This PDF file contains the editorial “Editorial: Reflection on JM3 over the Last Three Years” for JM3 Vol. 4 Issue 02

This PDF file contains the editorial “Editorial: Polarization and Hyper-NA Lithography” for JM3 Vol. 4 Issue 03

A technique using phase-shifting test mask patterns is introduced for monitoring polarization balance of the illumination in high numerical aperture (NA) and immersion projection printing systems. Test mask patterns are derived from high NA proximity effects and serve to scatter light into high angle spatial frequencies. This creates a central intensity dependent only on the local state of polarization. A test mask consisting of multiple patterns is proposed to monitor the polarization from any arbitrary illumination scheme. Proper calibration of the test reticle enables reasonable mask-making limitations and mask topography effects to be tolerated. A set of linear equations enable determination of the Stokes parameters from a series of resist images. Practical examples are simulated with rigorous electromagnetic theory. In resist, this technique is likely to monitor polarization with a sensitivity of more than 1.5% of the clear field per percent change in polarization state for on-axis illumination, or more than 0.75% for off-axis illumination. The effects of various realistic imaging conditions are discussed.

Water-based immersion lithography using ArF illumination is able to provide optical solutions as far as the 45-nm node, but is not able to achieve the 38- or 32-nm nodes as currently defined. Achieving these lithographic nodes will require new, higher refractive index fluids to replace the water used in first-generation immersion systems. We have developed a number of such second-generation high-index fluids for immersion lithography at 193 nm. These highly transparent fluids have 193-nm indices up to 1.664. To understand the behavior and performance of different fluid classes, we use spectral index measurements to characterize the index dispersion, coupled with Urbach absorption edge analysis and Lorentz Oscillator modeling. Interference imaging printers have long been available, and they now have a new use: a rapid, cost-effective way to develop immersion lithography, particularly at extremely high resolutions. Although interference printers will never replace classical lens-based lithography systems for semiconductor device production, they do offer a way to develop resist and fluid technology at a relatively low cost. Their simple image-forming format offers easy access to the basic physics of advanced imaging. Issues such as polarization of the image-forming light rays, fluid/resist interaction during exposure, topcoat film performance, and resist line edge roughness (LER) at extremely high resolutions, can all be readily studied. 32-nm 1:1 line/space (L/S) imaging is demonstrated using two of the second-generation fluids. These resolutions are well beyond current lens-based system capabilities. Results on the performance of various resists and topcoats are also reported for 32-nm L/S features.

The use of immersion technology will extend the lifetime of 193-nm technology by enabling numerical apertures (NAs) much greater than 1.0. The ultimate limits of NA are explored by analysis of polarization effects at the reticle and imaging effects at the wafer. The effect of reticle birefringence with polarized illumination is explored. The effects on critical dimension (CD) uniformity are mitigated if the maximum birefringence is <5 nm/cm. Hertzian or micropolarization due to the size of the reticle structures is examined through rigorous simulation. For the regime of interest, 20- to 50-nm imaging, it is found that dense features on a Cr binary reticle will polarize the light into the TE component upward of 15%. Below this regime, the light becomes polarized in the TM direction. The use of polarization in the illuminator for imaging will result in substantial gains in exposure latitude and mask error factor (MEF) when the NA~1.3 with 45-nm lines at 193 nm, with overall polarization effects increasing with decreasing k₁. The end-of-line pullback for 2-D patterns is reduced by the use of TE polarization in the illuminator. The interaction between the reticle-induced polarization and the illumination polarization is shown to be significant when an analysis is done using rigorous mask simulation instead of the more common Kirchhoff approximation. The impact of birefringence in the lens is analyzed using Jones pupil matrices to create a local polarization error in the pupil. The photoresist process is shown to interact with polarization. Different photoresists will show varying degrees of sensitivity to polarization variation.

Lincoln Laboratory has designed and fabricated a charge-coupled device (CCD) array capable of imaging both polarization and illumination uniformity. The device consists of an 1107-element linear array of UV-optimized silicon photodiodes readout by a three-stage CCD through a single ~1-MHz output amplifier. This yields an effective clock rate for the whole array of ~1 kHz. Each of the active diode surfaces within the 1107-element array is covered by a UV-opaque layer of polysilicon into which are patterned 140-nm, transmissive sampling slits. The orientation and location of the slits enables simultaneous determination of illumination uniformity, degree and direction of polarization, and polarization uniformity. The device is tested with a 193-nm excimer laser equipped with variably polarized illumination and the theoretical performance of the device was supported by finite-difference time domain optical simulations.

Optical proximity correction (OPC) tools are already equipped with the most advanced models for image formation and are capable of thin-film modeling, vector diffraction modeling, and polarization modeling. Accurate simulation of immersion lithography, even in the context of OPC, does not pose any particular difficulty. We use the optical simulator of a commercial OPC software to study source polarization and its impact in process latitude and in proximity and linearity curves. More than a 10-nm difference in both curves is observed versus source polarization at an immersion numerical aperture (NA)>1, projected to print the 45-nm node. Simulation of large and arbitrary layout snippets confirms these results and demonstrates the feasibility of using advanced models in the context of OPC. Also, dry and water-immersion lithography are compared at the same NA<1 and the main differences in imaging are highlighted. The depth-of-focus (DOF) increase in immersion is confirmed in both the ambient medium and the available DOF in resist. The DOF simulation results correlate closely with recent experimental work from other researchers.

Degradation in image contrast becomes a concern at higher numerical apertures (NAs) due to mask-induced polarization effects. We study how different photomask materials (binary and attenuated phase shift), feature sizes and shapes, pitch values, duty ratios (line to space), and wavelengths effect the polarization of transmitted radiation. Rigorous coupled-wave analysis (RCWA) is used to simulate the polarization of radiation by the photomask. The results show that higher NA leads to greater polarization effects in all cases. Off-axis illumination increases polarization in one of the first orders, decreasing it in the other. Nonvertical sidewall angles and rounded corners can also impact polarization, but the wavelength of incident radiation has no effect on polarization effects at the same NA values. In general, materials with higher refractive indices and lower extinction coefficients tend to pass more of the TM polarization state, whereas materials with lower refractive indices and a relatively wider range of extinction coefficients pass more TE polarized radiation. These properties can provide new design considerations for the development of next-generation masking materials.

We give a general introduction into polarized imaging and report on a Jones pupil approach for a complete evaluation of the resulting optical performance. The Jones pupil assigns a Jones matrix to each point of the exit pupil, describing the impact of both the global phase and the polarization on imaging. While we already can learn much about the optical system by taking a close look at the Jones pupil-and starting imaging simulations from it-a quantitative assessment is necessary for a complete evaluation of imaging. To do this, we generalize the concept of scalar Zernike aberrations to Jones-Zernike aberrations by expansion of the Jones pupil into vector polynomials. The resulting method is nonparaxial, i.e., the effect of the polarization-dependent contrast loss for high numerical apertures is included. The aberrations of the Jones matrix pupil are a suitable tool to identify the main drivers determining polarization performance. Furthermore, they enable us to compare the polarized and unpolarized performance of such a characterized lithographic system.

We present a novel, accurate methodology for investigating the kinetics of photoacid diffusion and catalyzed-deprotection of positive-tone chemically amplified resists during postexposure bake (PEB) by in-situ monitoring the change of resistance and capacitance (RC) of resist film during PEB. Deprotection converts the protecting group to a volatile group, which changes the dielectric constant of the resist. Thus, the deprotection rate can be extracted from the change of capacitance. The photoacid diffusivity is extracted from the resistance change because diffusivity determines the rate of change of the acid distribution. Furthermore, by comparing the RC curves, the dependence of acid diffusivity on the reaction state can be extracted. The experiment is done on Shipley UV5 resist with 248-nm wavelength illumination, an Agilent LCR meter is used to measure RC curves, and resist thickness loss is recorded. The kinetics of non-Fickean acid transportation, deprotection, free-volume generation and absorption/escaping, and resist shrinkage are analyzed. A comprehensive model of these chemical/physical mechanisms is proposed, whose parameters are extracted though a divide-and-conquer strategy with high accuracy. The theoretical RC curves fit the experiment with rms error less than 0.1%. The effects of free volume, polymer relaxation, and deprotection-induced polarity change on acid diffusion are also identified.

The ability of a critical dimension scanning electron microscope (CD-SEM) to resolve differences in the widths of two lines depends on the instrument's measurement repeatability and any sample-dependent biases. The dependence of repeatability and bias on eight different parameters is studied using the MONSEL Monte Carlo electron simulator. For each of 14,400 different combinations of values of eight parameters, three describing the sample and five describing characteristics of the instrument or measurement condition, an image is simulated, noise is added, and the edge positions are "measured" as would be done in a CD-SEM. From 100 repetitions of noise, the repeatability of such CD determinations is ascertained. Biases (i.e., average errors) are also determined. Noise amplitude, edge detection algorithm, and beam size are shown to be significant factors in measurement repeatability. The CD-SEM's measurement repeatability may be an order of magnitude better than its spatial resolution. For standard edge detection methods, the bias depends on the sample. This means that in a manufacturing environment in which the sample shape varies, there will be a random component of error that is not measured by the industry's usual same-sample tests of instrument precision.

Sidewall effects on critical dimensions (CD) of photomasks are studied. CDs of patterns are measured by a scanning electron microscope (SEM) and a deep-UV microscope. CDs of the patterns are also calculated by an aerial image simulation. The sidewall effects on the measured CDs are compared with effects on the calculated CD. The results indicate that the sidewall effect on CDs by SEM differs from the effect on the aerial image, and that the effect on CDs by a deep-UV microscope corresponds to the effect on the aerial image. The factor of the sidewall effect on an aerial image is studied using an electric field calculation simulator. The result reveals that the factor is transmissivity. SEMs have no information about transmission, and therefore the effects differ from each other. A deep-UV microscope employs transmitted light just like a wafer aligner, and therefore the effects correspond to each other.

Bubbles and droplets may be detrimental to the successful implementation of immersion lithography, depending in part on their lifetime in the system. In this work, a model is developed to estimate the dissolution times of nitrogen bubbles in pure water that may be free-floating or adhered to a solid surface. The model is then extended to small free and adhered droplets. Bubble dissolution and droplet evaporation times for typical immersion lithography conditions are presented.

Imaging interferometric lithography (IIL) is an optical resolution enhancement technique, based on wavelength-division multiplexing, that combines off-axis illumination with multiple exposures and pupil filtering. In prior experiments, IIL was shown to be capable of improving the resolution of optics to subwavelength regimes but only in isolated parameter settings. The frequency parsing scheme plays a critical role in the resolution of aerial images from IIL and a platform for optimizing these frequency coverage parameters is currently lacking. We present three approaches to the optimization of IIL frequency coverage parameters. First, the exhaustive search approach and, then, the dynamic programming and greedy versions are presented. For unobstructed lenses and large numerical apertures, these approaches converge to frequency coverages with illumination points along the vertical and horizontal axes, while a lens with an obscured center results in a tilted frequency coverage. Two particular frequency parsing strategies are analyzed from the perspective of the quadratic term in the aerial intensity. Simulation results comparing the off-axis exposures with and without the on-axis exposure are presented. The effect of the quadratic terms on the exposure latitude is shown to have a positive effect in the dense edge areas and a negative effect on isolated edges.

We perform experimental and theoretical studies of image placement error induced by aberrations of the projection lens. The goal is to experimentally determine the magnitude of the image placement errors, to compare experiments and simulations, and to screen possible correction strategies. The calculations and experiments are done for ArF lithography. Theoretically, we simulate image placement error using the projection lens aberration data and simulators such as Prolith or Solid-C. Features with low and high sensitivities to lens aberrations are identified, together with a reference feature that has low sensitivity for image placement error. Dedicated reticles are fabricated to print various features at different illumination conditions on the same substrate. The resulting patterns could be analyzed using top-down scanning electron microscopy (SEM), but also optically with the standard optical overlay tool KLA5200. For both techniques, the experimentally found image placement errors are in excellent agreement with simulations. In simulations, we calculate the dependency of the image placement error on pattern density, pattern orientation, and illumination conditions. These tendencies are experimentally reproduced. We conclude with a case study that demonstrates a possible correction strategy for image placement error.

For conventional wavelength (UV-vis-IR) lasers delivering radiation energy to the surface of materials, ablation thresholds, ablation (etch) rates, and the quality of ablated structures often differ dramatically between short (typically nanosecond) and ultrashort (typically femtosecond) pulses. Various short-wavelength (λ<100 nm) lasers emitting pulses with durations ranging from ~10 fs to ~1 ns have recently been put into routine operation. This makes it possible to investigate how ablation characteristics depend on pulse duration in the XUV spectral region. Four sources of intense short-wavelength radiation available in the authors' laboratories, including XUV and soft x-ray lasers, are used for the ablation experiments. Based on the results of the experiments, the etch rates for three different pulse durations are compared using the XUV-ABLATOR code to compensate for the wavelength difference. Comparing the values of etch rates calculated for nanosecond pulses with those measured for shorter pulses, we can study the influence of pulse duration on XUV ablation efficiency. The results of the experiments also show that the ablation rate increases while the wavelength decreases from the XUV spectral region toward x-rays, mainly due to increase of attenuation lengths at short wavelengths.

We present a new electro-thermal microactuator with multidirectional in-plane motion by selectively applying a single potential across two of four contacts. The design principle is based on the asymmetrical thermal expansion of the structure with 1. different lengths and widths of the beams, 2. varying resistivities of the beams by selectively doping, and 3. rigorous control of the thermal boundary conditions. Analytical models are derived to describe the electro-thermo-mechanical performances of the actuator. The commercial finite element package ANSYS is used to demonstrate the feasibility of the design, to verify the analytical results, and to characterize the actuator in details under complex heat transfer conditions. The design parameters that significantly affect the performance of the actuator are discussed, including the structural dimension, selective doping, and heat transfer condition. Conventional silicon-based micromachining techniques compatible with IC processes are used to fabricate the microactuator. The phosphorous-doped polycrystalline silicon film by low pressure chemical vapor deposition (LPCVD) is used to demonstrate the effectiveness of the microactuator. In conclusion, it is found that input voltages ≤7 V are required for the microactuator to achieve the maximum displacement in 8 µm and the maximum tip force in 8 µN with the operating temperature below 300 °C, the power as low as 10 mW, and the response time about 30 msec in average.

An accelerometer with concise structure having a resonant microbeam to measure 2-D acceleration is proposed. This structure is configured with a central proof mass suspended by four symmetrical and orthogonal high-aspect-ratio (HAR) microbeams. This dual-axis design is able to decouple a two-axis signal from a 2-D acceleration. An analytical model relating the linear relationship between the acceleration and the associated resonant frequency shift of microbeam is derived, and a finite element analysis (FEA) is also performed to confirm this model. The FEA result also shows that there is little cross talk between x and y directions of measurement, meaning that this structure is able to decouple a planar 2-D acceleration into two independent acceleration components, and therefore the 1-D analytical model can be used to evaluate the 2-D acceleration on the x-y plane. In addition, the model is verified by testing results of one conventional dual-axis natural frequency shifted microaccelerometer (DFSM). The simulation result also shows that the sensitivity of the proposed HAR accelerometer is triple that of a conventional DFSM.

A 2.45-GHz filter is fabricated by thin film bulk acoustic wave resonator (FBAR) technology. It is designed to minimize die size and simplify the fabrication process simultaneously by using inductive coupling plasma etching. The quality factor of a fabricated single resonator is 1567 and the electromechanical coupling coefficient is 5.7%. The fabricated filter has minimum insertion loss around 1.5 dB and maximum insertion loss at 3.6 dB in 83.5-MHz bandwidth, which is suitable for WLAN and Bluetooth applications.

Static thermal analyses for the earlier concept and the enhanced quasi-monolithic integration technology (QMIT) are performed in detail. The effects of several parameters such as the properties of the materials involved and different geometries in all possible structures are described. Simulation results confirm a very low thermal resistance for the enhanced QMIT structure and highlight its superiority to the earlier concept of QMIT structures. This leads to a longer lifetime, a higher reliability and a better performance of the packaging.

Ribbon structures on the submicrometer scale are of interest for the development of nanodevices in various fields. We fabricate periodic arrays of silicon ribbons using electron beam lithography and thin film deposition at highly oblique incident angles. A periodicity of 1 µm and a line width less than 100 nm is used for the lithographically prepatterned substrate seed layers to ensure that the planar fill factor was less than the equilibrium volume fill factor of the thin film. Individual ribbons exhibit a width of approximately 1.8 µm, controlled by the length of deposition, and a thickness of approximately 100 nm. The ribbons fabricated for this experiment have a length of 4 mm, and exhibit an amorphous structure with scattered crystallites throughout the matrix.