Development of compatible wet-clean stripper for integration of CoWP metal cap in Cu/low-k interconnects

Implementation of CoWP metal caps into Cu/low-k integration schemes requires a wet stripper that not only gives efficient cleaning but also has good compatibility to CoWP and low-k dielectrics. This paper describes a novel non-fluoride CoWP compatible stripper, developed based on a systematic study of the effect of stripper components, i.e. solvent, corrosion inhibitor, and stripper pH. Electrochemical methods were used to characterize galvanic corrosion of the CoWP/Cu couple and to estimate CoWP etch rate. Our studies showed that a traditional fluoride stripper caused severe damage to CoWP capping layer. The new stripper achieved a good balance between cleaning efficiency and compatibility to CoWP and low-k dielectrics, and demonstrated significant advantages in electrical properties over the traditional fluoride stripper.     

Efficient implementation of a frame aggregation unit for optical frame-based switching

Aggregating Ethernet frame or IP packet in large fixed-size frames allows for building scalable core network architectures. Classifying the arriving traffic based on destination core node information and quality of service parameters alleviates the need of performing table look-ups on packet basis. These advantages come at the cost of extra logic at the network egress, as regards implementation, and additional jitter due to the frame assembly process. This paper describes the efficient implementation of a frame aggregation unit that gathers Ethernet packets in G.709 containers, handles 10 Gb/s links, performs classification based on 24-byte headers, and includes a highly pipelined Queue Manager to cope with the considered rates while a specific scheduler controls the quality of service per core network flow. Based on the developed demonstrator, we provide results both as regards area and performance for an FPGA (field programmable gate array) Virtex-4 implementation as well as regarding the introduced jitter.     

Adaptive phase extraction: incorporating the Gabor transform in the matching pursuit algorithm

Short-time Fourier transform (STFT), Gabor transform (GT), wavelet transform (WT), and the Wigner-Ville distribution (WVD) are just some examples of time-frequency analysis methods which are frequently applied in biomedical signal analysis. However, all of these methods have their individual drawbacks. The STFT, GT, and WT have a time-frequency resolution that is determined by algorithm parameters and the WVD is contaminated by cross terms. In 1993, Mallat and Zhang introduced the matching pursuit (MP) algorithm that decomposes a signal into a sum of atoms and uses a cross-term free pseudo-WVD to generate a data-adaptive power distribution in the time-frequency space. Thus, it solved some of the problems of the GT and WT but lacks phase information that is crucial e.g., for synchronization analysis. We introduce a new time-frequency analysis method that combines the MP with a pseudo-GT. Therefore, the signal is decomposed into a set of Gabor atoms. Afterward, each atom is analyzed with a Gabor analysis, where the time-domain gaussian window of the analysis matches that of the specific atom envelope. A superposition of the single time-frequency planes gives the final result. This is the first time that a complete analysis of the complex time-frequency plane can be performed in a fully data-adaptive and frequency-selective manner. We demonstrate the capabilities of our approach on a simulation and on real-life magnetoencephalogram data.     

Compact Modules for Wireless Communication Systems in the E-Band (71–76 GHz)

The millimeter-wave spectrum above 70 GHz provides a cost-effective solution to increase the wireless communications data rates by increasing the carrier wave frequencies. We report on the development of two key components of a wireless transmission system, a high-speed photodiode (HS-PD) and a Schottky Barrier Diode (SBD). Both components operate uncooled, a key issue in the development of compact modules. On the transmitter side, an improved design of the HS-PD allows it to deliver an output RF power exceeding 0 dBm (1 mW). On the receiver side, we present the design process and achieved results on the development of a compact direct envelope detection receiver based on a quasi-optical SDB module. Different resonant (meander dipole) and broadband (Log-Spiral and Log-Periodic) planar antenna solutions are designed, matching the antenna and Schottky diode impedances at high frequency. Impedance matching at baseband is also provided by means of an impedance transition to a 50 Ohm output. From this comparison, we demonstrate the excellent performance of the broadband antennas over the entire E-band by setting up a short-range wireless link transmitting a 1 Gbps data signal.     

Morphology Transformation Pathway of Block Copolymer-Directed Cooperative Self-Assembly of ZnO Hybrid Films Monitored In Situ during Slot-Die Coating

Cooperative self-assembly (co-assembly) of diblock copolymers (DBCs) and inorganic precursors that takes inspiration from the rich phase separation behavior of DBCs can enable the realization of a broad spectrum of functional nanostructures with the desired sizes. In a DBC assisted sol–gel chemistry approach with polystyrene-block-poly(ethylene oxide) and ZnO, hybrid films are formed with slot-die coating. Pure DBC films are printed as control. In situ grazing-incidence small-angle X-ray scattering measurements are performed to investigate the self-assembly and co-assembly process during the film formation. Combining complementary ex situ characterizations, several distinct regimes are differentiated to describe the morphological transformations from the initially solvent-dispersed to the ultimately solidified films. The comparison of the assembly pathway evidences that the key step in the establishment of the pure DBC film is the coalescence of spherical micelles toward cylindrical domains. Due to the presence of the phase-selective precursor, the formation of cylindrical aggregates in the solution is crucial for the structural development of the hybrid film. The pre-existing cylinders in the ink impede the domain growth of the hybrid film during the subsequent drying process. The precursor reduces the degree of order, prevents crystallization of the PEO block, and introduces additional length scales in the hybrid films.     

Still Image Processing on Coarse-Grained Reconfigurable Array Architectures

Due to the increasing demands on efficiency, performance and flexibility reconfigurable computational architectures are very promising candidates in embedded systems design. Recently coarse-grained reconfigurable array architectures (CGRAs), such as the ADRES CGRA and its corresponding DRESC compiler are gaining more popularity due to several technological breakthroughs in this area. We investigate the mapping of two image processing algorithms, Wavelet encoding and decoding, and TIFF compression on this novel type of array architectures in a systematic way. The results of our experiments show that CGRAs based on ADRES and its DRESC compiler technology deliver improved performance levels for these two benchmark applications when compared to results obtained on a state-of-the-art commercial DSP platform, the c64x DSP from Texas Instruments. ADRES/DRESC can beat its performance by at least 50% in cycle count and the power consumption even drops to 10% of the published numbers of the c64x DSP.     

Measurement and Calculation of the Absolute Thermoelectric Power of Rhodium and Iridium

The thermoelectric power of Rh and Ir was redetermined between 100 K and 1400 K. It varies almost linearly from +1.7 μV K⁻¹ to −3.8 μV K⁻¹ for Rh and from +1.5 μV K⁻¹ to −2.2 μV K⁻¹ for Ir. The diffusive part of the thermopower could be calculated from the density of states. It is approximately equal to the temperature dependence of the electrochemical potential of the electrons divided by the electronic charge. This is attributed to the approximate establishment of local equilibrium between electrons and lattice atoms above 400 K—a condition not fulfilled in the phonon-drag regime below 300 K.     

Demonstration of the monolithic interconnection on CIS solar cells by picosecond laser structuring on 30 by 30 cm2 modules

In this paper, we present the selective structuring of all three patterns (P1, P2 and P3) of a monolithic interconnection of CIS (Cu(In,Ga)(S,Se)₂) thin film solar cells by picosecond laser pulses at a wavelength of 1064 nm. We show results for single pulse ablation threshold values and line scribing of molybdenum films on glass (P1), CIS on molybdenum (P2) and zinc oxide on CIS (P3). The purposes of these processes are the p‐type isolation (P1), cell interconnect (P2) and n‐type isolation (P3), which are required for complete cell architecture. The half micron thick molybdenum back electrode can be structured with a process speed of more than 15 m/s at about 15 W average power without detectable residues and damage by direct induced laser ablation from the back side (P1). The CIS layer can be structured selectively down to the molybdenum at process speeds up to 1 m/s at about 15 W average power, due to the precision of direct laser ablation in the ultrashort pulse regime (P2). The ZnO front electrode layer is separated by clean trenches with straight side walls at process speeds of up to 15 m/s at about 10 W average power, as a result of indirect induced laser ablation (P3). A validation of functionality of all processes is demonstrated on CIS solar cell modules (30 × 30 cm²). By replacing one state‐of‐the‐art process by a picosecond laser process at a time, solar efficiencies could be increased for P1 and P2 and stayed on a similar level for P3. After an optimization of the patterning processes in the R&D pilot line of AVANCIS, we achieved a new record efficiency for an all‐laser‐patterned CIS solar module: 14.7% as best value for the aperture area efficiency of a 30 × 30 cm² sized CIS module was reached. Copyright © 2014 John Wiley & Sons, Ltd.     

Low overhead assignment of symbolic coordinates in sensor networks

Approximate information on the location of nodes in a sensor network is essential to many types of sensor network applications and algorithms. In many cases, using symbolic coordinates is an attractive alternative to the use of geographic coordinates due to lower costs and lower requirements on the available location information during coordinate assignment. In this paper, we investigate different possible methods of assigning symbolic coordinates to sensor nodes. We present a method based on broadcasting coordinate messages and filtering using sensor events. We show in the evaluation that this method allows a reliable assignment of symbolic coordinates while only generating a low overhead.