Atomic layer deposition (ALD) is a layer by layer chemical vapor deposition (CVD) technique based on alternating purge-separated self-limiting surface reactions. ALD offers inherent atomic scale controlled growth of high quality conformal thin films at relatively low temperatures. A low thermal budget is often critical in microelectronics processing, such as for back-end-of-line processing, 3D integration, deposition on glass or flexible substrates for large area electronics, avoiding unwanted diffusion, and maintaining stable flat band voltages, effective work functions, threshold voltages, and effective oxide thickness in metal/insulator/metal (MIM) and metal/oxide/semiconductor (MOS) device structures. However, low deposition temperature can lead to incorporation of excess -OH groups or other residual impurities from unreacted ligands and result in poor stoichiometry of ALD films. Poor stoichiometry may in turn lead to sub-optimal physical, optical, and electrical properties. A number of approaches may be used to reduce impurities, increase density, improve properties, and achieve the desired morphology of ALD films. An obvious approach is to increase deposition temperature, however this may move a process out of the ALD window into the CVD regime, negating many of the benefits of ALD. The most common approach is post-deposition annealing (PDA) at elevated temperatures. The PDA temperatures that are typically required however, can exceed the maximum temperature limitations of the substrate or previously formed electronics. To maintain the low thermal budget of ALD while maximizing film properties, performing annealing during, rather than after, deposition can be beneficial. An alternate approach to help drive reactions and reduce impurity / ligand incorporation is thus to add extra energy as part of each (or every few) ALD supercycles. Methods to date include in-situ rapid thermal (MTA, DADA, etc.) annealing, flash lamp annealing, plasma exposure, and UV exposure. I will collectedly refer to these techniques as energy enhanced ALD (EEALD). [1-13] (Note that these methods are distinct from plasma enhanced ALD (PEALD), an excellent review of which can be found in [14].) Documented benefits of the various forms of EEALD include higher GPC, denser films, lower temperature, improved dielectric constant and refractive index, lower leakage, lower residual impurities. A potential downside of energy enhancement is the additional time required to implement these steps into the ALD super-cycle, particularly cool down time when in-situ annealing is incorporated. This invited talk will describe, compare, and contrast these various techniques focusing on mechanisms (thermal vs. chemical), placement in ALD supercycle, benefits and drawbacks, and challenges to be addressed in finding the ideal EE-ALD technique. Finally, I will introduce an entirely new method of EE-ALD. J.F. Conley, Jr., Y. Ono, D.J. Tweet, Appl. Phys. Lett. 84(11), 1913-1915 (2004).J.F. Conley, Jr., D.J. Tweet, Y. Ono, and G. Stecker, in High-k Insulators and Ferroelectrics for Advanced Microelectronic Devices, R.M. Wallace, D. Landheer, M. Houssa, and J. Morais, eds., MRS Proc. Vol. 811, 5 (2004).J.F. Conley, Jr., Y. Ono, D. Tweet, G. Stecker, R. Solanki, and W.W. Zhuang, in Physics and Technology of High-k Gate Diectrics II, S. Kar, R. Singh, D. Misra, H. Iwai, M. Houssa, J. Morais, and D. Landheer, eds., ECS Proc. Vol. 2003-22, 11 pgs.K.H. Holden, S.M. Witsel, P.C. Lemaire, and J.F. Conley, Jr. in preparation (2022).Henke et al., ECS J. Sol. Sta. Sci. Tech. 4(7), 277 (2015).R.D. Clark et al., ECS Transactions, 41(2), 79 (2011).Miikkulainen et al., ECS Tran. 80(3), 49 (2017).Chalker et al., ECS Tran. 69(7), 139 (2015).Kwak, Y.-H. Lee, and B.-H. Choi, Appl. Surf. Sci. 230, 249 (2004).Kim et al., Electrochemical and Solid-State Lett. 14(4), H146 (2011)No et al., J. ECS 153(6), F87 (2006).Shih et al., Sci. Rep. 7(1), 39717 (2017).Österlund et al. Vac. Sci. Tech. A 39, 032403 (2021).Ueda et al., Appl. Surf. Sci. 554, 149656 (2021).Profijt, et al., J. Vac. Sci. Technol. A 29(5), 050801 (2011).
Read full abstract