Gold metallizations for eutectic bonding of silicon wafers

Gold eutectic bonding of silicon wafers is a good candidate for wafer level vacuum packaging of vibrating MEMS: in this paper we investigated several e-beam evaporated metallizations stacks including a titanium adhesion layer, an optional diffusion barrier (Ni or Pt) and a gold film for eutectic bonding on Si and SiO₂/Si wafers. Interdiffusion in the multilayers for annealing temperatures (380–430°C) larger than the Au–Si eutectic temperature (363°C) and times corresponding to a bonding process was characterized by RBS, roughness and resistivity measurements. Au/Pt/Ti and Au/Ti/SiO₂ were found to have the best characteristics for bonding. This was confirmed by bonding experiments.     

Gold metallizations for eutectic bonding of silicon wafers

Gold eutectic bonding of silicon wafers is a good candidate for wafer level vacuum packaging of vibrating MEMS: in this paper we investigated several e-beam evaporated metallizations stacks including a titanium adhesion layer, an optional diffusion barrier (Ni or Pt) and a gold film for eutectic bonding on Si and SiO₂/Si wafers. Interdiffusion in the multilayers for annealing temperatures (380–430°C) larger than the Au–Si eutectic temperature (363°C) and times corresponding to a bonding process was characterized by RBS, roughness and resistivity measurements. Au/Pt/Ti and Au/Ti/SiO₂ were found to have the best characteristics for bonding. This was confirmed by bonding experiments.

     

Triple helix-forming oligonucleotides which make imperfect Watson-Crick duplexes that compete with the creation of the triplex

Because of the repetitive nature of the sequence, when titrating a G,A-rich, triple helix-forming oligonucleotide (TFO) with increasing concentrations of target duplex in order to obtain the dissociation constant of the complex, a duplex is sometimes first generated at intermediate concentrations of the target. This duplex is based on an imperfect Watson-Crick pairing of the TFO to the pyrimidine-rich strand of the target. An explanation is proposed for this duplex being obtained only in a certain domain of the titration range, before the triple helix becomes predominant.