Ellipsometry Porosimetry for Ultra Low K Material. Dr Dorian ZAHORSKI SOPRA UCLA July 2006

Transcription

1 Ellipsometry Porosimetry for Ultra Low K Material Dr Dorian ZAHORSKI SOPRA UCLA July

2 ULK characterization with EP Excellent dielectric properties of SiO 2 have been the cornerstone for microelectronic revolution during the past 30 years. Technology challenge: reduce RC delay RC = 2 2 2ρkε 4L + L o P2 T 2 ρ - metal resistivity k - relative dielectric constant L - line length P - metal pitch T - metal/dielectric thickness Better Interconnects require: 1. Metals with low resistivity (Cu) 2. Dielectrics with low dielectric constant (low-k) Introduction of porous low K materials 2

3 ULK characterization with EP EP is a unique technology to characterize CVD and spin coated porous ultra low K. EP measures the change of the optical properties and thickness of the materials during adsorption and desorbtion of an organic solvent. Porous material The analysis gives: porosity of the Low K pore size distribution Average pore size For both microporous and mesoporous, cumulative surface area, pore interconnectivity, Young modulus, thickness and refractive index. EP qualifies the sealing layer of both patterned and blanket wafers. + solvant Adrien Darragon - Product Manager, SOPRA sales@sopra-sa.com 3

4 Outline 1. Principle and measurement setup 2. Porosity and Pore size measurements Mesoporous Microporous 3. Young modulus 4. Pore killers and sealing Blanket wafers Patterned wafers Interconnectivity 4

5 Porosity measurement From the thickness and refractive index measurements on the basis of Lorentz- Lorenz equation: From adsorption/desorption isotherms : Porosity pore interconnectivity no need to know the dense skeleton material properties 5

6 Principle of Spectroscopic Ellipsometry linear polarisation elliptical polarisation E i n Ep Es E S E i E P fo E r r p r s E out =r E in Ep rp Es rs Ambient (n0, k0) Thin Film 1 (n1, k1, T1) Thin Film 2 (n2, k2, T2) Substrate (ns, ks) ELLIPSOMETRY is a method based on measurement of the change of the polarisation state of light after reflection at non normal incidence on the surface to study. E out =r E in = (TanY e idelta ) E in Ellipsometry measures TanY and Cos D 6

7 Physical parameters n,k Thickness Experimental Measurement Physical Model With an estimated sample structure TanY Cos D Calculated values - Film Stack and structure - Material n, k, dispersion - Composition Fraction of Mixture TanY Cos D Advanced minimization algorithms for quick convergence Experimental Measurement = Model Simulation? No, proposition for new parameters Yes Film structure and optical properties proposed are correct. 7

8 Porosity measurement Lorentz-Lorenz equations n re : refractive index measured when pores are empty n rl : refractive index measured when pores are filled n ads : refractive index of the solvent 8

9 Porosity measurement n re : refractive index measured when pores are empty n rl : refractive index measured when pores are filled n ads : refractive index of the solvent No need to know the refractive index of the skeleton V is the volume of open pores 9

10 Measurement principle Pressure P/Po=1 (P/Po) (P/Po) Acquisition of SE spectra time Analysis of SE spectra: Calculation of Thickness / Refractive index Thickness vs partial pressure Chemnitz LK3 SOPRA Refractive 633nm vs partial pressure Chemnitz LK3 SOPRA Thickness, µm Relative IPA Pressure Refractive Index Relative IPA Pressure 10

11 Measurement principle Thickness, µm B E Refractive Index C F Relative solvent Pressure 1.30 Relative solvent Pressure Solvent Volume B D Lorentz-Lorentz Equations => Solvent volume vs rel.pressure Porosity = 31% Relative Solvent Pressure 11

12 Meso Pores: Kelvin From pores volume to PSD Pore size with diameter form 2 to 50nm(IUPAC classification) The solvent is in a liquid phase Pore radius : r = 2γVL RT ln( P / P 0 ) Po : equibrium vapor pressure P/Po: relative equilibrium vapor pressure above meniscus g : Surface tension of adsorptive V L : molar volume of adsorptive r k : mean radius of curvature q : wetting angle Solvent Volume B D 0.00 Relative Solvent Pressure 2γVL r= RT.ln( P0/ P 2γV rn= RT.ln( P 2γV r1= RT.ln( P EP gives a distribution of pores for every relative pressure not a distribution a posteriori. 1 n / P 1 n 0) / P 0) 0) dv/d(lnr) Micro Micro TOLUENE Pore Radius (nm) 12

13 Mesoporous material 0.35 Solvent Volume B D From Solvent volume to Pore Size Distribution : Hysteresis in the isotherms=> mesoporous material (r >1nm) Relative solvent Pressure Kelvin dv/d(lnr) Micro Micro TOLUENE Pore Radius (nm) Kelvin 13

14 Mesoporous material Micro Micro TOLUENE dv/d(lnr) Pore Radius (nm) Kelvin The pore radius for the adsorption is 2.5nm (cavities) and the pore radius for desorption is 1.9nm (connections) 14

15 Microporous material 0.30 Micro pores: Dubinin Radushkevich Pore size with diameter below 2nm (IUPAC classification) The solvent is in a gaz phase { [ ] 2 } 2 RT ln( P / P) /( βe ) W / Wo exp O O Pore radius: r = 6/Eo W/Wo: fractional filling of the = micropore volume: Wo β: scaling factor (affinity coeff.) Eo: characteristic energy r: pore radius, nm, normal Distribution, average radius Solvent volume Relative Pressure of Solvent TOLUENE Ln(V) Ln²(p0/p) Microporous total volume : 20.5% Microporous percentage : 73% Microporous average radius : 0.62nm 15

16 Solvent Volume / P/P0 Microporous material 0.30 Solvent volume Mesoporous volume: 3.5% Mesoporous distribution calculated with kelvin equation 0.00 TOLUENE Relative Pressure of Solvent 0.09 Microporous volume: 20% Pore size: 0.75nm Total Porous Volume: microporous volume mesoporous volume Average pore size Pore size distribution dv/d(lnr) Micro Micro TOLUENE Pore Radius (nm) 16

17 Why distribution is important Courtesy of Dr. Shirataki. Asahi Kasei Corporation Same average density however different PSD Random porosity Periodic pore structure 17

18 PSD control for best integration Toluene Volume IPA Volume Relative Toluene Pressure 0.00 Relative IPA Pressure Tail of the distribution is very important as PSD can lead to problems in the integration dv/d(lnr) Low percentage of pores for R > 2 nm dv/d(lnr) Pore Radius (nm) Pore Radius (nm) Wide distribution 18

19 Exemple of porosity IPA Volume B D dv/d(lnr) 10 5 B D 36%, 1.8nm Relative IPA Pressure Pore Radius, nm IPA Volume B D dv/d(lnr) B D 28%, 1.1nm Relative IPA Pressure Pore Radius, nm IPA Volume B D dv/d(lnr) B D 12%, 1.0nm Pore Radius, nm 19

20 Ultra thin Ultra low K: 600A 1.47 Thickness (µm) Relative Methanol Pressure Refractive Index Relative Methanol Pressure Methanol Volume dv/d(lnr) Relative Methanol Pressure Porosity = 19.8 % Pore Radius (nm) Pore radius = 1.0 nm 20

21 BDII - K = Thickness (µm) Relative IPA Pressure Refractive Index Relative IPA Pressure IPA Volume dv/d(lnr) Relative IPA Pressure Open Porosity = 26.4% Pore Radius = 1.1 nm Pore Radius (nm) PALS results: Porosity: 26.6% Pore radius: 1nm 21

22 300mm Mapping EP features: Small spot of 250x250um to achieve edge exclusion better than 2mm Field viewing camera Mapping stage for 300mm and 200mm Sites coordinates are recipe parameters (0,60) (-60,0) (0,0) (60,0) Position Thickness (µm) Porosity (%) Pore Radius (nm) Top ± Left ± Center ± Right ± Bottom ± (0,-60) 22

23 Sealing of porous materials Sealing Techniques Additive or Surface densification Plasma deposition Atomic Layer Deposition Wet Chemicals -By plasma -By e-beam -By Giant Ion Clusters -By UV Ta/TaN Dielectric Polymer TaN WCN TiN etc. Micelles Silanols Thiols Siloxanes Source: M Baklanov - IMEC 23

24 Measurement on Pattern -sealing Sample: Trenches: lines of 140nm 1/1 ratio Sio2 (250nm) /ULK (300nm) /SiC (50nm) P/P0=1 Continuous change in SE spectra during adsorption cycle. The solvent fills in the materials inside the trench (through pore killers) 24

25 Measurement on Pattern -Sealing Sample: Trenches: lines of 140nm 1/1 ratio Sio2 (250nm) /ULK (300nm) /SiC (50nm) P/P0=1 The solvent fills the trenches at saturated pressure Change in Tan Psi only for P/P0 close to 1. The Solvent condenses on the trenches 25

26 Measurement on Pattern Sealing YES Pore sealing treatment is efficient NO Pore sealing treatment is not effective Tan(Psi) Peak Position, ev Tan(Psi) Peak Position, ev Relative Toluene Pressure Change in Tan Psi only for P/P0 >0.8 Relative Toluene Pressure Continuous change in Tan Psi 26

27 Pore sealing The piece of wafer is placed in the vacuum. Initial vacuum 60 sec under Solvent vapor 150 sec under Solvent vapor The solvent diffuses only through the edge and not from the Top of the film. => The porous layer is sealed. 27

28 Pores Sealing If the barrier is fully continuous, toluene cannot penetrate In the case of not fully dense barrier, toluene can penetrate through barrier pinholes and can be adsorbed by the Low K film.. 60 sec under Solvent vapor 150 sec under Solvent vapor => The porous layer is NOT sealed. 28

29 Interconnectivity Source: Shamyrian, Maex, MRS

30 Young Modulus calculation Pressure P/Po=1 (P/Po) (P/Po) Acquisition of SE spectra time Analysis of SE spectra: Calculation of Thickness / Refractive index Thickness, µm Fit of the curve with the Young Laplace equation: T =T 0 k.ln(p/p 0 ) K=T 0 RT/V.E Relative IPA Pressure V=Volume of the solvent molecule 30

31 Young Modulus calculation Effect of nanoindentor on porous low K 31

32 Correlation to electrical K Sample 5 Thickness (nm) Optical Index (633 nm) Porosity (%)Pore Radius (/) (n Low K ± 0.8 n = & k = Silicon n = & k = Sample 6 Thickness (nm) Optical Index (633 nm) Porosity (%)Pore Radius (/) (n Low K ± 1.0 n = & k = Silicon n = & k = Sample 7 Thickness (nm) Optical Index (633 nm) Porosity (%)Pore Radius (/) (n Low K ± 1.0 n = & k = Silicon n = & k = Sample 8 Thickness (nm) Optical Index (633 nm) Porosity (%)Pore Radius (/) (n Porosity (%) Porosity (%) K value Low K ± 1.0 n = & k = Silicon n = & k =

33 Summary table Solvent volume Micro porous TOLUENE Relative Pressure of Solvent dv/d(lnr) Micro Micro TOLUENE Pore Radius (nm) Micro Micro TOLUENE Meso porous IPA Volume RO. dv/d(lnr) Relative IPA Pressure Pore Radius (nm) Bi modal 33

34 Conclusion - EP 1. EP gives quantitative, accurate and repeatable measurement for: Porosity Pore size distribution for meso, microporous and bimodal distribution Film thickness and refractive index Mapping of this parameters Small spot for edge exclusion control 2. EP gives qualitative information for the pore sealing treatment of ULK layers. Measurement of blanket wafers Measurement of patterned wafers ULK behavior in a process integration scheme 34

35 Conclusions - EP "Low dielectric constant materials for microelectronics" K. Maex, M.R. Baklanov, D. Shamiryan and F. Iacopi, S.H. Brongersma, Z.S. Yanovitskaya J. Appl. Phys., Foc. Rev. 93, N11, 1 June 2003, pp

36 Bibliography "Ellipsometric Porosimetry of Porous Low-k Films with Quasi-Closed cavities" Mikhail R. Baklanov, Konstantin P. Mogilnikov, Jin-Heong Yim MRS Spring Meeting, April 2004, San Francisco, USA "Nondestructive Characterization of a Series of Periodic Porous Silica Films by in situ Spectroscopic Ellipsometry in a Vapor Cell C. Negoro, N. Hata, K. Yamada, T. Kikkawa Jap. J. of Appl. Phys. Vol 43 No 4, 2004, pp "cription of the porosity of inhomogeneous porous low-k films using solvent adsorption studied by spectroscopic ellipsometry in the visible range A. Bourgeois, A. Brunet Bruneau, V. Jousseaume, N. Rochat, S. Fisson, B. Demarets, J. Rivory Thin Solid Films , 2004, pp "Investigation of the microporous structure of porous layers using ellipsometric adsorption porometry F.N. Dultsev, Thin Solid Films 458, 2004, pp "Diffusion barrier integrity evaluation by ellipsometry porosimetry" D. Shamiryan, M.R. Baklanov, K. Maex, J. Vac. Sci. Technol. B 21(1), Jan-Feb 2003, pp "Low dielectric constant materials for microelectronics" K. Maex, M.R. Baklanov, D. Shamiryan and F. Iacopi, S.H. Brongersma, Z.S. Yanovitskaya J. Appl. Phys., Foc. Rev. 93, N11, 1 June 2003, pp "Porous Thin Film Metrology for Low-k Dielectric" Extrait de NIST Semiconductor Microelectronics and Nanoelectronics Programs, July 03, pp "Determination of Porosity of TiO 2 Films from reflection Spectra" T. Matsubara, T. Oishi and A. Katagiri J. of the ECS, 149 (2), 2002, pp C89-C93 36

37 CONTACTS For more product information : For any question please contact : Mr Laurent KITZINGER lkitzinger@soprainc.com 37