Search

One of the bottlenecks to building semiconductor chips is the increasing cost required to develop chemical plasma processes that form the transistors and memory storage cells1,2. These processes are still developed manually using highly trained engineers searching for a combination of tool parameters that produces an acceptable result on the silicon wafer3. The challenge for computer algorithms is the availability of limited experimental data owing to the high cost of acquisition, making it difficult to form a predictive model with accuracy to the atomic scale. Here we study Bayesian optimization algorithms to investigate how artificial intelligence (AI) might decrease the cost of developing complex semiconductor chip processes. In particular, we create a controlled virtual process game to systematically benchmark the performance of humans and computers for the design of a semiconductor fabrication process. We find that human engineers excel in the early stages of development, whereas the algorithms are far more cost-efficient near the tight tolerances of the target. Furthermore, we show that a strategy using both human designers with high expertise and algorithms in a human first–computer last strategy can reduce the cost-to-target by half compared with only human designers. Finally, we highlight cultural challenges in partnering humans with computers that need to be addressed when introducing artificial intelligence in developing semiconductor processes.

Atomic layer etching (ALE) is a multistep process used for removing ultrathin layers of the material. The removal step can be driven by ion bombardment, typically with energies of 1 s. Previously, we reported a new ALE operating regime where exposures to ion energies were >500 eV and step times were

A conceptual framework is introduced to gain insights into performance benefits and challenges of directional and isotropic Atomic Layer Etching (ALE).

In this paper, we report on plasma assisted thermal Atomic Layer Etching (ALE) of Al2O3. The surface was modified via a fluorine containing plasma without bias power. The removal was accomplished by a thermal reaction step using tin-(II) acetylacetonate Sn(acac)2. After a few cycles, material removal stopped and growth of a Sn-containing layer was observed. Insertion of a hydrogen plasma step was found to remove the Sn layer and a continuous material removal of 0.5 A/cycle was measured. The results show that plasma assistance can be used to realize thermal ALE of Al2O3. Specifically, plasma can be used both in the fluorination step and to keep the surface free from contaminations.

Following Moore’s Law, feature dimensions will soon reach dimensions on an atomic scale. For the most advanced structures, conventional plasma etch processes are unable to meet the requirement of atomic scale fidelity. The breakthrough that is needed can be found in atomic layer etching or ALE, where greater control can be achieved by separating out the reaction steps. In this paper, we study selective, directional ALE of silicon using plasma assisted chlorine adsorption, specifically selectivities to bulk silicon oxide as well as thin gate oxide. Possible selectivity mechanisms will be discussed. © The Author(s) 2015. Published by ECS. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/2.0031506jss] All rights reserved.

Plasma assisted atomic layer etching (ALE) has recently been introduced into manufacturing of 10 nm logic devices. This implementation of ALE is called directional ALE because ions transfer momentum to the etching surface during the removal step. Plasma assisted directional ALE can be described as sputtering of a thin modified layer on the surface of the unmodified material. In this paper, the authors introduce a collision cascade based Monte Carlo model based on sputtering theory which has evolved for over 50 years [P. Sigmund, Thin Solid Films 520, 6031 (2012)]. To test the validity of this approach, calculated near threshold argon ion sputtering yields of silicon and chlorinated silicon are compared to published experimental data. The calculated ALE curve for Cl2/Ar ALE of tantalum is in good agreement with the experiment. The model was used to predict the presence of salient sputtering effects such as ion mass and impact angle dependence, as well as redeposition in directional ALE. Finally, the author...

Challenges, new and old, are outlined for etching at the 45 nm technology node and below. The over-riding challenges pertain to improving selectivity over a wider range of materials and topographies and predicting pattern density behavior to reduce cost in mask design.

Relentless scaling of advanced integrated devices drives feature dimensions towards values which can be expressed in small multiples of the lattice spacing of silicon. One of the consequences of dealing with features on such an atomic scale is that surface properties start to play an increasingly important role. To encompass both dimensional as well as compositional and structural control, we introduce the term “atomic scale fidelity.” In this paper, we will discuss the challenges as well as new solutions to achieve atomic scale fidelity for patterning etch processes. Fidelity of critical dimensions (CD) across the wafer is improved by means of the Hydra Uniformity System. Wafer, chip and feature level atomic scale fidelity such as etch rate uniformity, aspect ratio dependent etching (ARDE) /1/, selectivity and surface damage can be addressed with emerging atomic layer etching (ALE) approaches /2/.

A semiempirical profile simulator to predict topographic evolution during Cl2/BCl3 plasma etching of photoresist patterned Al lines has been developed. Given incident flux distributions, the profile simulator uses a combination of a particle based Monte Carlo algorithm and analytic ray-tracing algorithm for solving feature-scale ion and neutral flux transport, respectively. We use angular and energy distributions for reflected ions that are consistent with experimental observation and molecular dynamic simulations. Etch yields with energy and angular dependence are experimentally determined for physical sputtering and ion-enhanced etching. The spontaneous etch rate of A1 by chlorine and the spontaneous desorption rate of Cl from photoresist are estimated from experimental results. Sticking coefficients for etchant, chlorine, and depositor, CClx, and depositing flux are determined by fitting simulated profiles to experimental data. A semiempirical site-balance model is developed to compute the surface cove...

Atomic layer etching (ALE) is a multistep process used today in manufacturing for removing ultrathin layers of material. In this article, the authors report on ALE of Si, Ge, C, W, GaN, and SiO2 using a directional (anisotropic) plasma-enhanced approach. The authors analyze these systems by defining an “ALE synergy” parameter which quantifies the degree to which a process approaches the ideal ALE regime. This parameter is inspired by the ion-neutral synergy concept introduced in the 1979 paper by Coburn and Winters [J. Appl. Phys. 50, 5 (1979)]. ALE synergy is related to the energetics of underlying surface interactions and is understood in terms of energy criteria for the energy barriers involved in the reactions. Synergistic behavior is observed for all of the systems studied, with each exhibiting behavior unique to the reactant–material combination. By systematically studying atomic layer etching of a group of materials, the authors show that ALE synergy scales with the surface binding energy of the bu...

Ion incidence angle distribution on surfaces in plasma etching reactors determines the shape evolution of via holes and trenches through its effects on the spatial variation of ion fluxes along the walls of these microscopic features. We describe a novel retarding-field energy analyzer design that is capable of measuring the energy and the incidence angle distributions of ions bombarding grounded surfaces in plasma reactors with sub-0.5° resolution. Using this analyzer we measured the energy and angle distributions of Ar ions incident onto a Si surface in a low-pressure helicon wave excited Ar plasma. Ion angle distributions are approximately Gaussian. In absence of collisions in the sheath, the width of the ion angle distribution function is determined by the ratio of the directed energy gained in the sheath to the random ion energy in the plasma. Variation of the ion angle distribution width as a function of plasma power and pressure is determined by the dependence of the sheath potential and the ion temperature on these externally controlled parameters. In low pressure Ar plasmas, the ion angle distribution broadens with increasing power and shows a maximum as a function of pressure in the range 0.5–4 mTorr.

The ever-shrinking dimensions of microelectronic devices has mandated the use of plasma processing in integrated circuit (IC) factories worldwide. Today the plasma-processing industry has grown to over $3 billion in revenues per year, well in excess of predictions made only a few years ago. Plasma etching and deposition systems are also found throughout flat-panel-display (FPD) factories despite the much larger dimensions of thin-film transistors (TFTs) that are used to switch picture elements (pixels) on and off. Besides the use of plasma in etching and depositing thin films, other processes include the following: removal of photoresist remnants after development (descumming), stripping developed photoresist after pattern transfer (ashing), and passivating defects in polycrystalline material. Why are plasma processes so prevalent?In etching, plasmas are used for high-fidelity transfer of the photolithographically defined pattern that defines the device or circuit. More generally, plasma provides the means to taper sidewalls. In Si processing, the sidewalls must be nearly vertical to obtain high density integration and faster performance. However in making FPDs, sidewalls are tapered to obtain uniform step coverage and reduce shorting. In deposition, plasmas are used to enable processing at low temperature. For both etching and deposition, only plasma processing provides an economically viable means for processing large area substrates: 300 mm for Si and more than 550 × 650 mm for FPDs. It is the ability to scale uniform reactant generation to larger areas that sets plasma apart from beam-based processes that might otherwise offer the desired materials modifications. The nonequilibrium characteristics of plasma further distinguish this processing method. Energetic electrons break apart reactant precursors while ions bombard the surface anisotropically.

As the IC industry approaches devices with half pitches below 10 nm, etching requires atomic scale fidelity because the device dimensions and their allowed tolerances are of the same order of magnitude as the inter-atomic distances in the crystal lattice. Atomic Layer Etching is emerging as a solution and plasma assisted or directional ALE is being adopted in the manufacturing of integrated devices /1/. Separation of the etching process into single unit processes allows to apply the large body of knowledge in surface science to improve fundamental understanding. In this talk we will focus on the fundamental processes at the wafer surfaces during etching of various materials with different ALE approaches. Strategies to obtain low physical and chemical surface damage will be presented. 1. K.J. Kanarik, T. Lill, E. Hudson, S. Tan, S. Sriraman, J. Marks, V. Vahedi, R.A. Gottscho; J. Vac. Sci. Technol. A 33, 020802 (2015)

Techniques available for real time, in sifu monitoring of surface chemistry during plasma processing are reviewed. Emphasis is given to recent uses of attenuated-total-reflection Fourier transform infrared (ATR-FTIR) spectroscopy to study surface bond vibrations during plasma cleaning and passivation of Si and GaAs surfaces. Examples are chosen to demonstrate the utility of real-time monitoring of surfaces for developing and optimizing plasma processes used in manufacturing of electronic and optoelectronic devices.

Electron cyclotron resonance (ECR) plasma reactors are being used for ultralarge scale integrated circuit fabrication to meet the stringent requirements on submicron feature etching. Three issues are critical for ECR reactor design: plasma uniformity, ion energy control, and wafer temperature control. Plasma uniformity is important for minimizing over etch times and reducing the probability of producing charging damage. Ion energy control is needed to optimize etching rate, anisotropy, and selectivity without compromising device yield. Wafer temperature control is important because large ion currents at low pressure can result in wafer heating and thereby alter the rates of surface chemical processes. An ECR plasma reactor is described that is designed to etch compound semiconductors and Si at low temperatures (−170 to 20 °C), where superior selectivity and linewidth control are achievable. By measuring dc bias, floating potential, and ion saturation current densities it is shown that ion energies in this...

Cleaning wafer surfaces in situ using gaseous reagents facilitates automated cluster tool processing of microelectronic devices. Moreover, to remain within stringent thermal processing budgets for making ultralarge scale integrated (ULSI) circuit devices, it is desirable to perform these dry cleaning processes at low temperatures. We report here the use of attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy to study and optimize room-temperature plasma cleaning of native oxide contaminated Si surfaces. By monitoring the vibrational modes of Si-H, H-SiO, O-H, and C-H in real-time, we develop a hybrid process where an upstream microwave discharge and the gas-flow rate are used to control the neutral atom flux while direct exposure to a radio-frequency plasma is used to control the ion energy flux

Although electron cyclotron resonance (ECR) plasma reactors are being used in the microelectronics industry for etching and deposition of thin films they are prone to instabilities that can wreak havoc with manufacturing processes. Plasma conditions are often unstable because of nonlinear coupling between wave propagation, power absorption, neutral density, and charge density profiles. We report on hysteresis, multiple steady states and abrupt transitions in an ECR plasma reactor that can alter the plasma properties drastically. Substrate platen floating potential and Langmuir probe measurements are used to identify several abrupt transitions and regions in the operating parameter space where the plasma can exist in either one of two different states under identical operating conditions (microwave power, pressure, flow rate, magnetic field, rf bias etc.) Depending on how the plasma is started either branch can be obtained. Abrupt transitions and hysteresis in plasma properties are observed both with and w...

As device design rules continue to shrink for the manufacturing of integrated circuits, unprecedented challenges for process inspection appear. No longer is optical microscopy adequate for determining if process results meet specifications. On the other hand, the alternative—scanning electron microscopy—is time consuming, destructive, and costly. Another approach is to measure scattered light intensity as a function of scattering angle, as opposed to imaging, to obtain distinct signatures for submicron structures. In this work, a set of Si wafers with photolithographically defined lines and spaces are reactively ion etched. By varying process conditions, a range of depths and sidewall profiles is generated and then inspected by detecting visible scattered laser light over 180°. The resultant scattergrams are then analyzed both by using discriminant analysis and by training a neural network to catalog the microstructures according to depth and profile. We find that this approach is a viable alternative to ...

Dry etching of compound semiconductors is becoming increasingly important as design ruler shrink for electronic devices. For photonic device applications, dry or plasma etching is used fin- device isolation, fine-line pattern transfer, and fabrication of optical quality interfaces. As has been well established for Si and W. plasma etching at reduced temperatures can provide superior critical dimension control and obviate the need for operating at high bias voltages that produce excessively energetic ion bombardment t. In this work, we explore low-temperature (−60 C to +60 C) etching of the compound semiconductors GaAs, AlGaAs, and AlAs, In addition to improving etch anisotropy, which provides critical dimension control, rye find thut processing at lower temperatures improves microuniformity and reduces loading effects. At high lemperaturcs, where larger samples are observed to etch more slowly than smaller pieces (loading effect), etching rates appear limited bv reactant transport to the wafer. In this regime, both microuniformity and macrouniformity arc poor. As the temperature is reduced, the etching rate becomes limited by surfitce processes us a residue containing the semiconductor elements, etchant gases, and residual background gases forms on the surface. hi this regime, the etch rare becomes independent of surface area and uniformity is improved.

Unprecedented demands for uniformity, throughput, anisotropy, and damage control in submicron pattern transfer are spurring development of new, low pressure, high charge density plasma reactors. Wafer biasing, independent of plasma production in these new systems is intended to provide improved ion flux and energy control so that selectivity can be optimized and damage can be minimized. However, as we show here, an inherent property of such discharges is the generation of significant densities of excited, metastable ionic states that can bombard workpiece surfaces with higher translational and internal energy. Absolute metastable ion densities are measured using the technique of self-absorption, while the corresponding velocity distributions and density scaling with pressure and electron density are measured using laser-induced fluorescence. For a low pressure, helicon-wave excited plasma, the metastable ion flux is at least 24% of the total ion flux to device surfaces. Because the metastable ion density scales roughly as the reciprocal of the pressure and as the square of the electron density, the metastable flux is largest in low pressure, high charge density plasmas. This metastable ion energy flux effectively limits ion energy and flux control in these plasma reactors, but the consequences for etching and deposition of thin films depend on the material system and remain an open question.

The electronics revolution of the past 30 years could not have taken place without plasma processing. Today millions of bits of information can be stored on a sliver of silicon no greater than a square centimetre. If alternative methods were used to pattern and deposit thin films, such chips and the products they comprise – from televisions and microwave ovens to the guidance systems of smart weaponry – would be prohibitively expensive. Plasma processing – the etching, deposition and cleaning of thin films – is essential for the economical fabrication of very-large-scale integrated (VLSI) electronic circuits. In 1990 the plasma processing industry had worldwide revenues of approximately $1 billion, and the microelectronics and electronics industries that it enabled had revenues of $50 billion and $750 billion respectively. With this enormous economic lever-age, the importance of understanding plasma processing, and fully exploiting its potential, cannot be over-emphasised.

Techniques available for real time, in sifu monitoring of surface chemistry during plasma processing are reviewed. Emphasis is given to recent uses of attenuated-total-reflection Fourier transform infrared (ATR-FTIR) spectroscopy to study surface bond vibrations during plasma cleaning and passivation of Si and GaAs surfaces. Examples are chosen to demonstrate the utility of real-time monitoring of surfaces for developing and optimizing plasma processes used in manufacturing of electronic and optoelectronic devices.

For applications in ultralarge scale integration, low pressure, high density plasmas are being developed for etching and deposition of thin films. To control critical parameters such as the flux and energy distribution of ions impacting surfaces, it is necessary to understand how these parameters are influenced by physical and electromagnetic design. In this work, we extend previous measurements of ion velocity distributions in Ar/He electron cyclotron resonance plasmas to Cl2/He plasmas. Using Doppler‐shifted laser‐induced fluorescence spectroscopy, we measure metastable Cl ion velocity distributions parallel and perpendicular to the magnetic field as a function of magnetic field amplitude, pressure, and microwave power. We also examine the effects of the wafer platen on the distribution functions by repeating the measurements after removing the platen. Surprisingly, little qualitative difference is seen when chlorine and argon discharges are compared; this is most likely a result of the low pressures em...

No abstract available

Electron cyclotron resonance (ECR) plasma reactors are being developed for etching and deposition of thin films during integrated circuit fabrication. To control critical parameters such as the flux and energy distribution of ions impacting surfaces, it is necessary to understand how these parameters are influenced by physical construction, electromagnetic design, and chemical kinetics. In this work, we report detailed measurements of ion velocity distributions in both the source and reactor regions of an ECR system using mixtures of Ar and He. Using Doppler‐shifted laser‐induced fluorescence spectroscopy, we measure metastable Ar‐ion velocity distributions parallel and perpendicular to the magnetic field direction as a function of magnetic field amplitude, pressure, rf bias voltage, and microwave power. The measurements, in turn, are used to estimate the magnitude of electrostatic potentials and fields parallel and perpendicular to the magnetic field. Indicative of ion trapping, we find nearly isotropic ion velocity distributions when the source is operated as a magnetic mirror and the He partial pressure is low; higher He pressures tend to cool the parallel velocity distribution. Downstream, we consistently observe bimodal ion velocity distributions: the fast component, created in the source, follows magnetic flux lines into the reactor; the slow component, created mostly where the plasma expands from the source into the reaction chamber, is more isotropic. The relative amplitudes of these two components, the average ion energy, and the ion energy distribution are easily controlled by changing pressure and magnetic field.

The average electron densities in CH4/H2/Ar and CCl2F2/O2 electron cyclotron resonance discharges typical of those used for etching InP, GaAs, and related materials have been measured approximately 4 cm downstream from the multipolar microwave source as a function of microwave power (50–300 W), additional radio‐frequency power (10–50 W), pressure (1–20 mTorr), flow rate (30–90 standard cubic centimeters per minute) and gas composition. At 1 mTorr pressure and 10 W rf, the electron densities (and semiconductor etch rates) increase rapidly with microwave power, from 1.3×1011 cm−3 for 5CH4/17H2/8Ar and 6×1010 cm−3 for 28CCl2F2/2O2 discharges at 50‐W microwave power, to 9×1011 cm−3 and 3×1011 cm−3, respectively at 300‐W microwave power. At the highest microwave power levels (≥200 W) the InP and GaAs etched surface morphologies are rough due to preferential removal of one of the lattice constituents from each material. The electron densities in both types of discharge show moderate increases with increasing rf...

The performance of today's electronic and photonic systems is largely limited by interconnection technology. Using silicon-based multichip modules and hybrid optical packaging as exemplary models, we show how high-performance interconnection technology can be achieved at low cost. The key is to choose materials and processes that are compatible with today's very-large-scale integrated-circuit manufacturing.

A long‐standing problem in plasma processing has been imprecise control and characterization of dissipated power. In particular, the partitioning of energy between gas phase and surface is not easily determined. Yet, for plasma process control and transfer from one reactor to another, it is important to monitor and control energy flux to the wafer. Using pulsed laser photoluminescence spectroscopy, we measure wafer temperatures nonintrusively as a function of time when radio frequency discharges are gated on and off. By so doing, the energy flux to the wafer surface is determined as a function of pressure, power, and frequency. This energy flux is then compared to the total power dissipated, which is calculated from measured current and voltage waveforms. We find that the fractional power dissipated at the wafer surface varies inversely with pressure regardless of applied power and frequency.A long‐standing problem in plasma processing has been imprecise control and characterization of dissipated power. In particular, the partitioning of energy between gas phase and surface is not easily determined. Yet, for plasma process control and transfer from one reactor to another, it is important to monitor and control energy flux to the wafer. Using pulsed laser photoluminescence spectroscopy, we measure wafer temperatures nonintrusively as a function of time when radio frequency discharges are gated on and off. By so doing, the energy flux to the wafer surface is determined as a function of pressure, power, and frequency. This energy flux is then compared to the total power dissipated, which is calculated from measured current and voltage waveforms. We find that the fractional power dissipated at the wafer surface varies inversely with pressure regardless of applied power and frequency.

Atomic layer etching (ALE) is a technique for removing thin layers of material using sequential reaction steps that are self-limiting. ALE has been studied in the laboratory for more than 25 years. Today, it is being driven by the semiconductor industry as an alternative to continuous etching and is viewed as an essential counterpart to atomic layer deposition. As we enter the era of atomic-scale dimensions, there is need to unify the ALE field through increased effectiveness of collaboration between academia and industry, and to help enable the transition from lab to fab. With this in mind, this article provides defining criteria for ALE, along with clarification of some of the terminology and assumptions of this field. To increase understanding of the process, the mechanistic understanding is described for the silicon ALE case study, including the advantages of plasma-assisted processing. A historical overview spanning more than 25 years is provided for silicon, as well as ALE studies on oxides, III–V compounds, and other materials. Together, these processes encompass a variety of implementations, all following the same ALE principles. While the focus is on directional etching, isotropic ALE is also included. As part of this review, the authors also address the role of power pulsing as a predecessor to ALE and examine the outlook of ALE in the manufacturing of advanced semiconductor devices.

We provide experimental evidence for a correlation between the characteristic energies of the exponential conduction‐ and valence‐band tails of hydrogenated amorphous silicon (a‐Si:H). We measured subgap optical‐absorption spectra of isolated a‐Si:H films in order to extract their valence‐band‐tail energies. Also, we modeled the current–voltage characteristics of thin‐film transistors which had a‐Si:H layers deposited identically to the isolated films, in order to extract their conduction‐band‐tail energies. When the quality of the a‐Si:H was varied, the characteristic energies of the two tails scaled linearly with each other.

Real-time, in situ observations of surface chemistry during the remote plasma passivation of GaAs is reported herein. Using attenuated total reflection Fourier transform infrared spectroscopy, the relative concentrations of -As-O, -As-H, -H2O, and -CH2 bonds are measured as a function of exposure to the effluent from a microwave discharge through NH3, ND3, H2, and D2. The photoluminescence intensity (PL) from the GaAs substrate is monitored simultaneously and used qualitatively to estimate the extent of surface state reduction. It was found that, while the -CHx(x = 2,3) and -As-O concentrations are reduced rapidly, the rates at which the -As-H concentration and the PL intensity increase are relatively slow. The concentration of -H2O on the GaAs surface increases throughout the process as surface arsenic oxides and the silica reactor walls are reduced by atomic hydrogen. These observations suggest that removal of elemental As by reaction with H at the GaAs–oxide interface limits the passivation rate.

Although plasmas are used throughout the microelectronics industry for etching, deposition, and cleaning of thin films, control of plasma processes has been a long‐standing problem. Because of the nonlinear properties of plasmas, such as the coupling between wave propagation, density profile, and power absorption, plasma reactors are prone to unstable operation, multiple steady states, and hysteresis. We report observation and suppression of an abrupt transition in the operating mode of an electron cyclotron resonance reactor that alters the ion flux to device wafers by more than twofold. While the origin of this mode change is not well understood, we show here that it is strongly correlated with the neutral gas density, which slowly decreases as the reactor temperature increases during a process or from run to run. By measuring the quartz liner temperature and adjusting the pressure to maintain an approximately constant neutral gas density, the mode change can be avoided indefinitely. In a simulated manu...

Plasma etching has been enabling nano-electronic fabrication since the 1980s; during this time, transistor size has shrunk by nearly two orders of magnitude, starting at 1.0 µm in the mid 80s to ~0.01 µm today. The manufacturing of these devices requires overcoming a series of challenges, ranging from continuous innovation on device integration to extend Moore's law to breaking tradeoffs on the perennial challenge of aspect ratio-dependent etching. In this paper, we will review four key areas in etch manufacturing: uniformity, defects, surface precision and 'sticky'/non-volatile etch materials. In the uniformity section, we will discuss the challenges for microscopic uniformity, such as localized feature dimension variations; macroscopic uniformity, such as performance at the extreme edge of the wafer; and repeatable uniformity, meaning wafer-to-wafer, lot-to-lot and chamber-to-chamber performance. While defect management is successful with in situ plasma cleans, one must be cognizant of the choice of clean chemistry. In surface precision, we look at the approach of atomic layer etching and how it can be successful in a manufacturing environment. Finally, in the non-volatile material section, we review technology drivers for DRAM (dynamic random access memory) and NAND flash memory in the microelectronics Si industry, with focus on the utilization of such materials and what it means to etch equipment manufacturers.

The role of hydrogen atoms in enhancing photoluminescence (PL) intensity of native‐oxide‐contaminated GaAs during room temperature, short‐time exposure to a H2 plasma is studied. Room‐temperature D2 plasma passivations, annealing experiments in vacuum at 400 °C, D atom depth profiling with secondary ion mass spectroscopy, and chemical etching of residual surface oxides after passivation are all consistent with a reduction in surface recombination velocity leading to enhanced PL. By contrast, passivation of deep level defects in bulk GaAs cannot account for all the observations. The passivated surface is stable up to 400 °C in vacuum for at least an hour with only a slight reduction in PL intensity. Thus, it is possible to reactivate inadvertently neutralized donors and acceptors after passivation without affecting the reduction in surface recombination velocity.

Ion and neutral temperatures are measured by high‐resolution laser‐induced fluorescence spectroscopy both in the source and downstream of an electron cyclotron resonance discharge through mixtures of Ar, Ar/Ne, and Ar/He. Contrary to previous reports, both ions and neutrals are found to be cold. In the source, ion temperatures perpendicular to the magnetic field are ≤0.5 eV; downstream they are ∼0.25 eV. Neutral temperatures in the source and downstream are 0.068 and 0.030 eV, respectively.

Electron cyclotron resonance plasma sources are gaining widespread use in plasma processing because they offer high ion flux with controllable energy and thereby high etching and deposition rates with minimal damage. However, it is unclear how ion energy distributions evolve from source to wafer as a function of plasma parameters such as pressure, microwave power, and magnetic field strength. Therefore, we used Doppler broadened and shifted laser‐induced fluorescent line profiles to measure Ar+ metastable ion velocity distributions downstream from a divergent magnetic field electron cyclotron resonance source. Spatially resolved distributions, measured at positions above and across a wafer platen, differ markedly from shifted Maxwell–Boltzmann functions. Ions are accelerated along the magnetic field direction by a weak (∼0.5 V/cm), ambipolar electrostatic field. The ion energy component perpendicular to the electric field corresponds to a temperature of only 0.46±0.10 eV. On the edges of the platen, the m...

By cooling substrates to low temperatures (–40 °C), plasma etching of AlGaAs/AlAs/GaAs structures is performed in an ion-activated, surface reaction limited regime. As a result, microscopic and macroscopic uniformity are vastly improved and etching is independent of gas flow patterns, plasma geometry, and reactor loading. Because the reactant is concentrated on the surface, etching rates remain large.

An array of novel hollow cathode plasma sources of 4 mm diameter and driven by up to 500 W of 13.56 MHz rf power was constructed. It was investigated using spatially resolved Langmuir probe measurements in the plasma expansion region. Initial results suggest the plasma observed downstream is the combination of a diffusion from the exit orifice of the 30 mm long cylindrical source region and a uniform plasma created in the expansion region. By measuring the ion density of the plasma plumes produced by two or more active hollow cathode sources, their mutual interaction has been inferred with the aim of determining whether a much larger array of sources could be envisioned. The effectiveness of the plasma in dissociating reactive species was tested using SF6 and measuring etch rates on unbiased silicon wafers. The results have been modelled and show that it is possible to produce a uniform spread of ±5%.

We evaluated the characteristics of a-Si:H/a-SiNx:H thin film transistors (TFTs), and of separately deposited a-Si:H films, as functions of the a-Si:H deposition power in a high-rate, large-area, 40 MHz PE-CVD system. TFT performance and a-Si:H film properties improve with decreasing power density and deposition rate. However, low defect density a-Si:H material was deposited at rates as high as 1500 Å/min. TFTs with gate nitride deposited at 1000 A/min show excellent I-V characteristics when the a-Si:H deposition power is low enough. The TFT electron mobility in the linear regime correlates well with the Urbach energy of the a-Si:H films, suggesting that the quality of the a-Si: H controls the performance of our TFTs.

We describe an experiment in which the etch depth of a diffraction grating is measured. A simulated experiment is used to develop and calibrate the measurement technique. A scatterometer was used to measure the diffraction patterns of a set of 5 wafers at 14 die locations. The estimator already developed is then used to find the etch depths at the 70 measured locations. Finally, a scanning force microscope is used as a reference method to validate the scatterometer measurements.

For applications in ultra-large scale integration, low pressure, high density plasmas are being developed for etching and deposition of thin films. To control critical parameters such as the flux and energy distribution of ions impacting surfaces, it is necessary to understand how these parameters are influenced by physical and electromagnetic design. In this work, we report measurements of ion velocity distributions in Ar/He and Cl2/He electron cyclotron resonance plasmas. Using Doppler-shifted laser-induced fluorescence spectroscopy, we measure metastable Ar and Cl ion velocity distributions parallel and perpendicular to the magnetic field as a function of magnetic field amplitude, pressure, and microwave power. We also examine the effects of the wafer platen on the distribution functions by repeating the measurements after removing the platen. We find nearly isotropic ion velocity distributions when the source is operated as a magnetic mirror and the He partial pressure is low; higher He pressures tend to cool the parallel velocity distribution. Downstream, we consistently observe bimodal ion velocity distributions: the fast component, created in the source, follows magnetic flux lines into the reactor; the slow component, created mostly where the plasma expands from the source into the reaction chamber, is more isotropic. The relative amplitudes of these two components, the average ion energy, and the ion energy distribution are easily controlled by changing pressure and magnetic field.

Light scattering methods are being used in semiconductor device fabrication for such diverse purposes as particulatemonitoring and linewidth control. In this overview, we focus on applications of light scattering for monitoring reactivelyion etched trench profiles and plasma passivation ofIII-V materials. 1. INTRODUCTION Light scattering methods have long been useful for process control and monitoring because they are relatively simple toimplement and are inherently non-intrusive. Here, we distinguish between light scattering and microscopy in that the latterentails the generation of an image. Similarly, we distinguish between light scattering and interferometry as the latter entailsthe use of either a transmitted or specularly reflected beam. By contrast, light scattering methods make use of visible, nearir, and near-uv radiation impinging on a sample and detection of inelastically and/or elastically scattered light at an angle orangles different from either the incident or specular angles. For semiconductor device fabrication, light scattering is usedfor detecting and counting particles in the processing medium (gas or liquid) and on wafer surfaces.' It has been used formonitoring surface roughness after sputter deposition of metal films,2 and for detecting latent images in photoresist as ameans of optimizing lithographic exposure equipment settings and controlling exposure and development processes.3 For

Laser scatterometry is a technique which involves shining a light beam on an area to be characterized and measuring the angular distribution of the light that is scattered from that area. A laser is used so that the incident light will be monochromatic and coherent. It is also valuable, in many applications, to be able to confine the probe beam to a selected area, with a diameter of 10 um or more. The scattered light distribution is typically measured either by scanning a detector over an arc, using a fixed array of photodiodes that are mounted along an arc, or by measuring the scattered light intensity distribution at a hemispherical screen when sample symmetry requires more than a one dimensional distribution measurement. In most cases, a single line measurement along a 90 deg arc is sufficient, either because of sample symmetry or because of the ability to align the sample in the direction of interest. The wavelength of the light used is a determining factor in the range of feature sizes that will be measured. Thus, different wavelength sources are sometimes valuable. Also, the angle of incidence of the probing light beam is a factor in the range of features that will be characterized.

not Available.