Electromigration of lower and upper Cu lines in dual-damascene Cu interconnects

2003 ◽  
Vol 766 ◽  
Author(s):  
Ahila Krishnamoorthy ◽  
Guo Qiang ◽  
Anand V. Vairagar ◽  
Subodh Mhaisalkar
Keyword(s):  
Activation Energy ◽  
Activation Energies ◽  
Resistance Change ◽  
Cu Interconnects ◽  
Dual Damascene ◽  
Different Types ◽  
The Asymmetry ◽  
Damascene Interconnects ◽  
Calculated Activation

AbstractElectromigration in upper M2 and lower M1 Cu lines in dual damascene interconnects was investigated by using two types of via terminated structures. Two different types of failures termed as abrupt and gradual failures were observed. Electromigration failures in M2 were mostly by gradual failures, while abrupt failures dominated in M1. Significantly different types of voids were observed in M1 as compared to those in M2, though similar resistance change was observed. Experimentally calculated activation energies for M1 and M2 indicated interfacial electromigration. Activation energy of M1 was found to be much lower than that of M2. This clearly demonstrates the asymmetry of electromigration in M2 and M1 Cu lines in dual damascene interconnects.

Mortality Dependence of Cu Dual Damascene Interconnects on Adjacent Segment

2004 ◽  
Vol 812 ◽  
Author(s):  
C.W. Chang ◽  
C.L. Gan ◽  
C.V. Thompson ◽  
K.L. Pey ◽  
W.K. Choi ◽  
...  
Keyword(s):  
Current Flow ◽  
Adjacent Segment ◽  
Cu Interconnects ◽  
Single Segment ◽  
Dual Damascene ◽  
Test Conditions ◽  
Damascene Interconnects ◽  
Interconnect Network

AbstractThree terminal ‘dotted-I’ interconnect structures, with vias at both ends and an additional via in the middle, were tested under a variety of test conditions. Failures (mortalities) were observed even when segments were tested under conditions that would not have led to failure in two-terminal structures. Mortalities were found in right segments with jL values as low as 1250 A/cm, which is lower when compared to the immortality condition (jL)cr of 3700 A/cm reported for similar via-terminated structures. Moreover, we found that the mortality of a dotted-I segment is dependent on the direction and magnitude of the current in the adjacent segment. These results suggest that there is not a definite value of the jL product that defines true immortality in individual segments that are part of an interconnect tree. More importantly, the critical jL value for a single segment of Cu interconnects may be reduced or increased by an adjacent segment. Therefore independently determined (jL)cr values cannot be directly applied to interconnects with branched segments, but rather the magnitude as well as the direction of the current flow in the adjoining segments must be taken into consideration in evaluating the immortality of interconnect segments in an interconnect network.

scholarly journals High-Temperature Deformation of Naturally Aged 7010 Aluminum Alloy

Metals ◽  
2021 ◽  
Vol 11 (4) ◽  
pp. 581
Author(s):  
Abdulhakim A. Almajid
Keyword(s):  
Activation Energy ◽  
Aluminum Alloy ◽  
High Temperature ◽  
Threshold Stress ◽  
Activation Energies ◽  
Temperature Deformation ◽  
High Temperature Range ◽  
Temperature Range ◽  
True Activation Energy ◽  
Two Temperatures

This study is focused on the deformation mechanism and behavior of naturally aged 7010 aluminum alloy at elevated temperatures. The specimens were naturally aged for 60 days to reach a saturated hardness state. High-temperature tensile tests for the naturally aged sample were conducted at different temperatures of 573, 623, 673, and 723 K at various strain rates ranging from 5 × 10−5 to 10−2 s−1. The dependency of stress on the strain rate showed a stress exponent, n, of ~6.5 for the low two temperatures and ~4.5 for the high two temperatures. The apparent activation energies of 290 and 165 kJ/mol are observed at the low, and high-temperature range, respectively. These values of activation energies are greater than those of solute/solvent self-diffusion. The stress exponents, n, and activation energy observed are rather high and this indicates the presence of threshold stress. This behavior occurred as a result of the dislocation interaction with the second phase particles that are existed in the alloy at the testing temperatures. The threshold stress decreases in an exponential manner as temperature increases. The true activation energy was computed by incorporating the threshold stress in the power-law relation between the stress and the strain. The magnitude of the true activation energy, Qt dropped to 234 and 102 kJ/mol at the low and high-temperature range, respectively. These values are close to that of diffusion of Zinc in Aluminum and diffusion of Magnesium in Aluminum, respectively. The Zener–Hollomon parameter for the alloy was developed as a function of effective stress. The data in each region (low and high-temperature region) coalescence in a segment line in each region.

scholarly journals Effect of CaO on catalytic combustion of semi-coke

2021 ◽  
Vol 10 (1) ◽  
pp. 011-020
Author(s):  
Luyao Kou ◽  
Junjing Tang ◽  
Tu Hu ◽  
Baocheng Zhou ◽  
Li Yang
Keyword(s):  
Activation Energy ◽  
Apparent Activation Energy ◽  
Combustion Rate ◽  
Activation Energies ◽  
Combustion Reaction ◽  
Tg Analysis ◽  
Conversion Rates ◽  
Apparent Activation Energies ◽  
Ozawa Integral Method ◽  
Addition Amount

Abstract Generally, adding a certain amount of an additive to pulverized coal can promote its combustion performance. In this paper, the effect of CaO on the combustion characteristics and kinetic behavior of semi-coke was studied by thermogravimetric (TG) analysis. The results show that adding proper amount of CaO can reduce the ignition temperature of semi-coke and increase the combustion rate of semi-coke; with the increase in CaO content, the combustion rate of semi-coke increases first and then decreases, and the results of TG analysis showed that optimal addition amount of CaO is 2 wt%. The apparent activation energy of CaO with different addition amounts of CaO was calculated by Coats–Redfern integration method. The apparent activation energy of semi-coke in the combustion reaction increases first and then decreases with the increase in CaO addition. The apparent activation energies of different samples at different conversion rates were calculated by Flynn–Wall–Ozawa integral method. It was found that the apparent activation energies of semi-coke during combustion reaction decreased with the increase in conversion.

scholarly journals The Effect of α-Al(MnCr)Si Dispersoids on Activation Energy and Workability of Al-Mg-Si-Cu Alloys during Hot Deformation

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Xiaoguo Wang ◽  
Jian Qin ◽  
Hiromi Nagaumi ◽  
Ruirui Wu ◽  
Qiushu Li
Keyword(s):  
Activation Energy ◽  
Aluminum Alloys ◽  
Constitutive Equation ◽  
Hot Deformation ◽  
Flow Instability ◽  
Void Formation ◽  
Activation Energies ◽  
Compression Tests ◽  
Softening Mechanism ◽  
Dynamic Softening

The hot deformation behaviors of homogenized direct-chill (DC) casting 6061 aluminum alloys and Mn/Cr-containing aluminum alloys denoted as WQ1 were studied systematically by uniaxial compression tests at various deformation temperatures and strain rates. Hot deformation behavior of WQ1 alloy was remarkably changed compared to that of 6061 alloy with the presence of α-Al(MnCr)Si dispersoids. The hyperbolic-sine constitutive equation was employed to determine the materials constants and activation energies of both studied alloys. The evolution of the activation energies of two alloys was investigated on a revised Sellars’ constitutive equation. The processing maps and activation energy maps of both alloys were also constructed to reveal deformation stable domains and optimize deformation parameters, respectively. Under the influence of α dispersoids, WQ1 alloy presented a higher activation energy, around 40 kJ/mol greater than 6061 alloy’s at the same deformation conditions. Dynamic recrystallization (DRX) is main dynamic softening mechanism in safe processing domain of 6061 alloy, while dynamic recovery (DRV) was main dynamic softening mechanism in WQ1 alloy due to pinning effect of α-Al(MnCr)Si dispersoids. α dispersoids can not only resist DRX but also increase power required for deformation of WQ1 alloy. The microstructure analysis revealed that the flow instability was attributed to the void formation and intermetallic cracking during hot deformation of both alloys.

New experiments on the relationship between light-induced defects and photoconductivity degradation

2001 ◽  
Vol 664 ◽  
Author(s):  
Stephan Heck ◽  
Howard M. Branz
Keyword(s):  
Activation Energy ◽  
Defect Density ◽  
Activation Energies ◽  
Dangling Bond ◽  
Fast Recovery ◽  
Slow Recovery ◽  
Degradation Model ◽  
Induced Defects ◽  
The Relationship ◽  
Hydrogenated Amorphous

ABSTRACTWe report experimental results that help settle apparent inconsistencies in earlier work on photoconductivity and light-induced defects in hydrogenated amorphous silicon (a-Si:H) and point toward a new understanding of this subject. After observing that light-induced photoconductivity degradation anneals out at much lower T than the light-induced increase in deep defect density, Han and Fritzsche[1] suggested that two kinds of defects are created during illumination of a-Si:H. In this view, one kind of defect degrades the photoconductivity and the other increases defect sub-bandgap optical absorption. However, the light-induced degradation model of Stutzmann et al.[2] assumes that photoconductivity is inversely proportional to the dangling-bond defect density. We observe two kinds of defects that are distinguished by their annealing activation energies, but because their densities remain in strict linear proportion during their creation, the two kinds of defects cannot be completely independent.In our measurements of photoconductivity and defect absorption (constant photocurrent method) during 25°C light soaking and during a series of isochronal anneals between 25 < T < 190°C, we find that the absorption measured with E ≤1.1 eV, first increases during annealing, then exhibits the usual absorption decrease found for deeper defects. The maximum in this absorption at E ≤1.1eV occurs simultaneously with a transition from fast to slow recovery of photoconductivity. The absorption for E ≤1.1eV shows two distinct annealing activation energies: the signal rises with about 0.87 eV and falls with about 1.15 eV. The 0.87 eV activation energy roughly equals the activation energy for the dominant, fast, recovery of photoconductivity. The 1.15 eV activation energy roughly equals the single activation energy for annealing of the light-induced dangling bond absorption.

scholarly journals Fatal Void Size Comparisons in Via-Below and Via-Above Cu Dual-Damascene Interconnects

2004 ◽  
Vol 812 ◽  
Author(s):  
Z. -S. Choi ◽  
C. L. Gan ◽  
F. Wei ◽  
C. V. Thompson ◽  
J. H. Lee ◽  
...  
Keyword(s):  
Numerical Simulation ◽  
Failure Mechanisms ◽  
Simulation Tool ◽  
Void Size ◽  
Dual Damascene ◽  
Preferential Nucleation ◽  
Failure Correlation ◽  
Void Volumes ◽  
Damascene Interconnects ◽  
Critical Void

AbstractThe median-times-to-failure (t50's) for straight dual-damascene via-terminated copper interconnect structures, tested under the same conditions, depend on whether the vias connect down to underlaying leads (metal 2, M2, or via-below structures) or connect up to overlaying leads (metal 1, M1, or via-above structures). Experimental results for a variety of line lengths, widths, and numbers of vias show higher t50's for M2 structures than for analogous M1 structures. It has been shown that despite this asymmetry in lifetimes, the electromigration drift velocity is the same for these two types of structures, suggesting that fatal void volumes are different in these two cases. A numerical simulation tool based on the Korhonen model has been developed and used to simulate the conditions for void growth and correlate fatal void sizes with lifetimes. These simulations suggest that the average fatal void size for M2 structures is more than twice the size of that of M1 structures. This result supports an earlier suggestion that preferential nucleation at the Cu/Si3N4 interface in both M1 and M2 structures leads to different fatal void sizes, because larger voids are required to span the line thickness in M2 structures while smaller voids below the base of vias can cause failures in M1 structures. However, it is also found that the fatal void sizes corresponding to the shortest-times-to-failure (STTF's) are similar for M1 and M2, suggesting that the voids that lead to the shortest lifetimes occur at or in the vias in both cases, where a void need only span the via to cause failure. Correlation of lifetimes and critical void volumes provides a useful tool for distinguishing failure mechanisms.

scholarly journals On the catalytic dehydrogenation of naphthenes II. Competition experiments and energies of C-H bonds

1947 ◽  
Vol 190 (1022) ◽  
pp. 309-328 ◽  
Keyword(s):  
Activation Energy ◽  
Active Sites ◽  
Catalyst Surface ◽  
Resonance Energy ◽  
Activation Energies ◽  
Theoretical Treatment ◽  
Catalytic Dehydrogenation ◽  
Simple Relationship ◽  
Hydrogen Atoms ◽  
Methyl Cyclohexane

The rates of dehydrogenation in competition experiments using mixtures of two naphthenes, or a naphthene and a cyclic mono-olefine or two cyclic mono-olefines, have been examined theoretically and experimentally for the stationary state conditions. Provided the two reactants can occupy the same sites on the catalyst surface, then the ratio of the rates should be directly proportional to the ratio of the partial pressures at any instant. Theory suggests that a constant which can be derived from these competition experiments should be independent of the overall pressures, or of the initial ratio of concentrations or of the overall extent of dehydrogenation. Further, the ratio of the rates in competition should bear no simple relationship to the ratio of the individual rates alone, but should be related to the slopes of the 1/rate against 1/pressure plot for the two components considered separately. Moreover, the constant should be a ratio of two functions each of which is characteristic of one of the naphthenes. The theoretical conclusions have been confirmed experimentally which proves either that the groups of active sites on the catalyst surface are widely separated or that any set of sites is available for the reaction of any molecular species, and no interference takes place between naphthene molecules adsorbed on adjacent sites. Proof that a naphthene and cyclohexene are dehydrogenated on the same sites is supplied by the observation that a constant is obtained when different mixtures of cyclohexene and trans -1:4-dimethyl cyclohexane are allowed to compete for the surface. The ratios for methyl, ethyl, the three dimethyl and the three trimethyl cyclohexanes in competition with cyclohexane have been accurately determined at temperatures of 400 and 450° C. From the constants so derived the activation energy differences for the removal of the first pair of hydrogen atoms has been obtained. These values are discussed in terms of the possible transition complexes, and it is shown that the reaction proceeds by the loss of a pair of hydrogen atoms simultaneously and not by a half-hydrogenated state mechanism. Using these activation energies and the experimentally found overall activation energy of 36 kcal./g. mol., the resonance energy per resonating structure was determined as 1-73 kcal. This is in good agreement with the energies of C-H bonds in alkyl radicals (2-2 kcal./g.mol./ resonating structure). The theoretical treatment suggests that the weakest C-H link in methyl cyclohexane should be in the three position to the methyl group. A study of the activation energies involved shows that the methyl cyclohexene produced from methyl cyclohexane is not 1-methyl-1-cyclohexene, thus confirming the theoretical deduction.

scholarly journals Length Effects on the Reliability of Dual-Damascene Cu Interconnects

2002 ◽  
Vol 716 ◽  
Author(s):  
F. Wei ◽  
C. L. Gan ◽  
C. V. Thompson ◽  
J. J. Clement ◽  
S. P. Hau-Riege ◽  
...  
Keyword(s):  
Current Density ◽  
Refractory Metal ◽  
Line Length ◽  
Complex Behavior ◽  
Cu Interconnects ◽  
Dual Damascene ◽  
Metal Liner ◽  
Growth Of Voids ◽  
Length Effects

AbstractWe have carried out experiments on dual-damascene Cu interconnects with different lengths. We find that at short lengths, similar to Al-based interconnects, the reliability of Cubased interconnects improves. Also like Al interconnects, some short Cu segments do not form voids that cause failure before back-stresses prevent the further growth of voids. However, unlike Al-based interconnects, there is no apparent deterministic current-density line-length product (jL) for which all lines are immortal. This is related to the absence of a conducting refractory-metal overlayer in Cu-technology that can shunt current around small voids. Also unlike Al, we find that at long lengths a sub-population of Cu lines is immortal. We propose that this is the result of rupture of the thin refractory metal liner at the base of the dual-damascene Cu vias. As a consequence of this complex behavior, median times to failure and lifetime variations are minimum at intermediate line lengths.

Sign in / Sign up

Export Citation Format

Share Document