This PDF file contains the editorial “How to Write a Good Scientific Paper: Style” for JM3 Vol. 15 Issue 03

This PDF file contains the editorial “Special Section Guest Editorial:Alternative Lithographic Technologies V” for JM3 Vol. 15 Issue 03

How surface geometries can be selectively manipulated through nanosphere lithography (NSL) is discussed. Self-assembled monolayers and multilayers of nanospheres have been studied for decades and have been applied to lithography for almost as long. When compared to the most modern, state-of-the-art techniques, NSL offers comparable feature resolution with many advantages over competing technologies. Several high-resolution alternatives require scan-based implementation (i.e., focused ion beams and e-beam lithography) while NSL is much more of a batch operation, allowing for full wafer or possibly even multiple wafer processing, potentially saving time and increasing throughput in a manufacturing environment. Additionally, NSL has continued to be of interest because it does not require expensive, complex equipment to be researched and realized, which continues to fuel interest in this approach. In spite of these advantages, applying NSL to specific, realizable devices is limited in the literature. The reason for this lack of application is not only unreliability in the self-assembly process, but also control of these patterned nanospheres within larger, multistep processes often required to fabricate most devices. Both of these items are addressed in this paper. The first issue was addressed through the development of a series of custom-designed nanosphere application vessels. These were designed based on the best published results from the literature, utilizing an alternate method of dip-coating but performed through draining the carrier fluid over the substrate rather than moving the substrate across the liquid–air boundary layer. This method is in the easier to perform, but arguably less-reliable spin-coating method also commonly employed. The key enabler in this effort lies in commercially available three-dimensional (3-D) printing technology, and how it was applied to rapidly prototype-improved deposition vessels. This was accomplished primarily with a single day turn-around between each 3-D printed design iteration. Each vessel design was incrementally improved, built, and tested to optimize the best performance in achieving the most reliable, repeatable self-aligned nanosphere layer formation. With an optimized design of this vessel in hand, the second challenge was addressed by using this vessel and the patterned nanosphere layers it produced with a patterned photoresist design to capture single layers of nanospheres in specifically designed locations and orientations. The hybrid mask produced from this approach can be integrated within virtually any multistep fabrication process. Additionally, other processing steps will be discussed, such as reactive ion etching (RIE), plasma ashing, and photoresist reflowing, and how they might be combined with these hybrid masks. Various results from combinations of these steps are presented. Finally, two potential applications which could benefit greatly from the resulting, engineered surface structures are discussed. These include a small-scale device application (engineering the contacting surfaces in a microswitch), as well as a much larger scale surface study application (surface engineering for controlling secondary electron emissions). The final results from this method allow for patterning groups of 500-nm polystyrene nanospheres formed in four to eight distinct rows each. These are positioned within patterned wells created in a 650-nm thick photoresist. The size and location of these wells are as precise as the photolithography process used to form them, in this case, ∼40-nm position error in the location of the edge of the laser using a Heidelberg laser lithography system. By combining multiple wells in close proximity, virtually any combination of nanosphere clusters become possible. Once patterned, postprocessing though RIE and deposition method selection together determine the final shape of the nanoscale features which result.

Directed self-assembly (DSA) of block copolymers (BCP) is considered a promising patterning approach for the 7-nm node and beyond. Specifically, a graphoepitaxy process using a cylindrical phase BCP may offer an efficient solution for patterning randomly distributed contact holes with subresolution pitches, such as found in via and cut mask levels. In any graphoepitaxy process, the pattern density impacts the template fill (local BCP thickness inside the template) and may cause defects due to over- or underfilling of the template. In order to tackle this issue thoroughly, the parameters that determine template fill and the influence of template fill on the resulting pattern should be investigated. Using three process flow variations (with different template surface energy), template fill is experimentally characterized as a function of pattern density and film thickness. The impact of these parameters on template fill is highly dependent on the process flow, and thus prepattern surface energy. Template fill has a considerable effect on the pattern transfer of the DSA contact holes into the underlying layer. Higher fill levels give rise to smaller contact holes and worse critical dimension uniformity. These results are important for DSA-aware design and show that fill is a crucial parameter in graphoepitaxy DSA.

Overcoming the optical limitations of 193-nm immersion lithography can be achieved using directed self-assembly (DSA) of block-copolymers (BCPs) as a low-cost and versatile complementary technique. The goal of this paper is to investigate the potential of DSA to address line and space (L/S) high-resolution patterning by performing the density multiplication of lines with the graphoepitaxy approach. As surface affinity is a key parameter in self-assembly, three variations, or “flavors,” of DSA template affinity are investigated regarding several success criteria such as morphology control or defectivity. More precisely, both the methodology to register DSA defects and the impact of process parameters on defectivity are detailed. Using the 300-mm pilot line available in LETI and Arkema’s advanced materials, we investigate process optimization of DSA line/space patterning of a 38-nm period lamellar PS-b-PMMA BCP (L38). Our integration scheme is based on BCP self-assembly inside organic hard mask guiding patterns obtained using 193i-nm lithography. Defect analysis coupled with the fine tuning of process parameters (annealing, brush material) provided the optimum conditions for the L38 self-assembly. Using such conditions, DSA using the three affinity flavors is investigated by means of SEM top-view and cross-section review. Lithographic performances of one selected flavor are then evaluated with the comparison of process windows function of either commensurability, morphology, or roughness. This work is meant as a guideline for the graphoepitaxy optimization of materials and process parameters on a 300-mm platform.

For directed self-assembly (DSA) to be deployed in advanced semiconductor technologies, it must reliably integrate into a full process flow. We present a methodology for using virtual fabrication software, including predictive DSA process models, to develop and analyze the replacement of self-aligned quadruple patterning with Liu–Nealey chemoepitaxy on a 14-nm dynamic random access memory (DRAM) process. To quantify the impact of this module replacement, we investigated a key process yield metric for DRAM, interface area between the capacitor contacts and transistor source/drain. Additionally, we demonstrate virtual fabrication of the DRAM cell’s hexagonally packed capacitors patterned with an array of diblock copolymer cylinders in place of fourfold litho-etch (LE⁴) patterning.

The high throughput of character projection (CP) electron-beam (EB) lithography makes it a promising technique for low-to-medium volume device fabrication with regularly arranged layouts, such as for standard-cell logics and memory arrays. However, non-VLSI applications such as MEMS and MOEMS may not be able to fully utilize the benefits of the CP method due to the wide variety of layout figures including curved and oblique edges. In addition, the stepwise shapes that appear because of the EB exposure process often result in intolerable edge roughness, which degrades device performances. In this study, we propose a general EB lithography methodology for such applications utilizing a combination of the CP and variable-shaped beam methods. In the process of layout data conversion with CP character instantiation, several control parameters were optimized to minimize the shot count, improve the edge quality, and enhance the overall device performance. We have demonstrated EB shot reduction and edge-quality improvement with our methodology by using a leading-edge EB exposure tool, ADVANTEST F7000S-VD02, and a high-resolution hydrogen silsesquioxane resist. Atomic force microscope observations were used to analyze the resist edge profiles’ quality to determine the influence of the control parameters used in the data conversion process.

We describe a lithographic approach—nanocrystal plasma polymerization-based lithography—in which colloidal nanocrystal assemblies (CNAs) are used as the inorganic resist and, potentially, the active material. The patterning process is based on a change in the dispersibility of the CNAs in solvents as a result of the exposure to plasmas. Plasmas can etch the capping ligands from the exposed area. During the development step, the unexposed area of CNAs is redispersed, leaving behind the patterned area, similar to what is expected from negative photoresist.

This paper demonstrates an approach for fabricating large-scale photonic crystal (PhC)-based devices using a combination of optical and focused ion beam (FIB) lithography techniques. Optical lithography along with reactive ion etching parameters is optimized to realize the layout of device structure and thereafter FIB milling is optimized to realize the designed PhC structure at those identified locations. At first, with the help of a specially designed mask and using optical lithography along with reactive ion etching, a number of rectangular areas of dimension of 10  μm×20  μm along with input and output waveguides of width ∼700  nm and thickness of ∼250  nm have been fabricated. Subsequently, use of FIB milling, a periodic PhC structure of lattice constant of 600 nm, having a hole diameter of ∼480  nm along with a defect hole diameter of ∼250  nm have been realized successfully on the selected areas. This method shows a promising application in fabricating PhC structure with device size <1  cm² at large scale, eliminating the problems of standard nanolithography techniques.

We propose and discuss a modular technology to reduce the edge-placement-error effect by combining selective etching and alternating-material (dual-material) self-aligned multiple patterning (altSAMP) processes. A geometrical cut-process yield model considering the joint effect of overlay errors, cut-hole, and line CD variations is developed to analyze its patterning performance. In addition to the contributions from the above three process variations, the impacts of key control parameters (such as cut-hole overhang and etching selectivity) on the patterning yield are examined. It is shown that the optimized altSAMP patterning process significantly improves the patterning yield compared with conventional SAMP processes, especially when the half pitch of device patterns is driven down to 7 nm and below. Moreover, the corresponding layout decomposition and synthesis strategy are also discussed. The experimental results of altSAMP process development and material screening are reported.

We present a directed self-assembly (DSA) compliant flow for contact/via layers with immersion lithography assuming the graphoepitaxy process for the cylinders’ formation. We demonstrate that the DSA technology enablement needs co-optimization among material, design, and lithography. We show that the number of DSA grouping constructs is countable for the gridded-design architecture. We use template error enhancement factor to choose DSA material, determine grouping design rules, and select the optimum guiding patterns. Our post-pxOPC imaging data show that it is promising to achieve two-mask solution with DSA for the contact/via layer using 193i at 5 nm node.

Nanoscale ridge apertures have been demonstrated to be applied for high-resolution lithography. We performed a numerical study of nanoscale bowtie apertures with different outline dimensions and gap sizes to analyze their detailed field distribution for near-field scanning optical lithography (NSOL). It is found that the high image contrast, which is necessary for good quality lithography, is obtained in the near-field region and decays quickly with increasing distance. Furthermore, a smaller gap size achieves higher image contrast and deeper depth of focus. With the NSOL system, static and scanning lithography experiments are conducted. Combined with the passive flexure stage for contact control, we achieved 18-nm lithography resolution.

The SHARP high-numerical aperture actinic reticle review project is a synchrotron-based, extreme ultraviolet (EUV) microscope dedicated to photomask research. SHARP emulates the illumination and imaging conditions of current EUV lithography scanners and those several generations into the future. An anamorphic imaging optic with increased mask-side numerical aperture (NA) in the horizontal and increased demagnification in the vertical direction has been proposed to overcome limitations of current multilayer coatings and extend EUV lithography beyond 0.33 NA. Zoneplate lenses with an anamorphic 4×/8× NA of 0.55 are fabricated and installed in the SHARP microscope to emulate anamorphic imaging. SHARP’s Fourier synthesis illuminator with a range of angles exceeding the collected solid angle of the newly designed elliptical zoneplates can produce arbitrary angular source spectra matched to anamorphic imaging. A target with anamorphic dense features down to 50-nm critical dimension is fabricated using 40 nm of nickel as the absorber. In a demonstration experiment, anamorphic imaging at 0.55 4×/8× NA and 6 deg central ray angle (CRA) is compared with conventional imaging at 0.5 4× NA and 8 deg CRA. A significant contrast loss in horizontal features is observed in the conventional images. The anamorphic images show the same image quality in the horizontal and vertical directions.

Extreme ultraviolet lithography (EUVL, λ=13.5  nm) is the most promising candidate to manufacture electronic devices for future technology nodes in the semiconductor industry. Nonetheless, EUVL still faces many technological challenges as it moves toward high-volume manufacturing (HVM). A key bottleneck from the tool design and performance point of view has been the development of an efficient, high-power EUV light source for high throughput production. Consequently, there has been extensive research on different methodologies to enhance EUV resist sensitivity. Resist performance is measured in terms of its ultimate printing resolution, line width roughness (LWR), sensitivity [S or best energy (BE)], and exposure latitude (EL). However, there are well-known fundamental trade-off relationships (line width roughness, resolution and sensitivity trade-off) among these parameters for chemically amplified resists (CARs). We present early proof-of-principle results for a multiexposure lithography process that has the potential for high sensitivity enhancement without compromising other important performance characteristics by the use of a “Photosensitized Chemically Amplified Resist™” (PSCAR™). With this method, we seek to increase the sensitivity by combining a first EUV pattern exposure with a second UV-flood exposure (λ=365  nm) and the use of a PSCAR. In addition, we have evaluated over 50 different state-of-the-art EUV CARs. Among these, we have identified several promising candidates that simultaneously meet sensitivity, LWR, and EL high-performance requirements with the aim of resolving line space (L/S) features for the 7- and 5-nm logic node [16- and 13-nm half-pitch (HP), respectively] for HVM. Several CARs were additionally found to be well resolved down to 12- and 11-nm HP with minimal pattern collapse and bridging, a remarkable feat for CARs. Finally, the performance of two negative tone state-of-the-art alternative resist platforms previously investigated was compared to the CAR performance at and below 16-nm HP resolution, demonstrating the need for alternative resist solutions at 13-nm resolution and below. EUV interference lithography (IL) has provided and continues to provide a simple yet powerful platform for academic and industrial research, enabling the characterization and development of resist materials before commercial EUV exposure tools become available. Our experiments have been performed at the EUV-IL set-up in the Swiss Light Source (SLS) synchrotron facility located at the Paul Scherrer Institute (PSI).

A photosensitive polyimide system based on amine catalyzed imidization of a precursor poly(amic ester) is described. The material is based on the meta ethyl ester of pyromellitic dianhydride and 2,2’ bis(trifluoromethyl)benzidine. It acts as a negative tone resist when formulated with a photobase generator. The material exhibits a dielectric constant of 3.0 in the gigahertz range, a coefficient of thermal expansion of 6±2  ppm/K, and can be patterned to aspect ratios of >2 when formulated with a highly quantum efficient cinnamide type photobase generator.

A high numerical aperture (NA) system with an NA larger than 0.5 is required to make patterns of 1X nm and below, even though extreme ultraviolet lithography uses a 13.5-nm wavelength source. To avoid the reflective efficiency loss and to avoid an increase in the chief ray angle of incident light, use of an anamorphic high-NA system is suggested. The suggested anamorphic NA system has nonisotropic magnification, x-magnification of 4× and y-magnification of 8×, and the mask NA shape is an ellipse due to the nonisotropic magnification distribution. Anamorphic NA systems have a nonconventional shadow effect due to nonisotropic incident angle distribution and magnification. These nonisotropic characteristics lead to the reduction of asymmetric shadow distribution and a reduction of horizontal–vertical bias. As a result, anamorphic NA systems can achieve balanced patterning results regardless of pattern direction and incident direction.

Extreme ultraviolet interference lithography records the interference pattern of two diffracted, coherent light beams, where the pattern resolution is half the diffraction grating resolution. The fabrication of diffraction grating masks by e-beam lithography is restricted by the electron proximity effect and pattern transfer limitations into diffraction efficient materials. By patterning HSQ lines at a relaxed pitch to avoid the electron proximity effect, depositing conformal iridium via atomic layer deposition, followed by ion milling the top and bottom iridium and HSQ removal, we fabricated iridium diffraction gratings at double the line spacing of the original HSQ lines. Line/space patterns of 6-nm half-pitch patterns were achieved using these masks, marking a new record resolution in photolithography.

The dynamic absorption coefficients of several chemically amplified resists (CAR) and non-CAR extreme ultraviolet (EUV) photoresists are measured experimentally using a specifically developed setup in transmission mode at the x-ray interference lithography beamline of the Swiss Light Source. The absorption coefficient α and the Dill parameters ABC were measured with unprecedented accuracy. In general, the α of resists match very closely with the theoretical value calculated from elemental densities and absorption coefficients, whereas exceptions are observed. In addition, through the direct measurements of the absorption coefficients and dose-to-clear values, we introduce a new figure of merit called chemical sensitivity to account for all the postabsorption chemical reaction ongoing in the resist, which also predicts a quantitative clearing volume and clearing radius, due to the photon absorption in the resist. These parameters may help provide deeper insight into the underlying mechanisms of the EUV concepts of clearing volume and clearing radius, which are then defined and quantitatively calculated.

The absorbed energy density (eV/cm³) deposited by extreme ultraviolet (EUV) photons and electron beam (EB) high-keV electrons is proposed as a metric for characterizing the sensitivity of EUV resist films. Simulations of energy deposition are used to calculate the energy density as a function of the incident aerial flux (EUV: mJ/cm², EB: μC/cm²). Monte Carlo calculations for electron exposure are utilized, and a Lambert–Beer model for EUV absorption. The ratio of electron flux to photon flux which results in equivalent energy density is calculated for a typical organic chemically amplified resist film and a typical inorganic metal-oxide film. This ratio can be used to screen EUV resist materials with EB measurements and accelerate advances in EUV resist systems.

We compare the speed and effectiveness of two genetic optimization algorithms to the results of statistical sampling via a Markov chain Monte Carlo algorithm to find which is the most robust method for determining real-space structure in periodic gratings measured using critical dimension small-angle x-ray scattering. Both a covariance matrix adaptation evolutionary strategy and differential evolution algorithm are implemented and compared using various objective functions. The algorithms and objective functions are used to minimize differences between diffraction simulations and measured diffraction data. These simulations are parameterized with an electron density model known to roughly correspond to the real-space structure of our nanogratings. The study shows that for x-ray scattering data, the covariance matrix adaptation coupled with a mean-absolute error log objective function is the most efficient combination of algorithm and goodness of fit criterion for finding structures with little foreknowledge about the underlying fine scale structure features of the nanograting.

The model accuracy of optical proximity-effect correction (OPC) was investigated by two modeling methods for a 10-nm node process. The first method is to use contours of two-dimensional structures extracted from critical dimension-scanning electron microscope (CD-SEM) images combined with conventional CDs of one-dimensional structures. The accuracy of this hybrid OPC model was compared with that of a conventional OPC model, which was created with only CD data, in terms of root-mean-square (RMS) error for metal and contact layers of 10-nm node logic devices. Results showed improvement of model accuracy with the use of hybrid OPC modeling by 23% for contact layer and 18% for metal layer, respectively. The second method is to apply a correction technique for resist shrinkage caused by CD-SEM measurement to extracted contours for improving OPC model accuracy. The accuracy of OPC model with shrink correction was compared with that without shrink correction, and total RMS error was decreased by 12% by using the shrink correction technique. It can be concluded that the use of CD-SEM contours and the shrink correction of contours are effective to improve the accuracy of OPC model for the 10-nm node process.

We present a comprehensive comparative study of the roughness metrics of different resists. Dense line/space of polymethyl methacrylate, hydrogen silsesquioxane, a metal oxide-based resist, and different chemically amplified resists (CARs) have been patterned by extreme ultraviolet interference lithography. All three line width roughness (LWR) metrics: the root-mean-square (r.m.s.) roughness value σ_LWR, the correlation length ξ, and the roughness exponent α, were extracted by metrological analysis of top-down SEM images. We found that all metrics are required to fully describe the overall roughness of each resist. Our measurements indicate that in fact, a few of the state-of-the-art resists tested here can meet the International Technology Roadmap for Semiconductors requirements for σ_LWR. The correlation length ξ was also found to be considerably higher in polymer-based materials in comparison to nonpolymers. Finally, the roughness exponent α, interpreted using the concept of fractal geometry, was found to be mainly affected by acid diffusion in CARs, where it produces line edges with a higher complexity than in non-CAR resists. These results indicate that the different resists platforms show very different LWR metrics and roughness is not manifested only in the σ_LWR but in all parameters. Therefore, all roughness metrics should be taken into account when comparing the performance among different resists since they ultimately have a substantial impact on device performance.

Java Monte Carlo Simulator for Secondary Electrons (JMONSEL) simulations are used to generate expected imaging responses of chosen test cases of patterns and defects with the ability to vary parameters for beam energy, spot size, pixel size, and/or defect material and form factor. The patterns are representative of the design rules for an aggressively scaled FinFET-type design. With these simulated images and resulting shot noise, a signal-to-noise framework is developed, which relates to defect detection probabilities. Additionally, with this infrastructure, the effect of detection chain noise and frequency-dependent system response can be made, allowing for targeting of best recipe parameters for multielectron beam inspection validation experiments. Ultimately, these results should lead to insights into how such parameters will impact tool design, including necessary doses for defect detection and estimations of scanning speeds for achieving high throughput for high-volume manufacturing.

The decreasing size of semiconductor features and the increasing structural complexity of advanced devices have placed continuously greater demands on manufacturing metrology, arising both from the measurement challenges of smaller feature sizes and the growing requirement to characterize structures in more than just a single critical dimension. For scanning electron microscopy, this has resulted in increasing sophistication of imaging models. For critical dimension atomic force microscopes (CD-AFMs), this has resulted in the need for smaller and more complex tips. Carbon nanotube (CNT) tips have thus been the focus of much interest and effort by a number of researchers. However, there have been significant issues surrounding both the manufacture and use of CNT tips. Specifically, the growth or attachment of CNTs to AFM cantilevers has been a challenge to the fabrication of CNT tips, and the flexibility and resultant bending artifacts have presented challenges to using CNT tips. The Korea Research Institute for Standards and Science (KRISS) has invested considerable effort in the controlled fabrication of CNT tips and is collaborating with the National Institute of Standards and Technology on the application of CNT tips for CD-AFM. Progress by KRISS on the precise control of CNT orientation, length, and end modification, using manipulation and focused ion beam processes, has allowed us to implement ball-capped CNT tips and bent CNT tips for CD-AFM. Using two different generations of CD-AFM instruments, we have evaluated these tip types by imaging a line/space grating and a programmed line edge roughness specimen. We concluded that these CNTs are capable of scanning the profiles of these structures, including re-entrant sidewalls, but there remain important challenges to address. These challenges include tighter control of tip geometry and careful optimization of scan parameters and algorithms for using CNT tips.

For the successful implementation of extreme ultraviolet (EUV) lithography in the upcoming technology nodes, a major challenge to overcome is the stable and reliable detection and characterization of mask defects. We have recently presented a reflective mode EUV mask scanning lensless imaging tool (RESCAN) which was installed at the XIL-II beamline of the swiss light source and showed reconstructed aerial images of test patterns on EUV masks. RESCAN uses scanning coherent diffractive imaging (SCDI) methods to obtain actinic aerial images of EUV photomasks and was designed for 80 nm onmask resolution. Our SCDI algorithm reconstructs the measured sample by iteratively solving the phase problem using overdetermined diffraction data gathered by scanning across the specimen with a finite illumination. It provides the phase and amplitude aerial images of EUV photomasks with high resolution without the need to use high numerical aperture (NA) lenses. Contrary to scanning microscopy and full-field microscopy, where the resolution is limited by the spot size or NA of the lens, the achievable resolution with our method depends on the detector noise and NA of the detector. To increase the resolution of our tool, we upgraded RESCAN with a detector and algorithms. Here, we present the results obtained with the tool that is capable of up to 40-nm onmask resolution. We believe that the realization of our prototype marks a significant step toward overcoming the limitations imposed by methods relying on imaging optics and shows a viable solution for actinic mask metrology.

A simple integration process of single-layer nanowire gratings (SLNGs) for polarization navigation sensor is proposed. The adaptive capacity of the nanoimprint process on uneven substrate is realized using compressed air and the flexible mold. The plasma ashing process is used to simplify the conventional SLNGs fabrication process. Based on the work mentioned above, six SLNGs areas with 1.3  mm×1.3  mm and six different directions are successfully integrated into a polarization sensitive structure applied to a polarization navigation sensor. The measurement result of quantum efficiency shows good sensitivity, and the angle of polarization measurement error is <±0.1  deg. The experiment result shows that the proposed process is applicable for the integration of large-area single-layer nanometal patterns.

A unified fabrication process was developed for four complex gratings. One period of each grating contains multiple pairs of ridges/grooves, and a ridge is usually composed of diversified materials. The process was suitable not only for mass fabrication but also for a large area. Moreover, with only one mold, four gratings could be obtained at once. Two 12×12  mm² samples were fabricated for demonstration. Profile uniformity of each sample was confirmed with broadband infrared (2.5 to 25  μm) transmittance spectra. A good agreement between measured and modeling results was observed.

A blue-ray laser lithography along with a dry etching process was presented to fabricate a three-dimensional (3-D) silicon structure with smooth surface. Main findings include the reduction of corrugations on photo resist profiles by adjusting the focal depth of grayscale laser lithography, optimization of reactive ion etching process to decrease the surface roughness after dry etching as well as keeping a low etching selectivity, and the introduction of a procedure to design and fabricate arbitrary 3-D silicon structures. Silicon microstructures with a curved surface were fabricated successfully applying this approach and the root-mean-square surface roughness was down to 3.0 nm. This result reduces approximately one order of magnitude than the previously reported surface roughness of arbitrary 3-D silicon microstructures. This approach is promising for fabricating arbitrary shape components in MEMS, optics, and microfluidic applications.

This paper demonstrates that another class of three-dimensional integrated circuits (3-D-ICs) exists, distinct from through-silicon-via-centric and monolithic 3-D-ICs. Furthermore, it is possible to create devices that are 3-D “at the device level” (i.e., with active channels oriented in each of the three coordinate axes), by performing standard CMOS fabrication operations at an angle with respect to the wafer surface into high aspect ratio silicon substrates using membrane projection lithography (MPL). MPL requires only minimal fixturing changes to standard CMOS equipment, and no change to current state-of-the-art lithography. Eliminating the constraint of two-dimensional planar device architecture enables a wide range of interconnect topologies which could help reduce interconnect resistance/capacitance, and potentially improve performance.

Waveguide (WG) photonic-bridge taper modules are designed for symmetric planar coupling between silicon WGs and single-mode fibers (SMFs) to minimize photonic chip and packaging footprint requirements with improving broadband functionality. Micromachined fabrication and evaluation of polymer WG tapers utilizing high-resolution focused ion beam (FIB) milling is performed and presented. Polymer etch rates utilizing the FIB and optimal methods for milling polymer tapers are identified for three-dimensional patterning. Polymer WG tapers with low sidewall roughness are manufactured utilizing FIB milling and optically tested for fabrication loss. FIB platforms utilize a focused beam of ions (Ga⁺) to etch submicron patterns into substrates. Fabricating low-loss polymer WG taper prototypes with the FIB before moving on to mass-production techniques provides theoretical understanding of the polymer taper and its feasibility for connectorization devices between silicon WGs and SMFs.

While automatic hot embossing systems are available for large- and small-scale productions of polymeric devices, one of the process challenges remains to be the manufacturing of precise, durable, and yet inexpensive hot embossing stamps. The use of metallic stamps manufactured by electroplating a photoresist pattern or by precision milling and their replication into silicone molds with UV-lithography, electroplating, and molding techniques is state of the art. Yet, there have been few, if any, thriving attempts to directly emboss polymers by means of bare photoresists, and in particular polyimide-based photoresists, without transferring the photoresist patterns into a different stamp material. We conduct a proof-of-concept by developing hot embossing stamps based on photosensitive polyimide. We focus primarily on the reliability of the aforementioned stamps throughout the hot embossing cycle and the fidelity of pattern transfer onto polymeric films for different microstructural patterns.

The use of optical tweezers (OTs) and spin angular momentum transfer to birefringent particles allows new mechanical measurements in systems where torque and rotation are relevant parameters at the single-molecule level. There is a growing interest in developing simple, fast, and inexpensive protocols to produce a large number of submicron scale cylinders of quartz, a positive uniaxial birefringent crystal, to be employed for such angular measurements in OTs. Here, we show that laser interference lithography, a method well known for its simplicity, fulfills these requirements and produces quartz cylindrical particles that we successfully use to apply and measure optical torque in the piconewton nm range in an optical torque wrench.

A dual-purpose nonresonant 2-degrees of freedom (DOF) drive-mode and 1-DOF sense-mode vibratory gyro/accelerometer fabricated using the economical ultraviolet-lithographie-galvanoformung-abformung (UV-LIGA) fabrication process using SU-8 photoresist is reported. The dual-purpose device presented is capable of detecting acceleration at the lower-frequency band and angular rate at the operating frequency band thereby functioning as both accelerometer and gyroscope. This is achieved by designing the structure such that the frequency response of the drive oscillator has two drive resonances with a flat zone between them, while the sense oscillator has one resonance, which is deliberately placed in the flat region between the two drive resonances. For angular rate detection, the device is operated in the flat zone at the sense resonance frequency at which the device is less susceptible to frequency variations due to both environmental variation and fabrication imperfections and hence is said to be operating in robust mode. The steady-state response and discrimination for angular rate and acceleration sensing have been devised using analytical modeling. The fabrication process is optimized to realize a gyro/accelerometer that has a 9-μm-thick nickel structural layer and 4-μm capacitive gaps. The overall miniature device size is 2.0  mm×1.9  mm. The experimental frequency response of the fabricated devices shows drive-mode resonances at 2.85 and 4.96 kHz and sense resonance at 3.85 kHz compared to the respective design values of drive-mode resonance frequencies 2.97 and 4.81 kHz and sense resonance frequency of 4 kHz. To demonstrate the dual-purpose capability of the device, acceleration characterization has been carried out and presented. The fabricated sensor is packaged in a ceramic package and interfaced with a MS3110 differential capacitive read out IC to characterize the acceleration response of the sensor, using an out-of-plane shaker. The bandwidth for acceleration detection is found to be 10 to 100 Hz and the sensitivity of the sensor at 30 Hz is found to be 57  mV/g.

Microspray generation by microfluidic nozzles comprising piezoelectric actuators featuring honeycombed grooves for antistiction was investigated. Microfluidic nozzles with 15-μm diameter were fabricated by the electroforming of nickel (Ni) with an estimated total deposition thickness of 40  μm. These nozzles were formed on the Ni plate in arrays of 31, 61, and 151; all nozzles were surrounded by a network array of honeycombed grooves with a line width of 30  μm. The piezoelectric actuators were bonded to the nozzle plates and filled with water to generate microsprays. The actuators were driven at electrical voltages of 40 to 142 V at 94 to 103 kHz, in which spray jets with volumetric rates of 4 to 65  ml/h were simply achieved and analyzed, agreeing with the proposed theory in the study. The use of particle image velocimetry (1500 fps) revealed that the turbulent flow of droplets from the sprays created strong recirculating vortices for a short time (0 to 12 ms). In addition, we experimentally demonstrated turbulence of droplets created at an average speed of 2.9±0.3  m/s, illustrating instable vortex-like motion. Hence the influence of turbulence on a variety of applications such as printing and cooling should be further concerned and investigated in the future.

The design, fabrication, and experimental test of an awl-shaped serpentine microspring (ASSM) for providing in-plane motion and low spring constant-to-layout area are investigated. The ASSM can provide low stiffness for in-plane motion under a restricted layout area. The spring constants of ASSMs for in-plane motion are theoretically analyzed by using Castigliano’s theorem, and validated through simulation using COMSOL Multiphysics. These ASSMs are successfully fabricated on a silicon-on-insulator wafer. The parameter K/A (spring constant-to-layout area) is used as the performance index. Smaller K/A induces larger deformation under the same layout area. It shows that K/A of ASSM is smaller than the classic one with the same dimensions (total length and number of turns). Geometric sizes having effect on K/A are discussed. The spring constants for in-plane and out-of-plane motion are compared and discussed. The taper angle and beam width-to-thickness ratio (w/h) are two key factors. As w/h being <1, the spring constant of in-plane motion (k_y) is always larger than that of out-of-plane motion (k_z). If w/h is <1, the spring constant k_z would be larger than k_y if the taper angle are larger than its critical angle φr.

We present the design and development of breath-mode silicon ring resonators fabricated using a commercial pure-play microfabrication process that provides ultraclean wafer-level vacuum-encapsulation. The micromechanical resonators are fabricated in MEMS integrated design for inertial sensors process that is developed by Teledyne DALSA Semiconductor Inc. The ring resonators are designed to operate with a relatively low DC polarization voltage, starting at 5 V, while providing a high frequency-quality factor product. We study the quality of the vacuum packaging using an automated testing setup over an extended time period. We study the effect of motional resistance on the performance of MEMS resonators. The fabricated devices had a resonant frequency of 10 MHz with the quality factor exceeding 8.4×10⁴.

We first investigate the effect of oxygen-doping concentration on resistive switching (RS) behaviors in SiC_xO_y, which were prepared by a radio frequency magnetron sputtering at the oxygen partial pressure ranging from 0% to 6%. Bipolar RS behaviors were achieved in all the fabricated devices and all these devices are valence change memories. With the oxygen partial pressure increasing from 0% to 6% (sample-0% has 40 at. % of oxygen), the mean R_HRS increases from 4.5 to 64.8  MΩ and then decreases to 1.5  MΩ, indicating that the device exhibits the largest ON/OFF ratio ∼500 at the oxygen partial pressure of about 2%. Based on the analyses of x-ray photoelectron spectroscopy, fitting current–voltage curves, and resistance–temperature measurements, it is clear that the trap filled limit space charge limited current and a Schottky barrier in the interface of the SiC_xO_y film and p+-Si are suggested to be dominant in the positive and negative biases, respectively. Most importantly, all devices can keep the data more than 10⁴  s and endure more than 10² continuous cycles, thus confirming the nonvolatile properties.

This paper studies the effect of microlens front-face angle on the performance of an optical system consisting of a planar-graded refractive index (GRIN) lens pair facing each other separated by a free-space region. The planar silica microlens pairs are designed to facilitate low-loss optical signal propagation in the free-space region between the opposing optical waveguides. The planar lens is fabricated from a 38-μm-thick fluorine-doped silica layer on a silicon substrate. It has a parabolic refractive index profile in the vertical direction, which is achieved by controlled fluorine incorporation in the silica film to collimate the optical beam in the vertical direction. Horizontal beam collimation is achieved by incorporating a horizontal curvature at the front face of the lens defined by deep oxide etch. A generalized 3×3ABCDGH transformation matrix method has been derived to compute the coupling efficiency of such microlens pairs to take front-face angles that may be present due to fabrication variations or limitations and possible input/output optical fiber offset/tilt into considerations. Pairs of such planar GRIN lens with various free-space propagation distances between them ranging from 75 to 2500 μm and with front-face angles of 1.5 deg, 2 deg, and 4 deg have been fabricated and characterized. Beam propagation method simulations have been carried out to substantiate the theoretical and experimental results. The results indicate that the optical loss is reasonably low up to 1.5 deg of front-face angles and increases significantly with further increase in the front-face angle. Analysis shows that for a given system with specific microlens front-face angle, the optical loss can be significantly reduced by properly compensating the vertical position of the input and output fibers.

Micromirror based spatial light modulators (SLMs) developed by the Fraunhofer Institute for Photonic Microsystems are well established in microlithography applications. Serving, e.g., as reflective, programmable photomasks in deep-UV mask writers, they enable highly flexible pattern generation. During operation, the micromirror bow significantly impacts contrast and the resolvable feature size of generated patterns. In some situations, MEMS micromirrors tend to change their bow during laser irradiation. A test regime including a characterization unit for the in situ analysis of MEMS micromirror topology has been developed to measure the bow change under various irradiation conditions. Experiments in which SLMs were irradiated by a 1-kHz, 248-nm pulse laser revealed that mirror bowing can occur in both directions (concave and convex). The bowing direction is dependent upon the applied irradiation parameters such as pulse-energy density, pulse number, and the deposited energy. Sustained irradiation at energy densities exceeding a certain limit can potentially become a limiting factor for the resolvable feature sizes of the patterns generated and, therefore, for the usable SLM lifespan.