This page contains a Flash digital edition of a book.
Interconnect Bonding


devices and the bondline has not changed significantly after stress testing.


Comparison of Electrical Results Table I summarizes and compares the electrical resistance results for three types of metal to metal bonding: Cu/Sn - Cu, CMP Cu - CMP Cu, and CMP Cu - Plated Cu. As expected the contact resistance of the Cu-Cu bonded samples was lower because there is no intermetallic compound formed. The Cu/Sn intermetallic increases the bond resistance by a factor of approximately three compared with CMP’d Cu-Cu. The total bond resistance for Cu/Sn-Cu bonding remains less than 100mΩ per bond.


Comparison of Shear Testing Results Table II summarizes and compares the bond strength results for the three types of metal to metal bonding that were used in this study: Cu/Sn - Cu, CMP Cu - CMP Cu and CMP Cu - Plated Cu. Die shear strength was measured using a Royce 550 instrument with maximum range of 10kg.


The devices were not underfilled for this


test. All three types of samples exhibited a bond strength of approximately 8kg +/- 1kg. For this die size, the minimum shear strength specified in MIL-STD-883E is 2.5kg. Failure occurred at the bond interface, as shown in Figs. 6 and 7.


Cu/Sn-Cu Shear Testing Fig. 6 shows representative top and bottom die surfaces from a Cu/Sn-Cu part after shear testing. The bottom die, right, shows the mechanical BCB key (dark circles around copper landing pads). Sn is visible on the copper pads and it can be seen to be offset towards the upper right corner, where further slippage was restrained by the BCB mechanical key. Shear failure occurred in the Cu/Sn IMC. The shear strength was above 10 kg.


Cu - Cu Shear Testing Fig. 7 shows representative top and bottom die surfaces from a CMP Cu - CMP Cu part after shear testing. The shear test result was 6 kg. All of the pads in this photo show unoxidized copper at the Cu-Cu bond. Shear failure occurred at the bondline.


Fig. 8 shows the bottom die surface for a Cu-Cu part after shear testing, revealing that the


corner and edge pads on the device did not bond, as shown by the surface oxidation on the pads. The lack of metal-metal bonding was caused by the within-die nonuniformity caused by dishing during CMP.


Low Temperature Bonding at 25µm To explore the feasibility of low temperature bonding, a set of devices was bonded at 210°C and compared with a set of devices otherwise similar but bonded at standard temperature (300°C).


These devices were similar in design to the devices tested for reliability above (325,632 interconnects in area array format) but were 25µm pitch instead of 10µm. The devices were measured for electrical yield, shear strength, and were cross sectioned for SEM and quantitative EDS analysis.


For both 300°C and 210°C bonding, electrical testing showed high channel yields were achieved. Average channel yield (each channel contains 1272 bonds) was 87%. The median resistance per bond was 127 mΩ/bond which includes the wiring and pad structures. In terms of yield and resistance the electrical behavior of the two sample sets was not significantly different.


Graphical maps of the Cu/Sn-Cu electrical yield show a random distribution, indicating a lack of misalignment, across-die nonuniformity or other systematic defects. Opens are most likely due to point defects. Individual bond yield is higher than 99.99%. Fig. 9 shows cross-sectional SEMs of dice bonded at 300°C and 210°C. The 300°C bonded sample used the standard process conditions described. The 210°C sample was bonded at 210°C for 5 min using 50MPa of pressure and


Table II. Comparison of the bond strength as measured by shear testing for 3 types of metal to metal


bonding: Cu/Sn - Cu, CMP Cu - CMP Cu and CMP Cu - Plated Cu


17


www.euroasiasemiconductor.com  Issue III 2011


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32