Search

The electromigration (EM) activation energy ( $\text {E}_{\text {A}}$ ) of alternative metals, such as Ru and Co, was obtained using low-frequency noise (LFN) measurements. High activation energies were expected, but values of ≈1 eV are found, most likely related to diffusion along with the metal-dielectric interface. Wafer-level accelerated EM tests were carried out to compare the LFN $\text {E}_{\text {A}}$ to the EM $\text {E}_{\text {A}}$ in the Ru wires. The calculation of the EM $\text {E}_{\text {A}}$ is found to be strongly dependent on the assumed temperature profile in the wire due to Joule heating (JH). The temperature profile was calculated analytically, assuming the contacts are at ambient temperature. For a void forming in direct proximity of the contact, the EM $\text {E}_{\text {A}}$ then matches the LFN $\text {E}_{\text {A}}$ . For a void at average wire temperature (ambient + JH), $\text {E}_{\text {A}}\approx $ 2 eV. In addition to demonstrating the application of LFN to study EM in alternative metals, this article also cautions for the impact of JH on the calculation of $\text {E}_{\text {A}}$ in interconnects.

A model explaining Lorentzian low-frequency noise spectra observed in electronic interconnects is presented. The model is based on the interaction of electrons with vacancies, whose thermodynamic equilibrium properties determine the temperature dependence of the noise. The model, therefore, allows calculating vacancy formation and migration enthalpies. It is experimentally verified on copper and tungsten interconnects; the sum of formation and migration enthalpies matches the electromigration activation energy obtained using standard accelerated tests.

Co has elicited a strong interest to replace Cu for future interconnect applications in microelectronic circuits due to its potentially lower resistivity and better reliability at scaled dimensions. Here, we demonstrate Co wires with electrical cross-sectional areas as small as 34 nm2 using a simple subtractive patterning technique. Semiclassical resistivity modeling of the wire resistivity indicates that grain boundary scattering is the dominant contribution for cross-sectional areas of the order of 100 nm2, while the contribution of surface scattering increases drastically below 50 nm2. Furthermore, wafer-level reliability tests of the Co wires show the high fusing current densities and a strong resistance to thermoelectric stress demonstrating the potential for robust reliability. The resistivity and the reliability of the Co wires are comparable with those of Ru wires.

In this paper we discuss a new EM test methodology, based on low-frequency noise (LFN) measurements. The main advantages of LFN over the standard accelerated EM tests are that they are non-destructive, much faster, closer to operation conditions and provide more fundamental understanding. Using the LFN technique, we study the EM properties in sub-30 nm line-width Cu interconnects with various metallization schemes. Furthermore, the EM activation energies of alternative metal interconnects (Ru, Co, W) are studied by means of LFN measurements.

© 2019 Author(s). 1.8-MeV proton irradiation to a fluence of 10 14 /cm 2 does not significantly affect the resistance or low-frequency noise of copper or ruthenium resistors fabricated via modern microelectronic fabrication techniques used to form metal lines. The room-temperature noise of these Cu and Ru resistors is surprisingly similar to that of Cu and Pt metal lines and wires fabricated using late-1970s nanofabrication techniques; however, measurements of the temperature dependence of the noise show that the defect kinetics are quite different among the various materials. A large increase in the noise magnitude observed above 200 K in Cu but not in Ru is consistent with the superior resistance to electromigration that Ru lines have shown, relative to Cu. ispartof: Applied Physics Letters vol:114 issue:20 status: published

This paper demonstrates the deteriorating electromigration (EM) reliability of Cu interconnects when scaling the metal pitch from 54 to 44nm. The study is carried out using a new EM test method based on low-frequency noise. Both EM lifetime and activation energy are found to decrease when scaling the line-width from 32 to 22nm. The decreasing activation energy is attributed to an increased number of grain boundaries and therefore enhanced grain boundary diffusion in narrow lines. The paper also shows that EM lifetimes can be qualitatively predicted based on the data obtained from LFN measurements.

Electromigration mechanisms of W-heaters with Cu connections at each line end to supply the current to the heater were studied. For making reliable lifetime estimates, a temperature profile is proposed based on infra-red (IR) microscopy, electromigration activation energies and failure location. For our structure, we show that the temperature of the void location is close to the ambient temperature and hence the lifetime prediction methodology should take this into account. Also, different test methodologies for this application are benchmarked, where we show that iso-thermal electromigration tests have only a limited impact on the lifetimes of our structures.

Reliability challenges of candidate metal systems to replace traditional Cu wiring in future interconnects are discussed. From a reliability perspective, a key opportunity is electromigration improvement: due to their high melting point and slower self-diffusion kinetics, higher current carrying capabilities are possible. Also, the higher cohesive energy and better resistance to oxidation of some metals potentially allows for barrierless integration, although adhesion properties must be carefully optimized. Besides avoiding small grain pinning and enabling high aspect ratio trench fill, the main processing challenges are identified to be a) avoiding seam voids, b) adhesion, c) CMP and d) disruptive metal etch. Main reliability challenges are related to higher mechanical stresses and higher joule heating which could lead to delamination during further processing and packaging and to enhanced electromigration in nearby metal lines.

We show the first electromigration (EM) failures of full ruthenium interconnects with a cross sectional area of 60nm2. The void is observed at the anode, which demonstrates that in p-type metals, such as Ru, the electromigration force acts in the direction of the electric field. The conventional representation of electromigration as electrons transferring their momentum onto the metal ions, thus has to be adapted for such metals. Moreover, we find that diffusion at the Ru-SiO 2 interface is the dominant diffusion mechanism for EM failure in these specific Ru lines.

The enhanced electromigration (EM) performance of 20 nm-wide Cu interconnects with a Mn-doped Cu seed and a Mn-based barrier is studied by means of low-frequency (LF) noise measurements and atom probe tomography (APT). While the EM activation energy of reference interconnects without Mn is 0.8 eV, standard EM tests revealed an activation energy of 1.0 eV for Cu(Mn) interconnects. The LF noise measurements confirm the activation energy of 1.0–1.1 eV in the Cu(Mn) interconnects, but also the activation energy of 0.8 eV is still visible, though less pronounced. Furthermore, the extent to which the mechanism at 0.8 eV is suppressed is strongly subjected to sample variations. These observations are confirmed by APT; Mn is found at the top surface and small clusters of Mn are present in the Cu bulk up to 5 nm away from the sidewalls. Mn segregation at the grain boundaries was not observed such that the hypothesis of Mn blocking grain boundary diffusion cannot be confirmed.The enhanced electromigration (EM) performance of 20 nm-wide Cu interconnects with a Mn-doped Cu seed and a Mn-based barrier is studied by means of low-frequency (LF) noise measurements and atom probe tomography (APT). While the EM activation energy of reference interconnects without Mn is 0.8 eV, standard EM tests revealed an activation energy of 1.0 eV for Cu(Mn) interconnects. The LF noise measurements confirm the activation energy of 1.0–1.1 eV in the Cu(Mn) interconnects, but also the activation energy of 0.8 eV is still visible, though less pronounced. Furthermore, the extent to which the mechanism at 0.8 eV is suppressed is strongly subjected to sample variations. These observations are confirmed by APT; Mn is found at the top surface and small clusters of Mn are present in the Cu bulk up to 5 nm away from the sidewalls. Mn segregation at the grain boundaries was not observed such that the hypothesis of Mn blocking grain boundary diffusion cannot be confirmed.

Electromigration (EM) strongly decreases the reliability of micro-electronics interconnects and becomes more problematic as scaling continues. Remedial measures are required, but therefore EM mechanisms first have to be understood. The standard, accelerated EM test methods are time-consuming, destructive and provide only limited physical understanding. We demonstrate that low-frequency (LF) noise measurements can be used to calculate EM activation energies, making it a fast and non-destructive alternative test method that leads to new insights into the underlying EM mechanisms. More specifically, we show 3 different approaches to calculate activation energies based on LF noise measurements and prove their equivalence. ispartof: pages:1-4 ispartof: 2017 International Conference on Noise and Fluctuations pages:1-4 ispartof: 2017 International Conference on Noise and Fluctuations (ICNF) location:Vilnius, Lithuania date:20 Jun - 23 Jun 2017 status: published

A Fluctuation Scaling (FS) law for the output current in Cu interconnects of 30nm line-width is presented. The scaling exponent p is estimated using the method of expanding bins. For an undamaged sample; a power law exponent p ≈ 2 is observed. Damaging the sample by electromigration induced voiding decreases p to values between 1 and 1.5. The p-exponent reflects the stationary distribution of the output current fluctuations. When 1 < p < 2, it follows a compound Poisson-gamma distribution and for p = 2 a gamma distribution. Current fluctuations in fresh samples thus follow a gamma distribution, whereas for voided samples their distribution is Poisson-gamma. As such it is shown that voids in a Cu line alter the stochastic process by which fluctuations in the output current are generated.

We show that the use of low-frequency (LF) noise measurements as a new, faster technique for electromigration (EM) characterization is not limited to providing EM activation energies (which was previously demonstrated) but can also explain and even predict EM lifetimes of interconnect lines. Two models are proposed: one is to predict EM void nucleation and one to predict void growth times. Predictions can be made for individual interconnects based on the results of non-destructive LF noise measurements, prior to actual EM stress, which is not possible with any other EM test method presently available.

© 2016 IEEE. The application of 1/f noise measurements to speed up electromigration (EM) testing and provide a better understanding of the underlying mechanisms of electromigration in advanced microelectronics interconnects is investigated. It is shown that 1/f noise measurements can be used for early EM damage detection during EM stress, before any changes in the resistance of the sample are observable. Also, the temperature dependence of the low frequency noise is used to calculate activation energies, which are then demonstrated to be similar to values found for electromigration using standard EM tests. Furthermore, the 1/f noise technique is used to assess and compare the EM properties of various advanced integration schemes and different materials. The 1/f noise measurements provide new evidence for the importance of grain boundary diffusion as a dominant EM failure mechanism in highly scaled interconnects. ispartof: pages:1-8 ispartof: 2016 IEEE INTERNATIONAL RELIABILITY PHYSICS SYMPOSIUM (IRPS) vol:2016-September ispartof: IEEE International Reliability Physics Symposium - IRPS location:Pasadena, CA USA date:17 Apr - 21 Apr 2016 status: published

Cu based metal trench is essential for the downscaling of interconnects for 20–10nm technologies, where it is important to optimize the materials and processing on the barrier and cap layer for a promising resistivity and electromigration (EM) performance. In order to shorten the period of process evaluation, a fast wafer-level EM characterization method is highly demanded rather than the conventional package-level EM lifetime test. In this paper, the low frequency noise measurement is proposed to investigate the EM reliability on the advanced copper interconnects. The noise expression was studied by an extensive experimental measurement. It was found that the low frequency noise measurement is a more sensitive method for studying EM degradation and can be an early indicator for EM characterization. In particular, the threshold current (I th ) and the power spectral density (PSD) is suggested to be used as a fast and useful tool for evaluating EM performance.

The use of 1/f noise measurements is explored for the purpose of finding faster techniques for electromigration (EM) characterization in advanced microelectronic interconnects, which also enable a better understanding of its underlying physical mechanisms. Three different applications of 1/f noise for EM characterization are explored. First, whether 1/f noise measurements during EM stress can serve as an early indicator of EM damage. Second, whether the current dependence of the noise power spectral density (PSD) can be used for a qualitative comparison of the defect concentration of different interconnects and consequently also their EM lifetime t50. Third, whether the activation energies obtained from the temperature dependence of the 1/f noise PSD correspond to the activation energies found by means of classic EM tests. In this paper, the 1/f noise technique has been used to assess and compare the EM properties of various advanced integration schemes and different materials, as they are being explored by the industry to enable advanced interconnect scaling. More concrete, different types of copper interconnects and one type of tungsten interconnect are compared. The 1/f noise measurements confirm the excellent electromigration properties of tungsten and demonstrate a dependence of the EM failure mechanism on copper grain size and distribution, where grain boundary diffusion is found to be a dominant failure mechanism. ispartof: Journal of Applied Physics vol:119 issue:18 pages:1-8 status: published