August_EDFA_Digital

edfas.org ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 21 NO. 3 32 In both EDX tomography examples, roughly equal scattering materials are replacing each other (W for Ge, and Ti for Si, respectively). This is easy to discern in the tomographic reconstruction of Figs. 6 and 7, but almost impossible to see as contrast in a normal TEM or STEM image. CONCLUSION The TEM has come a long way from merely imaging devices to evolving into a versatile tool, able to almost completely characterize all aspects of modern front-end and back-end structures. No other technique combines reasonable elemental sensitivity with high spatial resolu- tion like the TEM does. With amodern TEM tool, the FA engineer knows within less than 30minutes what the elemental composition of a defectmight be. PED enables detection of strain and grain orientation, and electron holography can be used tomap doping profiles. The move to 3D imaging and elemental mapping using tomographic techniques allows the FA engineer to gain insight into the nooks and crannies of each and every device and defect on the nanometer scale, without any projection effects. What is on the horizon? 3D EDX and EELS tomography are still in their infancy and will surely be improved and automated. New data processing techniques such as compressed sensing will reduce radiation exposure and increase S/N ratios. New cameras will open the door to measuring electric and magnetic fields on the atomic scale. Finally, ever shrinking dimensions will necessitate the use of critical dimension TEM (CD TEM) for process control of sub-7 nm technologies. All in all, electron microscopy is an exciting area to work in and there are no signs of slowing down in the further evolution of this versatile and fascinating tool. ACKNOWLEDGMENTS The author would like to thank M. Gribelyuk, B. Fu, Y.Y. Wang, and J. Bruley for interesting and inspiring discussions. REFERENCES 1. D.B. Williams and C.B. Carter: Transmission ElectronMicroscopy, 2nd ed., Plenum Press, New York, 1996. 2. H. Tan, et al.: “Advanced Industrial STEMAutomation andMetrology: Boundary of Precision,” 29th Annual SEMI Advanced Semiconductor Manufacturing Conference ( ASMC), IEEE, 2018, p. 131-135. 3. P.A. Midgley and A.S. Eggeman: “Precession Electron Diffraction – A Topical Review,” IUCrJ, 2015 2, p. 126-136. 4. J.Shlens:“ATutorialonPrincipalComponentAnalysis,”2014,[Online: accessed April 2019]. Available: https://arxiv.org/pdf/1404.1100.pdf. 5. F.H. Baumann: “Advanced TEM Imaging Analysis Techniques Applied to BEOL Problems,” Future Fab. Intl., 41, p. 2-12. 6. A. Genç, H. Cheng, J. Winterstein, L. Pullan, and B. Freitag: “3D Chemical Mapping using Tomography with an Enhanced XEDS System,” Microsc. Microanal., 2012, p. 23. 7. F.H. Baumann, J. Miller, B. Rhoads, A. Friedman, and B. Fu: “Characterization of VLSI Processing Defects using STEM-EELS Tomography,” Microsc. Microanal . , 22, (Suppl. 3), 2016. 8. J. Frank: Electron Tomography, Springer, New York, 2006. 9. M. Radermacher: Weighted Back-Projection Methods , PlenumPress, New York, 1992. ABOUT THE AUTHOR Frieder H. Baumann, a native of Namibia, studied physics in Bielefeld and Goettingen, Germany, where he earned his master’s degree and Ph.D. in physics from the University of Goettingen. He was amember of the technical staff for 14 years at Bell Laboratories in New Jersey, focusing on advanced electronmicroscopy, process simulation, and optical MEMS design and fabrication. After Bell Labs, he joined Philips Semiconductors inHamburg, leading the process and device simulation (TCAD) efforts for discrete bipolar devices. Since 2006, he has been a senior engineer in IBM’s microelectronics divi- sion in East Fishkill, N.Y., and worked on device characterization and physical failure analysis using advanced electron microscopic techniques. He is now a principal member of the technical staff at GlobalFoundries in Malta, N.Y.