Outgassing, photoablation and photoionization of organic materials by the electron-impact and photon-impact methods Grace H. Ho*, Yen-H. Huang, Hung-M. Lin, Chih-H. Shao, Chen-Y. Yeh, and Jia-J. Sung Department of Applied Chemistry, National University of Kaohsiung, Nanzih, Kaohsiung 811, Taiwan 1
OUTLINE 1. Introduction 2. Experimental 3. Results 3.1. The Accountability to Measure Gases at NSRRC 3.2. Resist Outgassing 3.3. Electron-impact or Photon-impact Source for Resist Testing? 4. Summary Acknowledgement
1. Introduction - Measuring Gases at NSRRC Detection How good ofit outgassed is to measure species gases from resist: qualitatively Typically using and a quadrupole mass quantitatively spectrometer at (QMS) NSRRC under in Taiwan? the residual gas analyzer (RGA) mode (i. e. by the electron-impact ionization (EI) method). - We measured EI spectra of twelve organic solvents and compared Outgassed our results species with those given by NIST and MSSJ. - = We Dissociated benchmarked ions and our neutrals quantitative liberated outgassing from the results resists into of counting vacuum. resist outgassing to those reported by other institutes. - We conducted outgassing measurement at OOB, EUV, and BEUV wavelengths.
1. Introduction Electron- or Photon-impact Source for Resist Testing? RESIST TESTING TOOLS FOR THE NXE PLATFORM E-gun Electron energy ( ev to ev) Matching EUV photon exposure? Reported by Rupert C. C. Perera, et al. of EUV Tech http://ieuvi.org/twg/resist/12/912/6 Pererra_Resist_TWG_12.pdf Electron- Electron vs. or photon, photon-stimulated do they have outgassing similar reaction (RGA/QMS) patterns? - We used 12 organic solvents as model samples. - Electron- We measured or photon-assist EI and PI spectra contamination the threshold (Witness plate) ev range.
2. Experimental Quadrupole mass spectrometer Vacuum compatible specular reflectometer Light source UV(OOB) from 4B-beamline, 13.5 and 6.7 nm from 8A-beamline at NSRRC in Taiwan. Resist preparation at NCKU. Thin-film metrology (a) specular ellipsometry at NCKU (b) α- step profilometer at NSYSU.
3.1. The Accountability to Measure Gases at NSRRC Branching ratio (%) This work NIST MSSJ 26 27 28 37 38 39 49 51 52 63 73 74 75 76 77 78 79 27 38 39 45 46 51 52 61 62 63 64 65 66 89 9 91 92 93 CH 3 C(O)----R, C 2 H 5 O---R and long-chain aliphatics tends to Branching ratio (%) Mass (amu) Benzene * This work O undergo dissociative pathways. NIST 15 26 27 28 29 39 41 42 43 44 54 55 56 57 58 59 71 72 85 99 114 Mass (amu) Mass (amu) Branching ratio (%) This work NIST MSSJ Vacher Mass (amu) MSSJ Vacher Aromatics and ketones can sustain their structural integrity by Mass (amu) MAK electrons at 7 ev. Our EI spectra are in a good agreement with those given by NIST and MSSJ. How good is good? * Branching ratio (%) This work NIST 15 19 26 27 28 29 31 42 43 44 45 46 47 56 75 3 118 EL Branching ratio (%) Branching ratio (%) This work NIST MSSJ OH OCH 3 Toluene 15 19 27 29 31 39 41 42 43 44 45 46 47 57 58 59 75 9 This work NIST Mass (amu) * PGME PGMEA 14 15 26 27 28 29 31 39 41 42 43 44 45 57 58 59 71 72 73 87 88 892117 * 6
3.1. The Accountability to Measure Gases at NSRRC How good it is to measure gases qualitatively at NSRRC? - Benchmarking most ion features of the EI spectra to those reported by 1.5 NIST Chemistry WebBook 1.5 JMMS MassBank Difference/Average. -1.5 68% 95%. -1.5.1 1....1 1... Absolute branching ratio (%) Instrument-to-instrument (ItI) uncertainty is mainly statistic; i. e. The ItI uncertainty of this work wrt JMSS (Hitachi series) is ~ 35% better than that wrt NIST (unspecified instruments). 7
3.1. The Accountability to Measure Gases at NSRRC How good it is to measure gases quantitatively at NSRRC? UL-SiOCH nm G. H. Ho, et al., JVSTB (12) 5162 Our facility at NSRRC can be one of the limited facilities worldwide capable of measuring resist outgassing in quantity.
3.2. Resist Outgassing - Wavelength Dependency Fragmentation pairs: 69 31() = C 4 H 5,4 O CH 3 O 59(6) 41 = HC(O)OCH 3 C 3 H 5 Wavelength effect on PMMA: Fragmentation partition within the unit of the resin base. 9
3.2. Resist Outgassing - Wavelength Dependency Wavelength effect on PS-tBA, GJ: Fragmentation partition within the deprotection group (tert-butyl). Outgassing from resin backbone is not important. PAG Wavelength effect on the added 5 wt% PAG ((C 6 H 5 ) 3 SC 4 F 9 SO 3 ), GJH: PAG outgassing is in the order of BEUV > EUV > OOB.
3.2. Resist Outgassing - Thickness Dependency 13.5 nm 6.7 nm Thickness dependency (% - nm -1 ) ΔIntensity Mean Intensity Thickness 11
3.2. Resist Outgassing - Thickness Dependency Linear or curvy dependence of resist outgassing on thicknesses??? Linear dependency is more likely to explain what we have found: - Outgassing is structural and absorption dependent. Thickness effect assessment: 15 nm and 8 nm films.4%-nm -1 ΔT ~ 26% 12
3.3. Electron-impact or Photon-impact Source for Resist Testing? Absolute branching ratio (ABR) of molecular and dissociative ionization over an extensive energy range Toluene MAK 9 EI EI ABR (%) ABR (%) 8 7 6 Solid line - This work Dash line - Vacher, et al., CPL (7) -MSSJ -NIST Electron-impact energy (ev) 9 8 7 6 C 7 H 7 + + H Toluene + C 5 H 5 + + C 2 H 3 PI Toluene Dash line - Shaw, et al., CP (1998) and - Stebbings and Tayloer, IJMS (1972) C 7 H 7 + + H Toluene + C 5 H 5 + + C 2 H 3 ABR (%) Electron-impact energy (ev) CH 3 CO + + C 5 H 11 CH 3 C(O)CH 3 + +C 4 H 8 Dash line - Vacher, et al., IJMS (5) C 5 H 11 + CH 3 CO + MAK + Very few studies are available for the comparison. Our results agree with the best available values. The extent of dissociative ionization varies smoothly and slowly when the excitation source is above ~ ev. Photo-impact energy (ev) 13 13
ABR (%) ABR (%) 3.3. Electron-impact or Photon-impact Source for Resist Testing? PI vs EI: Aromatics 9 8 7 6 C 5 H 5 + + C 2 H 3 Energy (ev) 9 8 7 6 Solid line PI Dash line EI C 6 H 5 + + H Energy (ev) Toluene C 7 H 7 + + H Toluene + Benzene Benzene + C 4 H 3 + + C 2 H 3 e - e - + Molecular ions: EI (less fragmentation) > PI Benzene has a negative electron Up affinity. to ev assessment about EI replacing PI for resist outgassing testing from aromatic moieties: - Less fragmentation by EI = Less outgassing by EI e - Under-estimate outgassing with respect to the real case using a photon-impact source? J.A.R. Samson, PRL (199) 14 14
3.3. Electron-impact or Photon-impact Source for Resist Testing? ABR (%) ABR (%) ABR (%) PI vs EI: few more examples 9 PGMEA 8 7 6 CH 3 CO + + C 4 H 9 O 2 CH 3 OCH + 2 + C 4 H 7 O 2 CH + 3 + C 5 H 9 O 3 Energy (ev) EL 9 8 7 6 C 2 H 5 + + C 3 H 5 O 3 Energy (ev) C-HXK + Energy (ev) C 2 H 5 O + + C 3 H 5 O 2 C 3 H 7 O 2 + + C 2 H 3 O C-HXK C 4 H 7 + + CH 3 CO CH 3 CH=CH 2 + + C 3 H 4 O Ketones can also sustain its molecular integrity but in a less extent than aromatics can. Molecular ions + direct dissociative ionization: PI (less fragmentation) > EI - Ketones (C-HXK, MAK, acetone) - Alcohols and ether (Ethanol, CH 3 OCH 3 ) - Esters (PGME, PGMEA, EL, Butylacetate, MMP) The fragmentation pattern by EI and PI sources can be alike (PGMEA), different (cyclo-hexanone, C-HXK), or similar (EL). Not easy to predict what if EI replaces PI for resist testing. 15
Summary The accountability to measure gases qualitatively and quantitatively. - What if the NSRRC site becomes one of the certified site for the outgassing study? Resist outgassing as functions of wavelengths, film thickness, and resist compositions. - Fragmentation partition evolved. - Weak thickness but structural and absorption dependency. We have investigated electron-impact and photon-impact ionization over an extensive energy range. - EI vs. PI, alike or dis-alike? 16 16
Acknowledgement Financial support by National Science Council in Taiwan contract Nos. NSC99-2113 -M-39-3, NSC-21 -M-9-, and 1-21-M-9-7. Nissan Chemical Industries, Ltd. and DuPont are thanked for providing free samples. 17 17