Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23452
J. Maiz, I. Segura
A DC resistometric method has been developed that provides a substantial reduction in electromigration test times of Al-Si thin films. The technique has been used to detect relative resistance changes in the IE-4 to IE-2 range. A simulation has been performed showing that resistance changes of this magnitude can be produced by the presence of small voids. The kinetics of the individual void nucleation and growth is believed to prevent the observation of constant rates of resistance change for individual units. The problem can be eliminated by the use of distributions where the rates of resistance change are constant for a total increase of up to 1%. Measurements of the activation energy and the current density dependence made by using the proposed methodology are presented. The correlation between the median-time-to-failure (MTF) and the rate of resistance change has been investigated and shows an almost inverse relationship for the range tested. Time-reduction factors of 20 to 50 appear practical with this technique, which should allow an easier evaluation of the electromigration under low-stress conditions like low J, low T, and bidirectional and pulsed currents.<>
{"title":"A resistance change methodology for the study of electromigration in Al-Si interconnects","authors":"J. Maiz, I. Segura","doi":"10.1109/RELPHY.1988.23452","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23452","url":null,"abstract":"A DC resistometric method has been developed that provides a substantial reduction in electromigration test times of Al-Si thin films. The technique has been used to detect relative resistance changes in the IE-4 to IE-2 range. A simulation has been performed showing that resistance changes of this magnitude can be produced by the presence of small voids. The kinetics of the individual void nucleation and growth is believed to prevent the observation of constant rates of resistance change for individual units. The problem can be eliminated by the use of distributions where the rates of resistance change are constant for a total increase of up to 1%. Measurements of the activation energy and the current density dependence made by using the proposed methodology are presented. The correlation between the median-time-to-failure (MTF) and the rate of resistance change has been investigated and shows an almost inverse relationship for the range tested. Time-reduction factors of 20 to 50 appear practical with this technique, which should allow an easier evaluation of the electromigration under low-stress conditions like low J, low T, and bidirectional and pulsed currents.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128977024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23445
P. Hefley, J. McPherson
Data retention studies were conducted on nonvolatile devices exposed to an external sodium diffusion source (sodium bicarbonate). A time-to-failure model is presented and the kinetics of failure are discussed. In this mobile ion contamination study, the integrity of the passivation overcoat (PO) was identified as the major factor affecting data loss. Even under a heavy concentration of sodium (3 wt.% Na sol.), pinhole-free PO units baked at 300 degrees C were observed to fail due to intrinsic charge loss before sodium compensation of the floating gate occurred. Units which had anomalies in the PO were found to fail rapidly with localized regions of failing bits appearing as circular areas. These circles of failed bits were found to grow in radius with the square root of time. Arrhenius plots of the rate of circular growth yielded an average value of 1.8 eV for the activation energy.<>
{"title":"The impact of an external sodium diffusion source on the reliability of MOS circuitry","authors":"P. Hefley, J. McPherson","doi":"10.1109/RELPHY.1988.23445","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23445","url":null,"abstract":"Data retention studies were conducted on nonvolatile devices exposed to an external sodium diffusion source (sodium bicarbonate). A time-to-failure model is presented and the kinetics of failure are discussed. In this mobile ion contamination study, the integrity of the passivation overcoat (PO) was identified as the major factor affecting data loss. Even under a heavy concentration of sodium (3 wt.% Na sol.), pinhole-free PO units baked at 300 degrees C were observed to fail due to intrinsic charge loss before sodium compensation of the floating gate occurred. Units which had anomalies in the PO were found to fail rapidly with localized regions of failing bits appearing as circular areas. These circles of failed bits were found to grow in radius with the square root of time. Arrhenius plots of the rate of circular growth yielded an average value of 1.8 eV for the activation energy.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129434372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23429
H. Kitagawa, T. Maeda, S. Murata, T. Maki, T. Kaeriyama, A. Hyslop, A. Nishimura
The effects of long-term temperature/humidity stress on a TiW/Al-Si-TiSi/sub 2/ metal contact system were investigated. Plastic packaged DRAM test sites were subjected to conventional unbiased humidity testing procedures which induced electrical opens. Failure analysis showed these opens to occur in the contact region. The nature of this phenomenon was studied. It was found that plastic-package-induced shear stress causes plastic deformation of Al-Si and forces it into the contact hole. The mass built up inside the contact hole increases, causing an alloy formation, and can eventually lift the metal from the diffusion surface, causing the electrical open circuit to occur. One proposed solution is to enhance the protective overcoat's ability to absorb this stress and thus minimize its transmittal to the metallization system.<>
{"title":"Plastic packaging stress induced failures in TiW/Al-Si metal to silicide contacts","authors":"H. Kitagawa, T. Maeda, S. Murata, T. Maki, T. Kaeriyama, A. Hyslop, A. Nishimura","doi":"10.1109/RELPHY.1988.23429","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23429","url":null,"abstract":"The effects of long-term temperature/humidity stress on a TiW/Al-Si-TiSi/sub 2/ metal contact system were investigated. Plastic packaged DRAM test sites were subjected to conventional unbiased humidity testing procedures which induced electrical opens. Failure analysis showed these opens to occur in the contact region. The nature of this phenomenon was studied. It was found that plastic-package-induced shear stress causes plastic deformation of Al-Si and forces it into the contact hole. The mass built up inside the contact hole increases, causing an alloy formation, and can eventually lift the metal from the diffusion surface, causing the electrical open circuit to occur. One proposed solution is to enhance the protective overcoat's ability to absorb this stress and thus minimize its transmittal to the metallization system.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131774410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23431
R. Lin, E. Blackshear, P. Serisky
The process compatibility and potential component reliability problems associated with surface-mount technology are investigated. Major reliability concerns observed during qualification of vendors' surface mount ICs are reported. Moisture-induced package cracking and its implication for the corrosion resistance of the package are discussed. The application of acoustic microscopy techniques such as scanning laser acoustic microscopy (SLAM) and C-mode scanning acoustic microscopy (C-SAM) for examining internal package cracking is described. The concept of threshold (safe) moisture level in a surface-mount package and the proper procedures for determining it are discussed.<>
{"title":"Moisture induced package cracking in plastic encapsulated surface mount components during solder reflow process","authors":"R. Lin, E. Blackshear, P. Serisky","doi":"10.1109/RELPHY.1988.23431","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23431","url":null,"abstract":"The process compatibility and potential component reliability problems associated with surface-mount technology are investigated. Major reliability concerns observed during qualification of vendors' surface mount ICs are reported. Moisture-induced package cracking and its implication for the corrosion resistance of the package are discussed. The application of acoustic microscopy techniques such as scanning laser acoustic microscopy (SLAM) and C-mode scanning acoustic microscopy (C-SAM) for examining internal package cracking is described. The concept of threshold (safe) moisture level in a surface-mount package and the proper procedures for determining it are discussed.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122403501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23428
T.B. Ching, W. Schroen
Bond pads typically have oxides below the metallization which act to insulate the pads from the substrate. These oxides are grown or deposited as part of the primary front-end process of the device. Different oxides are seen to have varying tolerance to the associated mechanical loading and ultrasonic scrubbing, resulting in bond pad cracking. The front-end process can thus influence bond quality. A study was done to characterize various oxides and also different metal systems consisting of aluminum and titanium tungsten alloy. The proper choice of metallization was found to give significant improvement in tolerance of bond-process-induced stresses.<>
{"title":"Bond pad structure reliability","authors":"T.B. Ching, W. Schroen","doi":"10.1109/RELPHY.1988.23428","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23428","url":null,"abstract":"Bond pads typically have oxides below the metallization which act to insulate the pads from the substrate. These oxides are grown or deposited as part of the primary front-end process of the device. Different oxides are seen to have varying tolerance to the associated mechanical loading and ultrasonic scrubbing, resulting in bond pad cracking. The front-end process can thus influence bond quality. A study was done to characterize various oxides and also different metal systems consisting of aluminum and titanium tungsten alloy. The proper choice of metallization was found to give significant improvement in tolerance of bond-process-induced stresses.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133061766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23443
F. Wulf, D. Braunig, W. Nickel
Similarities between bias-temperature (BT) and ionizing radiation stress results suggest a common microscopic nature of defects for oxide charge build-up and annealing. Therefore, experiments were performed to check the influence of technology and experimental conditions. A model based on chemical reaction between silicon and hydrogen can be used as a basis for understanding BT stress results and utilizing ionizing radiation as a rapid reliability screening tool.<>
{"title":"Reliability prediction of MOS devices: experiments and model for charge build up and annealing","authors":"F. Wulf, D. Braunig, W. Nickel","doi":"10.1109/RELPHY.1988.23443","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23443","url":null,"abstract":"Similarities between bias-temperature (BT) and ionizing radiation stress results suggest a common microscopic nature of defects for oxide charge build-up and annealing. Therefore, experiments were performed to check the influence of technology and experimental conditions. A model based on chemical reaction between silicon and hydrogen can be used as a basis for understanding BT stress results and utilizing ionizing radiation as a rapid reliability screening tool.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124646849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23430
R. E. Lund
An analysis and documentation method used to expose fibrous gold-silicon particle shorts was developed. The particles are a consequence of the die bond scrub-in being performed in an oxidizing atmosphere. Initially attached to the substrate, the fibers become detached during vibration or temperature cycling and cause intermittent shorts. It is shown that this particle generating tendency can be eliminated by using an adequate amount of cover gas concentration, which eliminates oxidation of the silicon. The bonding pressure, amplitude, duration, and temperature are other process parameters requiring optimization.<>
{"title":"Gold-silicon fiber shorts in VLSI devices","authors":"R. E. Lund","doi":"10.1109/RELPHY.1988.23430","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23430","url":null,"abstract":"An analysis and documentation method used to expose fibrous gold-silicon particle shorts was developed. The particles are a consequence of the die bond scrub-in being performed in an oxidizing atmosphere. Initially attached to the substrate, the fibers become detached during vibration or temperature cycling and cause intermittent shorts. It is shown that this particle generating tendency can be eliminated by using an adequate amount of cover gas concentration, which eliminates oxidation of the silicon. The bonding pressure, amplitude, duration, and temperature are other process parameters requiring optimization.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129228899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23439
T. R. Conrad, R.J. Mielnik, L. S. Musolino
A description is given of a test methodology called operational life testing (OLT), which has been implemented to monitor and quantify the early-life reliability of selected semiconductor technologies and identify early-life failure mechanisms. This monitor measures the effectiveness of screens and tests used to remove device infant-mortality failure modes. In addition, the early-life reliability monitor complements the data derived from highly accelerated long-term reliability tests since it highlights specific failure modes which are not predominant in highly accelerated long-term reliability tests. Information gained from the monitor can be used to implement tests and screens designed to eliminate certain failure modes in a more timely manner than accumulating and analyzing field return data.<>
{"title":"A test methodology to monitor and predict early life reliability failure mechanisms","authors":"T. R. Conrad, R.J. Mielnik, L. S. Musolino","doi":"10.1109/RELPHY.1988.23439","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23439","url":null,"abstract":"A description is given of a test methodology called operational life testing (OLT), which has been implemented to monitor and quantify the early-life reliability of selected semiconductor technologies and identify early-life failure mechanisms. This monitor measures the effectiveness of screens and tests used to remove device infant-mortality failure modes. In addition, the early-life reliability monitor complements the data derived from highly accelerated long-term reliability tests since it highlights specific failure modes which are not predominant in highly accelerated long-term reliability tests. Information gained from the monitor can be used to implement tests and screens designed to eliminate certain failure modes in a more timely manner than accumulating and analyzing field return data.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128256345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23447
J. Ondrusek, A. Nishimura, H. Hoang, T. Sugiura, R. Blumenthal, H. Kitagawa, J. McPherson
Electromigration (EM) in Ti:W/Al-1%Si two-layered and Ti/W/Al-1%Si three-layered metallizations was investigated. The activation energy and the EM performance were studied as functions of the change in resistance which results from EM damage. The effective activation energy was found to decrease monotonically with increasing test structure resistance under stress. The temperature coefficient of resistance supports a void formation model in the Al-Si layer as the resistance rise mechanism for the Ti:W/Al-Si multilayered film. An empirical form of the Black equation which incorporates resistance rise is proposed.<>
{"title":"Effective kinetic variations with stress duration for multilayered metallizations","authors":"J. Ondrusek, A. Nishimura, H. Hoang, T. Sugiura, R. Blumenthal, H. Kitagawa, J. McPherson","doi":"10.1109/RELPHY.1988.23447","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23447","url":null,"abstract":"Electromigration (EM) in Ti:W/Al-1%Si two-layered and Ti/W/Al-1%Si three-layered metallizations was investigated. The activation energy and the EM performance were studied as functions of the change in resistance which results from EM damage. The effective activation energy was found to decrease monotonically with increasing test structure resistance under stress. The temperature coefficient of resistance supports a void formation model in the Al-Si layer as the resistance rise mechanism for the Ti:W/Al-Si multilayered film. An empirical form of the Black equation which incorporates resistance rise is proposed.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123754563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-04-12DOI: 10.1109/RELPHY.1988.23425
S. Fan, J. McPherson
A wafer-level test for chlorine-induced corrosion is presented for VLSI multilayered metallization. The intrinsic corrosion rate for an arbitrary metallization can be determined by monitoring the electrical resistance rise when the test structure is stored in a fixed corrosive environment. The corrosion activity (rate constant) is found to be directly proportional to the amount of chlorine present on the slice and therefore can be used as an effective in-line process monitor for chlorine-based Al-alloy dry etches. The temperature dependence of the corrosion rate is described by an Arrhenius relationship with an activation energy of 0.3-0.4 eV. The corrosion activity is observed to be time-dependent, i.e. very low initially but rising rapidly to a maximum value, from which it decays slowly with time.<>
{"title":"A wafer-level corrosion susceptibility test for multilayered metallization","authors":"S. Fan, J. McPherson","doi":"10.1109/RELPHY.1988.23425","DOIUrl":"https://doi.org/10.1109/RELPHY.1988.23425","url":null,"abstract":"A wafer-level test for chlorine-induced corrosion is presented for VLSI multilayered metallization. The intrinsic corrosion rate for an arbitrary metallization can be determined by monitoring the electrical resistance rise when the test structure is stored in a fixed corrosive environment. The corrosion activity (rate constant) is found to be directly proportional to the amount of chlorine present on the slice and therefore can be used as an effective in-line process monitor for chlorine-based Al-alloy dry etches. The temperature dependence of the corrosion rate is described by an Arrhenius relationship with an activation energy of 0.3-0.4 eV. The corrosion activity is observed to be time-dependent, i.e. very low initially but rising rapidly to a maximum value, from which it decays slowly with time.<<ETX>>","PeriodicalId":102187,"journal":{"name":"26th Annual Proceedings Reliability Physics Symposium 1988","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114707194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}