Line Edge Profiles Research Articles

Tabletop extreme ultraviolet reflectometer for quantitative nanoscale reflectometry, scatterometry, and imaging.

Imaging using coherent extreme-ultraviolet (EUV) light provides exceptional capabilities for the characterization of the composition and geometry of nanostructures by probing with high spatial resolution and elemental specificity. We present a multi-modal tabletop EUV imaging reflectometer for high-fidelity metrology of nanostructures. The reflectometer is capable of measurements in three distinct modes: intensity reflectometry, scatterometry, and imaging reflectometry, where each mode addresses different nanostructure characterization challenges. We demonstrate the system's unique ability to quantitatively and non-destructively measure the geometry and composition of nanostructures with tens of square microns field of view and sub-nanometer precision. Parameters such as surface and line edge roughness, density, nanostructure linewidth, and profile, as well as depth-resolved composition, can be quantitatively determined. The results highlight the applicability of EUV metrology to address a wide range of semiconductor and materials science challenges.

Unbiased line edge roughness measurement using profile-averaging method for precise roughness parameters measurement

Background: Scanning electron microscopy (SEM)-based line edge roughness (LER) measurement suffers from an error due to noise in SEM image. Noise correction methods have been developed to obtain unbiased roughness results, however, there are still concerns in the viewpoint of measurement precision. Aim: To develop an unbiased roughness analysis for highly precise LER measurement. Approach: Combining the conventional unbiased roughness analysis with a profile-averaging method, where a line pattern is repeatedly measured and then the obtained profiles are aligned and averaged. The experimental result measured by SEM was verified using atomic force microscopy (AFM)-based LER metrology, which has higher reliability than SEM. Results: The experimental evaluation showed that the proposed method can obtain roughness parameters more precisely than the conventional method. Conclusions: When the noise in the line edge profile by SEM is too large, it is necessary to reduce the noise beforehand and then perform roughness analysis in order to obtain precise roughness results. The proposed method enables to measure LER with the highest precision using SEM. Additionally, the AFM-based LER metrology was demonstrated as a feasible technique to evaluate the performance of SEM-based LER metrology.

Direct comparison of line edge roughness measurements by SEM and a metrological tilting-atomic force microscopy for reference metrology

Background: Conventional scanning electron microscopy (SEM) that is used for 2D top-view metrology, a classical line edge roughness (LER) measurement technique, is incapable of measuring 3D structures of a nanoscale line pattern. For LER measurements, SEM measurement generates a single line-edge profile for the 3D sidewall roughness, although the line-edge profile differs at each height in the 3D sidewall. Aim: To develop an evaluation method of SEM-based LER measurement techniques and to verify how the 3D sidewall shape is reflected in the SEM’s 2D result. Approach: Direct comparison by measuring an identical location of a line pattern by SEM and an atomic force microscopy (AFM) with the tip-tilting technique that is capable of measuring the 3D sidewall. The line pattern has vertical stripes on the sidewall due to its fabrication process. Measured line edge profiles were analyzed using power spectral density, height-height correlation function, and autocorrelation function. Results: Line edge profiles measured by SEM and AFM were well matched except for noise level. Frequency and scaling analyses showed that SEM profile contained high noise and had lost a property of self-affine fractals in contrast to AFM. Conclusions: In the case of the line pattern with vertical stripes on the sidewall, SEM profile is generally consistent with 3D sidewall shape. The AFM-based LER measurement technique is useful as LER reference metrology to evaluate other LER measurement techniques.

Measurements of CD and sidewall profile of EUV photomask structures using CD-AFM and tilting-AFM

Accurate and traceable measurements of critical dimension (CD) and sidewall profile of extreme ultraviolet (EUV) photomask structures using atomic force microscopes (AFMs) are introduced in this paper. An instrument complementarily applied with two kinds of AFM techniques, the CD-AFM and the tilting-AFM, has been developed. High measurement stability of the instrument is demonstrated, for instance, the long-term CD stability is better than 1 nm over 500 successive measurements over 55 h. To traceably calibrate the effective tip geometry, transmission electron microscopes-based method is applied, which uses either the silicon crystal lattice or the structure pitch value calibrated by metrological AFMs as an internal scale. Several grating patterns with different nominal CDs and line/space ratios of an EUV photomask have been measured using the developed methods. A data evaluation method with considered higher order tip effect due to the non-vertical sidewall is introduced. Detailed measurement results of a test EUV photomask, such as middle CD, left and right sidewall angle, feature height, line edge roughness and edge profiles are given. Finally, the AFM results are compared to that of a PTB EUV scatterometer. The comparison of the middle CD yields a linear relation within a spread of only about ±2 nm and an offset of the absolute values below 3 nm. For the sidewall angle, both methods yield consistent results within a range of about 2°.

Linking EUV lithography line edge roughness and 16nm NAND memory performance

Resist roughness is one of the effects of process uncertainties in extreme UV lithography. All the lithographic elements, such as source, mask, optical system, resist and metrology, contribute to line roughness. As a result, line edge roughness is still a challenge to be tackled for 22 and 16nm transistor devices. In this paper, the impact of post-litho smoothing processes such as ion-beam implant and plasma etch treatment on state-of-the-art extreme UV lithography is reported. Top-down and cross section electron microscopy were used to evaluate low and high frequency components of line edge roughness by means of power spectral density analysis. To compute roughness impact on bit error rate performance of floating gate devices, line edge profiles were first rebuilt, and then injected into a Monte Carlo variability aware simulator for electrical evaluation of a 16nm multi-level NAND Flash memory array. Electrical variability simulations based on actual experimental roughness profiles were executed to quantify the impact of line edge roughness on the device reliability. It was found that inter-transistor low frequency roughness impacts the memory performance with a failure factor 3-5 times higher compared to the same electrical model without considering any stochastic variation. Intra-transistor high frequency roughness showed only a marginal effect compared inter-transistor variations. Moreover, it was demonstrated that the bit error rate exponentially depend on resist line edge roughness. Finally, by combining both experimental analysis and electrical assessment, it was possible to quantify the contribution of post-litho smoothing processes on the error correction code performance for such memory devices.

Performance and Yield Benefits of Quasi-Planar Bulk CMOS Technology for 6-T SRAM at the 22-nm Node

The performance and threshold voltage variability of quasi-planar bulk MOSFETs are compared against those of conventional bulk MOSFETs, via three-dimensional (3-D) device simulations with gate line-edge roughness and atomistic doping profiles, at 25 nm gate length. The nominal performance of six transistor (6-T) SRAM cells is studied via 3-D simulation of full cell structures. Compact (analytical) modeling is used to estimate SRAM cell yields. As compared to conventional bulk CMOS technology, quasi-planar bulk CMOS technology provides for enhanced SRAM cell performance and yield, and hence facilitates reductions in cell area and operating voltage. It also enables a notchless 6-T SRAM cell design which is advantageous for improved lithographic printability and either smaller area or lower standby power, and is projected to achieve 6-sigma cell yields at operating voltages down to ~0.8 V.

Comparison of line edge roughness and profile angles of chemical vapor deposited amorphous carbon etched in O 2/N 2/Ar and H 2/N 2/Ar inductively coupled plasmas

In this study, we compared the line edge roughnesses (LER) and profile angles of chemical vapor deposited (CVD) amorphous carbon ( a-C) patterns etched in an inductively coupled plasma (ICP) etcher produced by varying process parameters such as the N 2 gas flow ratio, Q (N 2), and dc self-bias voltage (V dc) in O 2/N 2/Ar and H 2/N 2/Ar plasmas. The tendencies of the LER and profile angle values of the etched CVD a-C pattern were similar in both plasmas. The LER was smaller in the O 2/N 2/Ar than in the H 2/N 2/Ar plasmas, and the profile angle was larger in the O 2/N 2/Ar than in the H 2/N 2/Ar plasmas under the same processes conditions. The use of O 2/N 2/Ar plasma was more advantageous than the H 2/N 2/Ar plasma for controlling LER and profile angle.

Multiscale characterization method for line edge roughness based on redundant second generation wavelet transform

We introduce a multiscale characterization method for line edge roughness (LER) based on redundant second generation wavelet transform. This method involves decomposing LER characteristics into independent bands with different spatial frequency components at different scales, and analyzing the reconstructed signals to work out the roughness exponent, the spatial frequency distribution characteristics, as well as the rms value. The effect of noise can be predicted using detailed signals in the minimum space of scale. This method was applied to numerical profiles for validation. Results show that according to the line edge profiles with similar amplitudes, the roughness exponent R can effectively reflect the degree of irregularity of LER and intuitively provide information on LER spatial frequency distribution.

Motif Parameters Based Characterization of Line Edge Roughness of a Nanoscale Grating Structure

Motif parameters were introduced to characterize line edge roughness (LER) of a nanoscale grating structure. Firstly with electron beam lithography employed the expected nano-scale grating structure with linewidth of 16 nm was fabricated on positive resist. Then the line edge profiles of the structure were extracted and their LERs were characterized. The results showed that the evaluation method is rather simple, effective and recommendable.

Study of the evolution of nanoscale roughness from the line edge of exposed resist to the sidewall of deep-etched InP∕InGaAsP heterostructures

The evolution of line edge roughness and sidewall roughness was monitored during the fabrication of deep-etched optical waveguides in InP∕InGaAsP heterostructures. Scanning electron microscopy was used to extract line edge profiles of the electron beam exposed resist and the lifted-off NiCr metal mask. Atomic force microscopy with an ultrasharp tip was utilized to extract the sidewall profiles of InP∕InGaAsP mesa waveguides that were etched using inductively coupled plasma reactive ion etching. The processing step that critically influences the roughness of the etched waveguides was determined by studying the evolution of the roughness.

Simulation of the negative chemical amplification deep-ultraviolet process in integrated circuit fabrication

Chemically amplified resists are able to produce nearly vertical line-edge profiles from low-contrast images, because acid generated upon exposure catalyses a crosslinking reaction during the subsequent post-exposure bake. A three-dimensional model for the chemical amplification process and resist etching has been developed in the framework of this research work. Based on this model, an algorithm for the simulation of the chemical amplification process and resist etching process has been implemented. The final resist profiles produced by this algorithm are in very good qualitative agreement with experimental data, and the dependence of the final resist profile on the post-exposure bake time and the number of crosslinking sites has been successfully reproduced.

Characterization and modeling of materials for photolithographic simulation

Photolithography is one of the critical processes in IC manufacturing. Modeling of resist materials and simulation of line edge profiles can play an important role in understanding and balancing complex trade-offs in optical lithography. There exist well understood and reliable techniques for calculation of the aerial image. We discuss here the models used to calculate the chemical modifications which occur in the resist and the models used to predict the profiles after development. We present the physical principles underlying these models and the assumptions inherent in them. In addition we examine the experimental techniques that are used to acquire modeling parameters.

Characterization of a photoresist with wavelength selected tone

The line edge profile simulation of a tone-switching resist system, obtained from the dissolution model of dual-sensitized, novolak-based resist in aqueous developer, is described. The model incorporates the actinic response of both a positive and a negative sensitizer in a two-pattern lithographic process that simultaneously exposes the same resist film. These response data are combined with dissolution rate measurements to establish a model for the resist and carry out SAMPLE simulation of resist line edge profiles for contact and projection printing. The model predictions are compared with SEM micrographs of exposed resist features.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Computer-aided resist modelling with extended XMAS in x-ray lithography

There is an increasing need for computer-aided lithography simulation and modelling in IC manufacture. Software programs are well-established tools in simulating the problems affecting line-edge profiles for different exposure arrangements and development processes. The simulation program XMAS (X-Ray Lithography Modelling and Simulation) is designed to evaluate a large variety of exposure and development situations in X-ray lithography. After the imaging procedure, a simulation of the development process can be performed resulting in two or three-dimensional resist profiles at various stages in the process. Results are presented concerning the development behaviour of standard and experimental positive three-component-system (3CS) resists as well as experimental negative tone resists. The influence of secondary electrons backscattered from different substrate layers is investigated. The algorithm for three-dimensional development based on a ray-tracing formalism is described.

Monte Carlo simulation of line edge profiles and linewidth control in x-ray lithography

X-ray lithography exposures of PMMA on silicon have been simulated with a Monte–Carlo method including the energy backscattering from the substrate. The line-edge acuity of the resist image has been modeled with experimentally determined x-ray spectra of various sources (aluminum, tungsten, and palladium) and various source diameters (0.1, 1.5, and 3 mm). The intrinsic line-edge degradation due to secondary electron exposure is derived from the point source images while the finite-source effect (penumbra) is extracted from the additional degradation at finite source diameters. The penumbra effect is found to be rather small in PMMA. Depth profiles of the energy absorbed in PMMA demonstrate the significance of the energy backscattering from the substrate for energetic x rays. This effect can be significantly reduced by an organic spacer layer. The extent of the geometric projection effect has been modeled with the aluminum source. The results show that a finite angle of incidence of the x rays causes a sizable enlargement of the mask shadow due to the finite absorber thickness ultimately restricting the usable field size. The resultant linewidth variation is more pronounced for a mask with vertical profiles than for one with 20° wall angle. It can be reduced by thinning the absorber although at the expense of the mask contrast. The angle of incidence is replicated in the resist profiles causing an increasing tilt of the lines further away from the center of the wafer. This also restricts the field size but can be controlled to some extent by the thickness of the imaging resist layer.

Computer simulation of line edge profiles undergoing ion bombardment

A computer model of dry etching was used to predict line edge profiles obtained by the ion etching device built in our Institute. For the simulation, the etch rates as a function of impact angle data for gold, silicon, and chromium, resists AZ 1350 and PMMA were determined. These characteristics were approximated by means of analytical functions. It has been found that the accuracy of the etch rate versus impact angle is of great importance. The influence of this dependence on the predicted contour evolution has been investigated for the case of a gold hemisphere approximated by two-dimensional semicircular profile. The model was compared with known analytical results. The experimental verification of the model has been made on patterns developed in AZ 1350 on silicon, AZ 1350 on Al, and PMMA on Au. SEM observations of the evolution of edges under ion milling show good agreement between the experimentally obtained pattern profile and computer predictions.

Planar Plasma Etching of Polysilicon Using CCl4 and NF 3

A comparison is made between carbon tetrachloride and nitrogen trifluoride as etchants for plasma etching phosphorus‐doped polysilicon over silicon dioxide. The comparison is based upon optimal gas pressures and flow rates of and for a commercial planar plasma etcher. The etch rates of doped polysilicon and silicon dioxide were measured as a function of power density for both and . Polysilicon to silicon dioxide etch rate ratios of 15 to 1 were found for with accompanying very high etch rates (1–2 μm/min). Ten to one etch rate ratios were obtained with using low power densities and low etch rates (400 Å/min). Line edge profiles were examined and found to be quite isotropic for polysilicon etched with and anisotropic . Also studied was the effect on etch rates due to varying wafer loads and the rates of photoresist erosion.

Characterization of positive resist development

The development of positive photoresist is assumed to be a surface limited reaction whose development rate is determined by the local inhibitor concentration M, as defined by Dill et al. This paper describes a technique for determining development rate as a function of inhibitor concentration and demonstrates the usefulness of the technique for quality control. Resist development parameters for the simulation of line edge profiles in resist are presented.

Application of line-edge profile simulation to thin-film deposition processes

A computer simulation program based on a segment-motion model has been developed to investigate thin-film-deposition processes. This simulation program is used to study the physical phenomena in the thin-film-deposition process as well as the effects of different tooling configurations and operating conditions. The simulated results are compared with experimental results obtained from an electron-gun evaporator with planetary tooling and from a planarizing sputtered quartz system to verify the accuracy of the simulation and to illustrate imporant phenomena during the deposition process. The simulation program is also used in exploratory studies to find the optimum source angle for step coverage in a typical planetary evaporator. The step coverage is found to change only slightly for wafers at different locations on the planet, but can change significantly depending on the pattern orientation on the wafers if the wafer rotation is not independent of the planet rotation. A study of the sputtered quartz planarization process shows that the growth of different topological features can be accurately predicted from the change in deposition and etching rate as a function of the angle of incidence. In particular, the increased deposition near a conductor edge is attributed to the propagation of a growth front from the sidewall boundary conditions.

Developer Temperature Effects on E‐Beam and Optically Exposed Positive Photoresist

The effect of developer temperature on the development rates and line edge profiles of electron beam and optically exposed diazo‐type photoresists were investigated. An in situ thickness measurement tool (FTA) was used to monitor the development rates; and activation energies based on a simple Arrhenius dependence were calculated. For the positive diazo‐resist system AZ‐1350J exposed to e‐beam, the activation energies and line edge profiles are temperature dependent. Both optically and e‐beam exposed resists give greater contrast, straighter line edge profiles, and linewidth control at low temperatures (10°–15°C). For AZ‐2400 exposed optically and with e‐beam, the activation energies and line edge profiles are independent of temperature, indicating that developer temperature is a less critical parameter for AZ‐2400 than for AZ‐1350J.