Yasumitsu Orii

IMC bonding for 3D interconnection

We performed stacking experiments on Si dies using annular tungsten TSVs (Through Silicon Vias) and Cu studs with low-volume solder micro-bumps. Unlike standard 100-micron C4 (Controlled Collapse Chip Connection) solder balls, very small solder volumes (< 6 microns in height) form IMC (InterMetallic Compounds) in the junctions during the bonding or reflow processes. The two interconnect metallurgies of Cu/Ni/In and Cu/Sn joints were considered for low-volume lead-free solder micro-bumps for 3D integration. A previous study on these metallurgies [5] showed that the Cu/Sn joints form thermally stable intermetallics while in the Cu/Ni/In joints, some indium solder remains unreacted due to the presence of the Ni barriers. The shear testing on the stacked systems showed that the die stacks with Cu/Sn joints exhibit higher shear strengths than those with Cu/Ni/In joints. However the impact shock testing on the systems revealed that the die stacks with Cu/Sn joints are less resistant to mechanical shocks than the systems with Cu/Ni/In joints. This new work focuses on thermal cycle testing of the die stack systems with the Cu/Ni/In and Cu/Sn interconnections. Preliminary thermal cycle testing on the die stack systems with Cu/Ni/In joints showed that the joints are stable against thermal cycle stresses for thousands of cycles. To quickly compare the systems with two metallurgies, we mounted the Si die stacks onto organic substrates to impose additional stresses on the systems. In addition to standard DTC (Deep Thermal Cycle) tests, we also conducted a HAATS (Highly accelerated Air to Air Thermal Shock) test [23] with a short cycle time to reduce the testing time. The DTC and HAATS tests showed that the stacked systems with Cu/Ni/In joints had fewer failures and smaller increases in the electrical resistances of the joints during the tests than the systems with Cu/Sn joints.

Ultrafine-pitch C2 flip chip interconnections with solder-capped Cu pillar bumps

PoP structures have been used widely in digital consumer electronics products such as digital still cameras and mobile phones. However, the final stack height from the top to the bottom package for these structures is higher than that of the current stacked die packages. To reduce the height of the package, a flip chip technology is used. Since the logic chips of mobile applications use a pad pitch of less than 80 µm or less, an ultra-fine-pitch flip chip interconnection technique is required. C4 flip chip technology is widely used in area array flip chip packages, but it is not suitable in the ultrafine-pitch flip chips because the C4 solder bumps melt and collapse on the wide opening Cu pads. Although the industry uses ultrafine-pitch interconnections between Au stud bumps on a chip and Sn/Ag pre-solder on a carrier, this flip chip technique has two major problems. One is that the need for bumps on both die and carrier drives up material costs. The other is that the long bonding process time required in the individual flip chip bonding process with associated heating and cooling steps demands large investments in equipment. To address these problems, we developed the mount and reflow with no-clean flux processes, and new interconnection techniques were developed with Cu pillars and Sn/Ag solder bumps on Al pads for wirebonding, were developed. It is very easy to control the gap between die and substrate by adjusting the Cu pillar height. Since it is unnecessary to control the collapse of the solder bumps, we call this the C2 process for direct Chip Connection (C2). The C2 bumps are connected to Cu substrate pads, which are a surface treated with OSP (Organic Solder Preservative), with reflow and no-clean processes. This technology creates the SMT/Flip Chip hybrid assembly for SoP (System on Package) use. We have produced 50 µm-pitch C2 interconnections and tested their reliability. The interconnection resistance increase caused by the reliability testing is quite small. It is clear that C2 flip chip technology provides robust solder connections at low cost. Also the C2 structure with a low-k device was evaluated and no failures were observed at 1,500 cycles in the thermal cycle test. This indicates that low-k C2 structures seem robust. For finer pitch flip chip interconnections, a wafer-level underfill process is needed to overcome the limitations of the standard capillary underfill process for ultra-narrow spaces. To date, a wafer- level underfill process exists for the C2 process with an 80-µm pitch. In addition to fine pitch interconnections, a die thickness of 70 µm is required to reduce the final stack height. Such thin die cannot be processed by the C2 process because such dies slip too easily during the reflow process. To resolve this issue, a Post-Encapsulation Grinding (PEG) method was developed. In this method the die is ground to less than 70 µm after joining and underfilling. This report presents the PEG method and reliability test results for die thicknesses 20 µm, 70 µm and 150 µm.

Solder joint reliability evaluation of chip scale package using a modified Coffin–Manson equation

Abstract Thermal cycle tests were performed for chip scale package (CSP) solder joints with Sn–37mass%Pb under several thermal cycle conditions. Under the conventional thermal cycle conditions, which heat up to approximately 100 °C, microstructure coarsening occurred and solder joints were fractured. The thermal fatigue lives followed the modified Coffin–Manson equation. The exponential factors m and n, and the activation energy Q in that equation were evaluated as 0.33, −1.9 and 15.5 kJ/mol, respectively. When the maximum temperature is room temperature and the temperature range is very narrow, the solder joint fracture occurred without microstructure coarsening, and the thermal fatigue life does not follow the modified Coffin–Manson equation.

Investigations of cooling solutions for three-dimensional (3D) chip stacks

Three-dimensional (3D) chip stacks are receiving more attention for system performance enhancements. However, because of the higher circuit density, the cooling of 3D chip stacks gets more challenging. In conventional cooling methods, a heat sink or a micro-channel cooler is located at the top of the chip to dissipate the generated heat in a chip. In this paper, possible cooling methods from the bottom of a silicon interposer and cooling from the peripheral of a silicon interposer were proposed and evaluated. Based on the experimentally obtained thermal resistance of lead-free (SnAg) interconnections, the cooling performances of the above two cooling solutions were investigated by modeling and the requirements were clarified.

Development of vacuum underfill technology for a 3D chip stack

We developed a vacuum underfill technology for 3D chip stacks and for flip chips in high performance system integration. We fabricated a 3D prototype chip stack using the vacuum underfill technology to apply the adhesive. The underfill was injected into each 6 µm gaps in a 3-layer chip stack and no voids were detected in acoustic microscopy images. Electrical tests and thermal reliability tests were used to measure the resistance of the vertical interconnections and the impact of the underfill. The results showed there was minimal difference in the average interconnection resistance of the chip stack with and without underfill.

Experimental thermal resistance evaluation of a three-dimensional (3D) chip stack

To propose an appropriate cooling solution for a three-dimensional (3D) chip stack at the design phase, it is necessary to estimate the total thermal resistance of a 3D chip stack. The interconnection between stacked chips is considered as one of the thermal resistance bottleneck of a 3D chip stack, but it is not experimentally clear yet. We have previously measured the thermal conductivity of SnAg with Cu post to be 37–41W/mC by a steady state thermal resistance measurement method, using the sample which was simply composed of two Si chips and SnAg with Cu post between two Si chips. In this study, 3D stacked test chips are fabricated, which are implemented with PN junction diodes for temperature sensors and diffused resistors for heating, and the thermal conductivity of the interconnection in actual 3D stacked structure is experimentally obtained. The temperature distributions of two 3-layer-stacked-test-chips are measured and the equivalent thermal conductivity of the interconnection is experimentally obtained to be 1.6W/mC. This value is compared with the measured thermal conductivity of SnAg with Cu post (37–41W/mC) and its adequacy is examined.

Experimental thermal resistance evaluation of a three-dimensional (3D) chip stack, including the transient measurements

For the thermal management of three-dimensional (3D) chip stack, its thermal resistance needs to be clearly understood. In this study, 3D stacked test chips are fabricated, which are implemented with PN junction diodes for temperature sensors and diffused resistors for heating. At SemiTherm2011, the equivalent thermal conductivity of the interconnection, including BEOL (Back-End-Of-the-Line, wiring layer) is experimentally obtained to be 1.6W/mC and this time, we measure the thermal effect of Cu TSVs and it is experimentally supported that as the Cu TSV area ratio increases, the thermal conductivity of chip with TSVs in the vertical direction increases, on the contrary, that in the horizontal direction decreases. Also, the transient thermal measurement is performed and its result is compared with steady state measurement result. Further, the thermal capacitance measurement of 3D stacked test chip with hot spot heating is performed, which is essential to determine the transient thermal performance of 3D chip stack.

Thermal Stresses of Through Silicon Vias and Si Chips in Three Dimensional System in Package

Rbased on finite element method (FEM) was used to simulate the effects of voids formed inside Cu TSVs on the thermal conduction and mechanical stresses in the TSV structure. The thermal performance that was required in 3D SiP was estimated to ensure the reliability. Simulations for thermal stresses in the TSV structure in 3D SiP were carried out under thermal condition due to power ON/ OFF of device. In case that void was not present inside the TSV, the stresses in TSV were close to the hydrostatic pressure and the magnitude of the equivalent stress was lower than the yield stress of copper. Maximum principal stress of the Si chip in the TSV structure for the case without voids was lower than that of the bending strength of silicon. However, the level of the stresses in the Si chips should not be negligible for damages to Si chips. In case that void was present inside the TSV, stress concentration was occurred around the void in the TSV. The magnitude of the equivalent stress in the TSV was lower than the yield stress of copper. The magnitude of the maximum principal stress of the Si chip was lower than that of the bending strength of silicon. However, its level should not be negligible for damages to TSVs and Si chips. The stress on inner surfaces of Si chip was slightly reduced due to the presence of a void in the TSV. [DOI: 10.1115/1.4006515]

Through silicon via process for effective multi-wafer integration

We propose a novel 3D integration method, called Vertical integration after Stacking (ViaS) process. The process enables 3D integration at significantly low cost, since it eliminates costly processing steps such as chemical vapor deposition used to form inorganic insulator layers and Cu plating used for via filling of vertical conductors. Furthermore, the technique does not require chemical-mechanical polishing (CMP) nor temporary bonding to handle thin wafers. The integration technique consists of forming through silicon via (TSV) holes in pre-multi-stacked wafers (> 2 wafers) which have no initial vertical electrical interconnections, followed by insulation of holes by polymer coating and via filling by molten metal injection. In the technique, multiple wafers are etched at once to form TSV holes followed by coating of the holes by conformal thin polymer layers. Finally the holes are filled by using molten metal injection so that a formation of interlayer connections of arbitrary choice is possible. In this paper, we demonstrate 3-chip-stacked test vehicle with 50 × 50 μm-square TSVs assembled by using this technique.

Wafer IMS (Injection molded solder) — A new fine pitch solder bumping technology on wafers with solder alloy composition flexibility

In this paper, we will describe a new low cost solder bumping technology for use on wafers. The wafer IMS (injection molded solder) process can form fine pitch solder bumps on wafers, while offering greater solder alloy flexibility. This method is also applicable to form uniform solder bump heights when a wafer has different size and shape of I/O pads. The wafer IMS bumping process uses a solder injection head that melts the desired bulk solder alloy composition and then dispenses the molten solder into resist material cavities on wafers within a nitrogen environment. The injected molten solder contacts and wets to the metal pads without flux, thus forming intermetallic compounds at the solder/pad interface. After stripping the resist material, solder bumps exhibit straight side walls and round tops as the solders have solidified inside the cavities of this resist film. This particular geometry is unique and offers a ready-for-substrate bonding condition without an additional reflow step. In the case of using Cu pillars, one resist material is used for both Cu electroplating and molten solder injection. After patterning the resist material, the Cu pillars are electroplated to the desired height, and the remaining cavities of resist material are filled by the injection of molten solder. The final bump height is defined by the thickness of the resist material. Therefore, any non-uniformity of Cu pillar height across a wafer is masked by the final solder bump uniformity. A prototype tool for wafer IMS bumping technology has been developed and solder bumping has successfully been demonstrated with Sn-3.0Ag-0.5Cu solder on 200mm wafers. The test wafer employed interconnects pads of four different diameters and three different shapes. Other solder compositions have also been tried successfully.