The Effect of Metal Area and Line Spacing on TDDB Characteristics of 45nm Low-k SiCOH Dielectrics

Abstract

Low-k time-dependent dielectric breakdown (TDDB) is rapidly becoming one of the most important reliability issues in Cu/low-k technology development and qualification. Although considerable progress has been made in recent years in addressing the electric field dependence of low-k time-to-breakdown (t BD ), there has been very little comprehensive work done on the effect of metal area and line spacing on low-k TDDB. The lifetime of a product chip is typically obtained by extrapolating TDDB data from small test structures to large chip areas, and the low-k TDDB line spacing scaling rule normally should be considered for the definition of operating voltages for various technologies to assure long-term reliability. Therefore, both area scaling and line spacing scaling relations are of great importance, in order to have a robust technology qualification. In this study, a thorough investigation into the 45 nm low-k SiCOH TDDB was conducted in order to understand the breakdown failure statistics, to model the area dependence, and to explore the line spacing scaling. With the help of experimental results and computational simulations, the effect of line-to-line spacing on low-k TDDB was clearly identified and a methodology for accurate determination of Weibull shape factor is proposed. [

Methodology to determine the impact of linewidth variation on chip scale copper/low-k backend dielectric breakdown

“…It has been noted that the characteristic lifetime is dependent on line edge roughness (LER) [4,5,[14][15][16]. LER reduces the effective line space, i.e.…”

A methodology to extract failure rates for low-k dielectric breakdown with multiple geometries and in the presence of die-to-die linewidth variation

“…Instead, as in [3], the area scaling formula is used to extract the Weibull shape parameter. The relationship between characteristic lifetimes for structures with different areas is given by Fig.…”

“…1, the data points do not fall on a straight line, as expected for Weibull statistics. In [3] it was noted that linewidth variation can be as large as ±30%. This variation distorts the Weibull curves used to determine the lifetime of a structure.…”

“…As a result, direct extraction of the parameters, g and b, is inaccurate. It is suggested in [3] that die-to-die linewidth variation can be eliminated during Weibull parameter extraction through calibration of lifetime measurements based on capacitance measurements, from which the mean space between the lines of the comb structure can be computed. However, because of the complexity of the structure, capacitance is also impacted by variation in the low-k dielectric constant as a function of the composition of the stack, and consequently, variation in the dielectric constant is confounded with variation in distance.…”

An efficient approach to quantify the impact of Cu residue on ELK TDDB

“…Both spacing and surface quality affect the IMD TDDB lifetime; however, it is difficult to identify which is the dominant factor on the IMD TDDB result. Recently, many publications have described the spacing effect on the IMD TDDB lifetime [1] [2]. Some data can be simulated well using a real space simulation model; however, some ultimately worse case can't, as shown in Fig.1.…”