Microstructure characterization of Sn-Ag solder joints between stud bumps and metal pads

The microstructure of the flip-chip solder joints fabricated using stud bumps and Pb-free solder was characterized. The Au or Cu stud bumps formed on Al pads on Si die were aligned to corresponding metal pads in the substrate, which was printed with Sn-3.5Ag paste. Joints were fabricated by reflowing the solder paste. In the solder joints fabricated using Au stud bumps, Au-Sn intermetallics spread over the whole joints, and the solder remained randomly island-shaped. The δ-AuSn, ε-AuSn₂, and η-AuSn₄ intermetallic compounds formed sequentially from the Au stud bump. The microstructure of the solder joints did not change significantly even after multiple reflows. The AuSn₄ was the main phase after reflow because of the fast dissolution of Au. In the solder joints fabricated using Cu stud bumps, the scallop-type Cu₆Sn₅ intermetallic was formed only at the Cu interface, and the solder was the main phase. The difference in the microstructure of the solder joints with Au and Cu stud bumps resulted from the dissolution-rate difference of Au and Cu into the solder.     

Wetting interaction between Sn-Zn-Ag solders and Cu

The wetting interaction of Sn-(7.1–9)Zn-(0–3)Ag solders with Cu was investigated from 230°C to 300°C. The wetting time, wetting forces, and activation energy of the wetting reaction were studied. The wetting time decreases with increasing temperature and increases with Ag content. The wetting force exhibits a disproportional correlation to temperature rise, while no trend was observed with respect to Ag content. The wetting behavior was ascribed to the interaction between Cu and Zn. The AgZn₃ compound was formed at the interface when the solder contains 0.3% Ag and above, while it was formed within the bulk solder at 2% Ag and above.     

Thermal diffusion of lithium acceptors into ZnO crystals

Electron paramagnetic resonance (EPR) has been used to monitor the diffusion of lithium ions into single crystals of ZnO. The in-diffusion occurs when a crystal is embedded in LiF powder and then held in air at temperatures near 750°C for periods of time ranging up to 22 h. These added lithium ions occupy zinc sites and become singly ionized acceptors (because the material is initially n type). A corresponding reduction in the number of neutral shallow donors is observed with EPR. To monitor the lithium acceptors, we temporarily convert them to the EPR-active neutral acceptor state by exposure to laser light (325 nm or 442 nm) at low temperatures. Also, after each diffusion treatment, we monitor the EPR signal of singly ionized copper acceptors and the photo-induced EPR signal of neutral nitrogen acceptors. These nitrogen and copper impurities are initially present in the crystal, at trace levels, and are made observable by the thermal anneals. Infrared-absorption measurements at room temperature in the 2–10 μm region show that the concentration of free carriers decreases as lithium is added to the crystal. After 22 h at 750°C in the LiF powder, the free-carrier absorption is no longer present, and the crystal is semi-insulating.     

Low-energy electron-enhanced etching of HgCdTe

Low-energy electron-enhanced etching (LE4) is applied to HgCdTe to eliminate ion-induced surface damage. First, LE4 results for patterned samples are illustrated. The LE4 mechanism is understood from a mechanistic study in terms of three etch variables: direct current (DC) bias, gas composition, and sample temperature. For this paper, the effects of DC bias (electron energy) and gas composition (CH₄ concentration) are summarized qualitatively, followed by quantitative evidence. Etch rate, the amount of polymer, surface stoichiometry, and surface roughness have specific relations with each etch variable under competition between pure LE4 and polymer deposition.     

Layer-transfer process for silicon-on-insulator with improved manufacturability

We observe hydrogen platelet buildup in single-crystalline silicon caused by hydrogen-plasma processing. The platelets are aligned along a layer of lattice defects formed in silicon before the plasma processing. The buried-defect layer is formed by either silicon-into-silicon or argon-into-silicon implantation. We discuss the platelet nucleation, growth, and merge phenomena and discuss applicability of the plasma hydrogenation to silicon-on-insulator (SOI) wafer fabrication by layer transfer.     

Solid-state intermetallic compound layer growth between copper and 95.5Sn-3.9Ag-0.6Cu solder

Long-term, solid-state intermetallic compound (IMC) layer growth was examined in 95.5Sn-3.9Ag-0.6Cu (wt.%)/copper (Cu) couples. Aging temperatures and times ranged from 70°C to 205°C and from 1 day to 400 days, respectively. The IMC layer thicknesses and compositions were compared to those investigated in 96.5Sn-3.5Ag/Cu, 95.5Sn-0.5Ag-4.0Cu/Cu, and 100Sn/Cu couples. The nominal Cu₃Sn and Cu₆Sn₅ stoichiometries were observed. The Cu₃Sn layer accounted for 0.4–0.6 of the total IMC layer thickness. The 95.5Sn-3.9Ag-0.6Cu/Cu couples exhibited porosity development at the Cu₃Sn/Cu interface and in the Cu₃Sn layer as well as localized “plumes” of accelerated Cu₃Sn growth into the Cu substrate when aged at 205°C and t>150 days. An excess of 3–5at.%Cu in the near-interface solder field likely contributed to IMC layer growth. The growth kinetics of the IMC layer in 95.5Sn-3.9Ag-0.6Cu/Cu couples were described by the equation x=x_o+Atⁿexp [−ΔH/RT]. The time exponents, n, were 0.56±0.06, 0.54±0.07, and 0.58±0.07 for the Cu₃Sn layer, the Cu₆Sn₅, and the total layer, respectively, indicating a diffusion-based mechanism. The apparent-activation energies (ΔH) were Cu₃Sn layer: 50±6 kJ/mol; Cu₆Sn₅ layer: 44±4 kJ/mol; and total layer: 50±4 kJ/mol, which suggested a fast-diffusion path along grain boundaries. The kinetics of Cu₃Sn growth were sensitive to the Pb-free solder composition while those of Cu₆Sn₅ layer growth were not so.     

Concentrations of native and gold defects in HgCdTe from first principles calculations

We have studied the defect formation energies of the various native (vacancies, interstitials, and antisites) and Au defects in Hg_1−xCd_xTe using density functional-based total energy calculations with ultrasoft pseudo-potentials. These studies are important for infrared (IR) detection technology where the device performance can be severely degraded because of defects. To calculate formation energies, we modeled the neutral and charged defects using supercells containing 64 atoms. From the formation energies, we have determined the defect concentrations as a function of stoichiometry and temperature. We find the prevalent neutral defects to be Au at the Hg site (Au_Hg ), Hg vacancies (V_Hg ), and Te antisites (Te_Hg ). We have also explicitly studied charged defects and have found Te _Hg ²⁺ , Au _Hg ¹⁻ , V _Hg ¹⁻ , V _Hg ²⁻ , and V _Te ²⁺ to have low formation energies. We have identified Au_Hg to be the prevalent Au defect, having concentrations several orders of magnitude greater than the other Au defects. We find that the charge state of V_Hg is primarily (1−) or (2−) depending on the electronic chemical potential.     

Phase equilibria of the Sn-Ag-Cu-Ni quaternary system at the sn-rich corner

Knowledge of phase equilibria of the Sn-Ag-Cu-Ni quaternary system at the Sn-rich corner is important for the understanding of the interfacial reactions at the Sn-Ag-Cu/Ni contacts, which are frequently encountered in recent microelectronic products. Various Sn-Ag-Cu-Ni alloys were prepared and equilibrated at 250°C. The alloys were then quenched and analyzed. The phases were determined by metallography, compositional analysis, and x-ray diffraction (XRD) analysis. No quaternary phases were found. The isoplethal sections at 60at.%Sn, 70at.%Sn, 80at.%Sn, and 90at.%Sn at 250°C are determined. The phase equilibrium relationship was proposed based on the quaternary experimental results and the 250°C isothermal sections of the four constituent ternary systems, Sn-Ag-Cu, Sn-Ag-Ni, Sn-Cu-Ni, and Cu-Ag-Ni. Because there are no ternary phases in all these three systems, all the compounds are in fact binary compounds with various solubilities of the other two elements.     

Rapid thermal annealing effects on the photoluminescence properties of molecular beam epitaxy-grown GaIn(N)As quantum wells with Ga(N)As spacers and barriers

This article describes the effects of rapid thermal annealing (RTA) on the photoluminescence (PL) emission from a series of GaIn(N)As quantum wells. Indium compositions of both 20% and 32% were examined with nominal N compositions of 1% or 2%. The N location was varied within our quantum structure, which can be divided into three regions: (1) quantum well, (2) Ga(N)As spacer layers at the barrier-to-well interface and well-to-barrier interface, and (3) barriers surrounding each quantum well. Eight combinations of samples were examined with varying In content, Ga(N)As spacer layer thickness, N content, and N location in the structure. In the best cases, the presence of these Ga(N)As spacer layers improves the PL properties, due to annealing, with a reduction in the emission wavelength blueshift by ～400 Å, a reduction of the decrease in the full-width at half-maximum (FWHM) by ～5 meV, and a threefold reduction of the increase in integrated intensity. It was also observed that relocating N from the quantum wells to the barriers produces a comparable emission wavelength both before and after annealing. Our results further show that the composition of incorporated N in the material is most influential during the stages of RTA in which relatively small amounts of thermal energy is present from our lower annealing times and temperatures. Hence, we believe a low thermal-energy anneal is responsible for the recovery of the plasma-related crystal damage that was incurred during its growth. However, the In composition in the quantum well is most influential during the latter stages of thermal annealing, at increased times and temperatures, where the wavelength blueshift was roughly independent of the amount of incorporated N. As a result, our investigations into the effects of RTA on the PL properties support other reports that suggest the wavelength blueshift is not due to N diffusion.     

Influence of Si<Subscript>3</Subscript>N<Subscript>4</Subscript> passivation on surface trapping in SiC metal-semiconductor field-effect transistors

The SiC metal-semiconductor field-effect transistors (MESFETs) have been reported to have current instability and strong dispersion caused by trapping phenomena at the surface and in the substrate, which degrade direct-current (DC) and radio-frequency (RF) performance. This paper illustrates the change in electrical characteristics of SiC MESFETs after Si₃N₄ passivation. Because of a reduction of surface trapping effects, Si₃N₄ passivation can diminish current collapse under pulsed DC conditions, increasing the RF power performance. The reduction of surface trapping effects is verified by the change in the ratio of the drain current to the gate current under pinch-off conditions.