Literatura científica selecionada sobre o tema "Surface Activated Bonding"

The surface activated bonding (SAB) technique enables room temperature bonding of metals, such as Au, by forming metal bonds between clean and reactive surfaces. However, the re-adsorption on the activated surface deteriorates the bonding quality, which limits the applicability of SAB for actual packaging processes of electronics. In this study, we propose and demonstrate the prolongation of the surface activation effect for room temperature bonding of Au by utilizing a self-assembled monolayer (SAM) protection. While the bonding without SAM fails after exposure of the activated Au surface to ambient air, the room temperature bonding is achieved using SAM protection even after 100 h exposure. The surface analysis reveals that the clean and activated Au surface is protected from re-adsorption by SAM. This technique will provide an approach of time-independent bonding of Au at room temperature.

Surface Activated Bonding (SAB) is interesting for strong silicon to silicon bonding at room temperature without any annealing needed, afterwards (1). Although it is a well-known technique, the activation step, in particular, is scarcely documented. This paper offers insights about the impact of soft activation parameters on the amorphous region at the bonding interface. In addition, the adherence energy of hydrophobic silicon bonding with SAB is quantified to better understand bonding mechanisms. With very low dose and acceleration activation parameters, the surface preparation prior to bonding becomes of paramount importance. Indeed, the silicon native oxide is typically removed during the activation step. The thin amorphous silicon region is a side effect of this singular surface preparation(2). In order to work around this potential roadblock, we used instead hydrophobic surface preparation to remove the native oxide, before entering into the activation step. Two types of preparation were evaluated in this study. First, a standard “HF-Last” chemical treatment was used on standard silicon wafers. This treatment removed the silicon native oxide and passivated the surface with Si-H and, to a lesser extent, Si-F bonds (3). We otherwise used epitaxy-reconstructed silicon wafers with fully hydrophobic surfaces (4). Silicon native oxide was removed thanks to an ultra-pure H2 bake at 1100°C, 20 Torr for 2 minutes in an epitaxy chamber. Then, several tens of nm of Silicon were deposited at 950°C to obtain, after another H2 bake, a silicon surface fully passivated by hydrogen atoms with atomically smooth terraces and mono-atomic step edges. Our EVG®ComBond® bonding tool, operating under ultra-high vacuum (UHV), is equipped with an accelerated argon ion beam to perform the activation step. The softest functional settings, on our set up, are 50V (acceleration) and 26 mA (dose). After beam initialization, the two sets of substrates pass through the activation chamber. Activated substrates are then transferred to the bonding chamber within 5 minutes of handling. The exposure time in the activation chamber was evaluated, the aim being to remove adsorbed hydrogen atoms on the silicon surface without any amorphous silicon generation. Different characterization techniques such as transmission electron microscopy or FTIR-MIR were used to quantify the amorphous layer formation and the potential Si-H bonds remaining (after activation). The adherence energy of the bonded pair was measured by a double cantilever beam method under prescribed displacement control in anhydrous atmosphere (5). Figure 1 shows the adherence energy (Gc=2γc) in mJ/m² as a function of activation exposure time with soft activation parameters for both wafer preparations. The 0s reference bonding was conducted without passing through the activation module. We then had very low adherence energies, around 50 mJ/m², as expected for standard hydrophobic silicon wafer bonding under UHV (6). Upon Ar+ exposure, behaviors were very different depending on surface preparation. The adherence energy barely increased with the Ar+ exposure time for “HF-Last” surfaces. Meanwhile, even 1s of exposure to Ar+ had a definite impact on the adherence energy of epi-reconstructed, atomically smooth silicon surfaces, which was definitely higher. The maximum difference between both wafer preparations occurred for 30 up to 60 seconds exposure times. This indicate a change in the bonding mechanism as the comparatively high roughness of the “HF-Last” silicon wafer started to be counter-balanced by activation. The experimental set up, the manufacturing process, as well as further characterizations will be presented. Cross-sectional TEM imaging of the bonding interface, FTIR-MIR and AFM measurements after surface preparation will help us better understand the specificities of such soft activation process on the SAB of hydrophobic surfaces. The impact of the amorphous silicon layer on bonding will be discussed. Suga T et al. STRUCTURE OF A1-A1 A N D A1-Si3N4 INTERFACES BONDED AT ROOM TEMPERATURE BY MEANS OF THE SURFACE ACTIVATION METHOD. Acta Metallurgica et Materialia 1992. Takagi H et al. Surface activated bonding of silicon wafers at room temperature. Appl Phys Lett. 1996. Abbadie A et al. Low thermal budget surface preparation of Si and SiGe. Appl Surf Sci. 2004. Sordes D et al. Nanopackaging of Si(100)H Wafer for Atomic-Scale Investigations. 2017. Maszara WP et al. Bonding of silicon wafers for silicon‐on‐insulator. J Appl Phys. 15 nov 1988;64(10):4943-50. Tong QY et al. The Role of Surface Chemistry in Bonding of Standard Silicon Wafers. J Electrochem Soc. 1997. Figure 1

Surface Activated Bonding (SAB) is interesting for strong silicon to silicon bonding at room temperature without any annealing needed, afterwards. This technique has been recognized by the scientific community for more than two decades now and was used for numerous reviewed applications. Although it is a well-known technique, the activation step, in particular, is scarcely documented. This paper offers insights about the impact of soft activation parameters on the amorphous region at the bonding interface. In addition, the adherence energy of hydrophobic silicon after SAB bonding is quantified, to better understand bonding mechanisms. Soft activation parameters on hydrophobic silicon substrates yield exceptionally thin bonding interfaces with acceptable bonding energy at room temperature. According to cross-sectional Transmission Electron Microscopy imaging, a 0.53 nm thick amorphous silicon interface was achieved with an adherence energy of 1337 ± 137 J/m² measured by the Double Cantilever Beam method.

In order to optimize the process parameters of Si-Si wafer direct bonding at room temperature, Si-Si surface activated bonding (SAB) was performed, and the effect of the argon ion beam current for surface activation treatment on the Si-Si bonding quality was investigated. For the surface activation under the argon ion beam irradiation for 300 s, a smaller ion beam current (10~30 mA) helped to realize a lower percentage of area covered by voids and higher bonding strength. Especially with the surface activation under 30 mA, the bonded Si-Si specimen obtained the highest bonding quality, and its percentage of area covered by voids and bonding strength reached <0.2% and >7.62 MPa, respectively. The transmission electron microscopy analyses indicate that there exists an ultrathin amorphous Si interlayer at the Si-Si bonding interface induced by argon ion beam irradiation to Si wafer surfaces, and its thickness increases as the argon ion beam current rises. The investigation results can be used to optimize the SAB process and promote the applications of SAB in the field of semiconductor devices.

In order to optimize the process parameters of Si-Si wafer direct bonding at room temperature, Si-Si surface activated bonding (SAB) was performed, and the effect of the argon ion beam current for surface activation treatment on the Si-Si bonding quality was investigated. For the surface activation under the argon ion beam irradiation for 300 s, a smaller ion beam current (10~30 mA) helped to realize a lower percentage of area covered by voids and higher bonding strength. Especially with the surface activation under 30 mA, the bonded Si-Si specimen obtained the highest bonding quality, and its percentage of area covered by voids and bonding strength reached <0.2% and >7.62 MPa, respectively. The transmission electron microscopy analyses indicate that there exists an ultrathin amorphous Si interlayer at the Si-Si bonding interface induced by argon ion beam irradiation to Si wafer surfaces, and its thickness increases as the argon ion beam current rises. The investigation results can be used to optimize the SAB process and promote the applications of SAB in the field of semiconductor devices.

In order to optimize the process parameters of Si-Si wafer direct bonding at room temperature, Si-Si surface activated bonding (SAB) was performed, and the effect of the argon ion beam current for surface activation treatment on the Si-Si bonding quality was investigated. For the surface activation under the argon ion beam irradiation for 300 s, a smaller ion beam current (10~30 mA) helped to realize a lower percentage of area covered by voids and higher bonding strength. Especially with the surface activation under 30 mA, the bonded Si-Si specimen obtained the highest bonding quality, and its percentage of area covered by voids and bonding strength reached <0.2% and >7.62 MPa, respectively. The transmission electron microscopy analyses indicate that there exists an ultrathin amorphous Si interlayer at the Si-Si bonding interface induced by argon ion beam irradiation to Si wafer surfaces, and its thickness increases as the argon ion beam current rises. The investigation results can be used to optimize the SAB process and promote the applications of SAB in the field of semiconductor devices.