Roland A. - PVD for microelectronics (779636), страница 66
Текст из файла (страница 66)
Proc. 312:193-198 (1993).9.45. C.W. Kaanta, S. G. Bombardier, W. J. Cote, W. R. Hill, G. Kerszykowski, H. S. Landis, D. J.Pindexter, C. W. Pollard, G. H. Poss, J. G. Ryan, S. Wolff, and J. E. Cronin, "Dual damascene:A ULSI wiring technology," in Technical Proc. of the VLSI Multilevel lnterconnection ConS,pp. 144-152 (1991); see also P. Singer, "Making the move to dual damascene processing,"Semicond. Int., 79-82 (Aug. 1997).9.46. W. M. Posadowski and Z.
J. Radzimski, "Sustained self-sputtering using a direct current magnetron source," J. Vac. Sci. & Tech. All(6): 2980-2984 (1993).9.47. T. Asamaki, R. Mori, and A. Takagi, "Copper self-sputtering by planar magnetron," Jpn J. Appl.Phys.
33:2500-2503 (1994).9.48. N. Hosokowa, T. Tsukada, and H. Kitahara, "Effect of discharge current and sustained self-sputtering," Le Vide, supplement 201 (Proc. of the 8th Int. Vacuum Congress, Cannes, France), pp.11-14 (Sept. 1980).9.49. Z. J. Radzimski, O. E. Hankins, J. J. Cuomo, W. P. Posadowski, and S. Shingubara, "Opticalemission spectroscopy of high density metal plasma formed during magnetron sputtering," J.Vac. Sci. & Tech.
B15(2): 202-208 (1997).9.50. P. C. Zalm, "Some useful yield estimates for ion beam sputtering and ion plating at low bombarding energies," J. Vac. Sci. & Tech. B2(2): 151-152 (1984).9.51. S.-Q. Wang, "Barriers against copper diffusion into silicon and drift through silicon dioxide,"Materials Res. Soc. Bull. 19(8): 30--40(1994).9.52. K.-H. Min, K.-C.
Chun, and K.-B. Kim, "Comparative study of tantalum and tantalum nitrides(Ta,N and TaN) as a diffusion barrier for Cu metallization," J. Vac. Sci. & Tech. B14(5):3263 3269 (1996).9.53. M. Takeyama, A. Noya, T. Sase, A. Ohta, and K. Sasaki, "Properties of T a N films as diffusionbarriers in the thermally stable Cu/Si contact systems," J. Vac.
Sci. & Tech. B4(2): 674--678(1996).9.54. B. Mehrotra and J. Stimmell, "Properties of direct current magnetron reactively sputtered TAN,"J. Vac. Sci. & Tech. B5(6): 1736-1740 (1987).9.55. M Biberger, S. Jackson, and E. Klawuhn, "Low pressure sputtering of copper and related barriers for seed layers and complete planarization," in Technical Proc. of the ISSP Conf, June1997; see also "Processing and integration of copper interconnects," R.
L. Jackson, E.Broadbent, T. Cacouris, A. Harrus, M. Biberger, E. Patton and T. Walsh, Solid State Tech. 59-59(March 1998).9.56. S.M. Rossnagel, C. Nichols, S. Hamaguchi, D. Ruzic, and R. Turkot, "Thin high atomic weightrefractory film deposition for diffusion barrier, adhesion layer and seed layer applications," J.Vac. Sci. & Tech. B14(3): 1819-1827 (1996).9.57. I. Beinglass and M. Naik, "Advanced interconnect and via technology utilizing integration ofCVD AI and PVD AICu," Technical Proc. of SEMICON-Korea session 2, part II, pp. 73-78(February 1997).9.58. G.A.
Dixit, A. Paranjpe, Q.-Z. Hong, L. M. Ting, J. D. Luttmer, R. H. Havemann, D. Paul, A.Morrison, K. Littau, M. Eienberg, and A. K. Sinha, "A novel 0.25/xm via plug process usinglow temperature CVD AIFFiN," Technical Digest of the International Electron DevicesMeeting, pp. 1001-1003 (1995).9.59.
H. Horie, M. Imai, A. ltoh, and Y. Arimoto, "Novel high aspect ratio aluminum plug for352R. POWELLAND S. M. ROSSNAGELlogic/DRAM LSIs using polysilicon-aluminum substitute (PAS)," Technical Digest of theInternational Electron Devices Meeting, pp. 946-948 (1996).9.60. A.J. Babriarz, "Key process controls for underfilling flip chips," Solid State Tech., 77-83 (Apr.1997).9.61. D.
R. Marx, A. Lateef, and A. Clarke, "Sputtering deposits evaporation-quality UBM for flipchip," Semicond. Int., 97-102 (Mar. 1998).Chapter 10 Process Modeling for Magnetron DepositionOver the past 10 years, considerable effort has been put into the computersimulation of the sputter deposition process. The underlying goal has always been to perform a virtual experiment that predicts some aspect of thedeposition process without actually conducting the physical experiment.This goal has been met in some cases, and yet each of the modeling approaches in use has various assumptions or shortcuts that make the resultsless than ideal.
In this chapter we will present a brief overview of some ofthe major approaches to the modeling of sputter deposition. More detailedinformation is available in a recent companion volume [10.1] and otherreferences at the end of this chapter.There are three areas of interest for modeling in sputter deposition, andmodels of each region have been developed. At the cathode surface, it isinteresting to examine the sputtering process as well as the emission or reflection of various other species. In the region between the cathode and thesample, modeling is appropriate to examine the transport and/or scatteringof sputtered atoms as well as geometrical filtering effects due to distanceor collimation.
Finally, at the sample, modeling can be appropriate to examine the topographical nature of the deposition, the microstructure of thefilm, and other related physical characteristics.Models for each of these three areas generally come in two forms: analytic and Monte Carlo. Many of the analytic models are two-dimensional,but some of the Monte Carlo models have been developed in three dimensions. In the first two regions (cathode and transport), the Monte Carlomodels are atomic in nature: The model simulates single particles (mostlyatoms) and deals with simple physical interactions.
At the substrate location, all models of both types treat the film growth process in terms of largegroups of atoms rather than single particles. The physics of these ensembles of particles is generally tied to atomic or traditional solid statephysics, but it is clearly not a first-principles atomic interaction approach.This is mostly a computer-capacity problem, not a physics problem. As anexample, a via with an aspect ratio of 3:1 and a diameter of 0.35 micronshas a net volume of 9.5 x 10 -14 c m 3.
To fill this volume requires 5 x 10 9atoms, plus a similar amount or more on the surrounding field region nearthe via ~ clearly a number that will strain most practical computer models. It is anticipated that the widespread advent of parallel-processingcomputers along with smaller feature dimensions should lead to a morerigorous, atomic approach to film modeling. As an alternative example, afeature 1000-A wide (0.1 microns) is equivalent to roughly 400 atoms inwidth. If one were to consider the deposition of very thin films, such as353354R. POWELL AND S.
M. ROSSNAGELdiffusion barriers that were only a few tens of atoms thick, and also makeuse of the intrinsic symmetries of trenches, the computational problem becomes orders of magnitude more manageable, and it is possible to considera first-principles, single-atom-based model.10.1 Cathode Surface ModelsThe modeling of physical sputtering is generally done for the energy rangeof interest to magnetron sputtering by using a Monte Carlo approach. Themost widely used program is known as TRIM (transport of ions in matter),and this program was originally developed to examine the world of energetic ion implantation (i.e., at much higher energies [10.2-10.4]).
However, the basic physics at low energies are essentially the same, and TRIMhas been widely used to model and predict the sputter yield, the energy dependence of the sputtered atoms, and the angular distribution of the sputtered atoms, as well as the probability, angular, and energy distributions ofinert gas ions that are neutralized and reflected from the surface at moderate energies.As an example, Eckstein has examined the case of 1 keV Ar + onto Ni.The energy distributions of the sputtered Ni atoms are shown in Fig.
10.1at two different angles of incidence: 0 ~ (normal incidence) and 75 ~ whichis close to grazing [10.4]. The simulation shows that the majority of sputtered atoms have energies of under 10 eV and that a high-energy tail existsup to about 200 eV for normal incidence almost all the way to the incidentenergy (1 keV) for grazing bombardment.This general approach to modeling has been quite successful at predicting the various physical properties of the emitted species from the cathode,and the program has been adapted to run on a conventional PC. The resultsfrom TRIM and its many progeny are often now accepted as legitimatedata.
This data can then be used without experimental confirmation is otheraspects of the film deposition process.The TRIM approach is also useful in exploring the physical processesthat occur at the sample surface. For example, a fractal variant of TRIMhas been used to model reflection of depositing metal atoms as a functionof the inclination angle of the surface [10.5-10.6]. TRIM is also appropriate for use in modeling the resputtering of surface atoms during either inertgas ion bombardment of a growing film (e.g., bias sputtering) or energeticdeposition, such as may occur during I-PVD (see Chapter 8). The resultsof the sputtering and ion-impact models are in the form of particle fluxesemitted from the surface with known energy and angular distributions.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
