinked materials, notably that the rosin skeleton has a great influence on the free volume and k value of the material. The fractional free volume and k value of the former are 26% and 2.44, respectively, and those of the latter are 14% and 2.84, respectively. In addition, the distances between molecular chains and the density of the former are 0.60 nm and 1.06 g cm-3, respectively; those of the latter are 0.56 nm and 1.28 g cm-3, respectively. These data show that introducing hydrogenated phenanthrene rings occupies part of the space and hinders the packing of molecular chains, which increases the distance between molecular chains and reduces the density of the polymer, resulting in an increasing free volume and a reducing k value. Notably that introducing hydrogenated phenanthrene rings cannot affect other properties of the material. Therefore, this research indicates that introducing rosin skeletons can prepare high-performance materials, which provide some promising low-k materials for the development of electronics and microelectronics.

Comparative Analysis of Transcriptome Profiles Reveals the Mechanisms in the Difference of Low Potassium Tolerance among Cultivated and Tibetan Wild Barleys

Potassium (K) deficiency is a bottleneck for crop production. Thus, developing low K (LK)-tolerant crop cultivars to relieve the issue is extremely urgent. Our previous studies had found that Tibetan annual wild barley accessions showed a higher LK tolerance than the cultivated barley. In this study, RNA-sequencing was performed on three barley genotypes, wild (XZ153, LK tolerance; XZ141, LK sensitivity) and cultivated (ZD9, LK sensitivity) barley genotypes, to compare the transcriptome profiles of their shoots at two time points after LK stress. In total, 4832 genes displayed differential expression at 48 h and 15 d among three genotypes after K stress treatment, with XZ153 having much more differentially expressed genes (DEGs) at 48 h than 15 d, but it was the opposite in ZD9. Meanwhile, GO annotation analysis and KEGG pathway enrichment were implemented on 555 and 814 LK tolerance-associated DEGs at 48 h and 15 d after LK stress, respectively. Three barley genotypes differed significantly in transcriptional level after LK treatment. The high tolerance in wild genotype XZ153 could be attributed to many factors, mainly including K channels, Ca2+ signaling pathway, ethylene biosynthesis process, TCA cycle, glycolysis, pentose phosphate pathway, and photosynthesis. Furthermore, some candidate genes identified in this study may be used to improve the LK tolerance of barley.

Charge Transport Mechanism in a PECVD Deposited Low-k SiOCH Dielectric

Charge Transport Mechanism in a PECVD Deposited Low-k SiOCH Dielectric

Functionalization of low-k surfaces with low-energy Ar ions

In this work low-energy Ar impacts on low- k surfaces were simulated with ab initio density functional theory method. The obtained results reveal the mechanism of CH3-group removal under Ar atom/ion irradiation; the threshold energy of this process was estimated. Keywords: low-k dielectrics, functionalization, simulation, diffusion barriers, methyl groups.