Comparison between XAS (Xanes-Exafs), AWAXS, ASAXS and DAFS applied to nanometer scale supported metallic clusters. Definition of a methodology D. Bazin 1 Plan I. Introduction I-1. Nanometer scale metallic clusters. I-2. Structural parameters : N & R. II. X-ray Absorption Spectroscopy II-1. K edge of 3d transition metals. II-2. L edge of 5d transition metals. II-3. Extended X-ray Absorption Fine Structure (Exafs). II-3a. Monometallic clusters & II-3b Bimetallic clusters. II-4 Comparison between Xanes and Exafs. III. Anomalous Wide Angle X-ray Scattering III-1. Generalities III-2. Monometallic clusters III-3. Comparison between Xas and Awaxs III-4. Size distribution & morphology III.-5. Bimetallic clusters. IV/ Anomalous Small Angle X-ray Scattering IV-1. Monometallic clusters IV-2. Bimetallic clusters. V/ Generalities on Diffraction Anomalous Fine Structure V-1. Generalities V-2. First DAFS experimental results at the ESRF on metallic particles. VI. Summary VII. Future prospects 2 I.1 Nanometer scale metallic clusters I Monometallic clusters Morphology : Icosahedra are preferred to cubooctahedra, Relaxation : Large compressions of the central atom are observed for the icos geometries (6%) while small core compressions are found for the fcc geometries (1%) Bimetallic clusters Distribution of the metals inside the cluster : statistical distribution or core/shell distribution Interaction between the cluster and the support 3 I.2 Structural parameters N & R I Monometallic cluster Coordinations (N1,N2,N3,N4) --> Size & Morphology Interatomic distances --> Compression/dilatation Bimetallic cluster Coordination (NAA, NAB, NBA, NBB) --> Segregation Very high reactivity --> In situ methods to preserve the physico chemistry integrity of the metallic cluster • In-situ spectroscopy of catalysts Ed. B.M. Weckhuysen, American Scientific Publishers,2004. • ”Xas for catalysts and surfaces “, Ed. Y. Iwasawa,World Scientific, 1996. • X-ray absorption: principles, applications, techniques of EXAFS, SEXAFS, and XANES. Ed. DC Koningsberger, R Prins, Wiley, 1988. 4 I.3 L’outil d’investigation : une source de rayons X I 100000 km/s --> 1000km/s Brillance Nb de photons/cône Tube : 107 Synchrotron : 1020 la fluorescence X (identifier les éléments) ou la diffraction X (identifier les phases) Temps d’acquisition, Cartographie, Sensibilité la spectroscopie d’absorption X. Spéciation 5 1.4 R.S. à travers le monde & Recherche finalisée Denmark: France: ISA (Aarhus). ESRF(Grenoble), Soleil (Orsay). Germany: ANKA (Karlsruhe), BESSY (Berlin), DELTA (Dortmund), ELSA (Bonn), HASYLAB (Hamburg). Italy: Elettra (Trieste). Spain: LLS (Barcelona). Sweden: MAX (Lund). Switzerland: SLS (PSI) (Villigen). U.K.: Diamond (Didcot), SRS (Daresbury). Brazil: Canada: USA: LNLS (Campinas SP). CLS (Saskatoon). ALS (Berkeley CA), APS (Argonne IL), CAMD (Baton Rouge LA), DFELL (Durham NC), CHESS (Ithaca NY), SLS (Upton NY), SRC (Madison WI), SSRL (Stanford, CA) ,SURF II (Gaithersburg MD). China (PR): BSRF (Beijing). India: INDUS (Indore). Japan: Nano-Hana (Ichihara), Photon Factory (Tsukuba), SPring-8 (Nishi Harima). Russia: SSRC (BINP) (Novosibirsk). South Korea: Pohang Acc. Lab (Pohang). Taiwan: SRRC (Hsinchu). Australia: Australian Syn. (Melbourne). 6 II.1 X-ray Absorption Spectroscopy II E µx(E) =log(Ι0/Ι1) I0 S I1 2,5 Pt, L III 2 1,5 Absorption 1 0,5 2p3/2 2p1/2 2s 0 11400 11600 11800 12000 12200 12400 Energy (eV) Sayers D. A., Lytle F. W. and Stern E. A., Advances in X-ray Analysis, (Ed. Plenum, New-York, 13, 1970). 1s 7 II.2 The associated formula II 2,5 Pt, L III 2 0,3 Pt, L III edge 1,5 0,2 Absorption 1 0,1 0,5 0 -0,1 0 11400 11600 11800 12000 12200 12400 Energy (eV) -0,2 -0,3 2 4 6 8 10 12 14 16 -1 k( χ(k) = Σterms j Nj/kR Electronic j fj(k), φj(k), λ 2 ) Structural jterms fj(k)exp(-R /λ)exp(-2σj2k2)sin(2kRj +φj(k)) Nj , Rj , σj. R(0)-a Πτ N σ 0 1 2 3 4 5 6 Ρ(0 ) Modulus of the Fourier Transform or Pseudo radial distribution function 8 II.3 Nanometer scale metallic cluster & Xas II 25 L N1 N2 N3 N4 20 III 15 Pt Au 10 Absorption (a.u.) 5 0 0 1 2 3 4 5 6 7 Diameter (nm) -50 0 50 E-E0(eV) 100 • D. Bazin, D. Sayers, J. Rehr, Comparison between Xas, Awaxs, Asaxs & Dafs applied to nanometer scale metallic clusters. J. Phys. Chem. B 101, 11040 (1997). • D. Bazin, D. Sayers, J. Rehr, C. Mottet Numerical simulation of the Pt LIII edge white line relative to nanometer scale clusters, J. of Phys. Chem. B 101, 5332 (1997). • D. Bazin, J. Rehr, Limits and advantages of X-ray absorption near edge structure for nanometer scale metallic clusters. J. Phys. Chem. B 107, 12398 (2003). 9 II.4a Some examples II 1%wt Pt/oxide Impregnation Cl _ H2PtCl6 .. Pt/Y-SiO 2 Modules de T.F. (u.a.) _ PtO2 Pt 0 1 2 3 5 R(F) 6 .. Pt/Y-SiO 2 Modules de la T.F. (u.a.) O Pt O O 0 1 Pt Pt Pt Pt Pt Pt Pt Reduction 4 _ H2PtCl6 O Calcination Xas studies of bimetallic Pt-Re(Rh)/Al2O3 catalysts in the first stages of preparation. D. Bazin et al., J. of Cat. 110, 209 (1988). • Bimetallic reforming catalysts : Xas of the particle growing process during the reduction. D. Bazin et al., J. Cat. 123, 86 (1990). • In situ high temperature and high pressure Exafs studies of Pt/ Al2O3 catalysts : Part I. N. S. Guyot-Sionnest et al., Cat. Let 8, 283 (1991). • In situ high temperature and high pressure Exafs studies of Pt/ Al2O3 catalysts : Part II. N. S. Guyot-Sionnest et al., Cat. Let 8, 297 (1991). • Investigation of dispersion and localisation of Pt species in mazzite using Exafs. A. Khodakov et al. J. of Phys. Chem. 101, 766 (1996). • In situ study by Xas of the sulfuration of industrial catalysts : the Pt & PtRe/Al2O3 system. A. Bensaddik et al., Applied Cat. A 162, 171, (1997). • Xas of electronic state correlations during the reduction of the bimetallic PtRe/Al2O3 system. D. Bazin et al., J. of Synchrotron Radiation 6, 465, 1999. • Influence of the H2S/H2 ratio and the temperature on the local order of Pd atoms in the case of a highly dispersed multimetallic catalyst : Pd-Ni-Mo/Al2O3. D. Bazin et al., J. de Physique IV, 12-6, 379, (2002). • Structure & size of bimetallic PtPd clusters in an hydrotreatment catalyst. D. Bazin et al., Accepted in Oil & Gas Science and Technology – Rev. IFP • Pt, L III _ PtO2 2 3 4 __ PtO 2 5 R(0) 6 Modules de la T.F. (u.a.) ... Pt/Y-SiO 2 _ Pt mŽtal R(0)-a Πτ N Reforming …Pt, PtPd, PtRh, PtMo,PtSn,PtRe, PtIr, Fischer–Tropsch, Co, CoPt, CoPd, CoRu, σ 0 0 1 Modulus of the Fourier Transform or Pseudo radial distribution function 2 3 4 5 6 Ρ(0) 1 2 3 4 5 R(0 ) 6 10 II.4b Experimental set up 4 3 1 2 3 5 Catalyst Water cooling X-ray beam Quick disconnect for gas flow 5 3 2 Thermocouple Catalyst Gas input Gas output Figure 1 : Partial view of the sample holder 1, 2 : Boron nitride sample holder and cover plate 3 : Carbon foil for gas leakage 4,5 : Stainless steel mounting block and cover plate Sample holder Figure 2 : General view of the sample holder QuickTimeª et un dŽcompresseur Photo - JPEG sont requis pour visualiser cette image. 11 II.4c Some examples II Pt/Al2O3 H2 T=450°C NPtO=6 2,5 Pt, L III 2 1,5 Absorption 1 0,5 0 11400 11600 11800 12000 12200 12400 Energy (eV) NPtO= 6. NReO=4.1 NPtO=0.4 NPtPt=4.5 NPtO=0.3 NPtMet=3.9 NReO=0.9 NReMet=3.7 NPtS=2.8 NPtS=2.0 NPtMet=4.2 NReS=2.4 NReMet=4.5 1%H2S/H2 T=450°C H2 T=450°C PtRe/Al2O3 NPtS=2.8 NPtPt=1.5 NPtS=1.0 NPtMet=4.0 NReS=2.0 NReMet=4.7 12 II.5a Xanes K edge of 3d transition metals : Size II In the case of a cluster, it is essential to consider each kind of atom since the signal coming from the surface and the central atom are definitely not the same. A 13 Cu environment is not enough to produce the resonance C. Xanes : size effect B C Cu metal Absorption (u.a.) 55 Cu 13 Cu -10 0 10 20 30 40 50 60 E(eV) • For nanometer scale materials, it is definitely not possible to simulate their Xanes part for K edge with a linear combination of the Xanes of well crystallised reference compounds. G. N. Greaves et al., Nature 294,139 (1981). D. Bazin et al. J. de Physique C4-481, 6 (1996). 13 II.5b Xanes L edge of 5d transition metals II L III 43Pt 19Pt 13Pt From an experimental point of view, the LIII white line is at the centre of electronic charge transfer between either the nanometer scale metallic particle and the support or between the two metals which are present inside the cluster. Ab initio calculation of DOS regarding Pt clusters show that the DOS of clusters is clearly different from the DOS of the Pt bulk. Two physical phenomenon can affect the intensity of the white line : the size of the cluster which can be considered as an intrinsic effect and a possible charge transfer which can be considered as an extrinsic effect E(eV) 11550 11560 11570 11580 11590 11600 Size effect on the Xanes spectrum for Pt 13 , Pt 19 and Pt 43. 4 3,5 3 2,5 D.O.S.2 1,5 1 0,5 00 923 13 6 8 E F 10 E(eV) Averaged D.O.S. of 13 Pt, 55 Pt , 147 Pt , 309 Pt , 923 Pt cuboctahedra. 2 4 F. W. Lytle et al. Phys. Rev B 15-4, 2426 (1975). 14 II.5c Dynamical processes 15 II.5c Xanes L edge : Dynamical process II B Pt-Re/Al2O3 Pt LIII Re LII bsorption of the sample (a.u.) 11400 11540 11680 11820 11960 Energy of the photons (eV) 12100 Preparation data (%wt ) Samples Pt Re Cl A 0.97 1.2 1.2 A' 1.0 0.5 1.0 B B' B" 0.97 1.0 1.0 1.0 1.0 1.0 1.2 1.0 1.0 Fig 1 : Absorption spectrum collected during an in-situ reduction of a 1 weight% Pt-1weight%Re/Al 2O3 sample. Water controls the migration of the Re oxide phase, we distinguish : ■ Hydrated calcined samples (samples A, A') which were reduced after a three month exposure to air, implying that they have been largely rehydrated, Dehydrated calcined samples (samples B, B', B'') which were reduced after a new drying operation of 15 hours at 100°C. For samples B' and B'', an in situ drying at 150°C (120 min for sample B' and 360 min for sample B") was carried out immediately before reduction. 16 II.5c Xanes L edge : Dynamical process II Samples with water : mobility of the Re oxide species Preparation data (%wt ) Samples Pt Re A 0.97 1.2 A' 1.0 0.5 Sample A’ Cl 1.2 1.0 Pt L III edge Re L II edge Sample A White line area of the two metals (a.u.) Pt L III edge Re L II edge Region I III II White line area of the two metals (a.u.) 0 125 250 375 500 Temperature (¡C) Fig. 6. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the hydrated sample AÕ. 0 100 200 300 400 Temperature (¡C) 400 Fig 4 : Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the hydrated sample A. 400 17 II.5c Xanes L edge : Dynamical process II Samples without water : Re oxide species “fixed on the support” Pt,L III edge Sample B Re, L II edge B B' B" White line area of the two metals(a.u.) 50 100 150 200 250 300 350 0.97 1.0 1.0 1.0 1.0 1.0 1.2 1.0 1.0 400 Temperature (¡C) Fig. 8. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the dehydrated sample B. Sample B” Sample B’ Calcination N2 + O 2 Reduction H2 Reduction H2 Cal. N2 + O 2 te line area of the two metals (a.u.) white line area of the two metals (a.u.) Pt, L III edge Re, L II edge Pt, L III edge 100 150 200 250 300 350 Temperature (¡C) Fig. 9. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the dehydrated sample B'. Re, L II edge 400 150 150 200 250 300 350 400 Temperature (¡C) Fig. 10. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the dehydrated sample BÒ. 18 II.5c Xanes L edge : Dynamical process II Pt L III edge Re L II edge Pt L III edge Re L II edge White line area of the two metals (a.u.) White line area of the two metals (a.u.) Region I 0 0 100 200 300 400 Temperature (¡C) 400 Fig 4 : Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the hydrated sample A. 400 125 III II 250 375 500 Temperature (¡C) Fig. 6. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the hydrated sample AÕ. PtRe bimetallic clusters Considering the results obtained on the sample A, we can assume that the first regime is linked to the building of the Pt/Re alloy, the atomic ratio being equal to 1. Pt,L III edge Re, L II edge Monometallic clusters White line area of the two metals(a.u.) 50 100 150 200 250 300 Temperature (¡C) 350 Fig. 8. Evolution of the white line intensities of the two metals (platinum and rhenium) during the reduction of the dehydrated sample B. 400 19 II.6a Exafs : Disavantages : Monometallic : polydispersity II Nt 13 147 1415 13 & 1415 25 N1 N2 N3 N4 20 15 10 5 N1 5.5 8.9 10.5 N2 1.8 4.0 5.0 N3 3.7 13.0 18.5 N4 0.9 6.1 9.1 8.9 4.0 13.8 6.5 0 0 1 2 3 4 5 6 7 Diameter (nm) One of the limitations is the fact that Xas is insensitive to polydispersity. It is clear that the results given by mixing clusters which have 13 & 1415 atoms are similar to the coordination number associated with a cluster of 147 atoms. J. Moonen et al., Physica B 208, 689 (1995). 20 II.6b Bimetallic clusters : repartition of the two metals II Bimetallic system ■N A*NAB=NB*NBA ■σ B B B ■R B A AB=RBA NAA+NAB=12 NAA+NAB>NBA+NBB B B AB=σBA 12 N AA N AB N BA N BB 10 These relations are no longer 8 valid when monometallic clusters 6 coexist with bimetallic ones. Coordination 4 NAB decreases as NMono increases. The distribution of the 2 two metals inside the cluster 0 0 given by the different coordination is simply false. 1 2 3 4 5 Nmono 21 II.7. Comparison between Xanes and Exafs for NSMC II Xanes : size effect B N = 13 C N = 55 Pt metal F.T. (a.u.) Cu metal Absorption (u.a.) 55 Cu 13 Cu 0 -10 0 10 20 30 40 50 1 2 3 4 60 5 6 R() E(eV) •These two figures show that Xas (Xanes) and (Exafs) are well adapted to metallic cluster containing a few atoms. •Xanes as well as Exafs is very sensitive to the size of the particle. 22 III.1 Anomalous Wide Angle X-ray Scattering III 80 f 0 (Pt) 70 60 Complex atomic factor f(q,E) = f0(q) + f'(E) + i f"(E) between f0(q) & the density ρ(r) f0(q)=4π∫0∞ρ(r)sin(qr)/qr r2dr 50 40 30 20 10 Relation Kramers-Kronig relation f’(E)=2/πP∫0∞ef”(e)/(e 2-E2 )de Optical theorem f"(q,E) = E σ(E)/2hcre 0 5 10 20 q( 15 -1 ) 20 Pt L III f"(Pt) 10 0 f'(Pt) -10 -20 -30 11000 11500 12000 12500 13000 E(eV) Debye equation I(q)=ΣiΣj fi(q)fj(q)sin(qRij)/qRij D. Raoux in “Resonant anomalous X-ray scattering : Theory and applications” Ed. G. Materlik et al. North Holland. D. Bazin D. Sayers, Jpn J. Appl. P., Vol. 32, Suppl. 32-2 p 249 (1993). 23 III.1 Monometallic clusters. 111 5 10 311 200 220 7 N = 2869 Pt N = 3871 Pt N = 5083 Pt N = 6525 Pt N = 8217 Pt N = 10179 Pt 4 10 7 222 3 10 N=2057 7 Ιντενσιτψ N=1415 2 10 7 N=923 Intensity (a.u.) 1 10 7 N=561 N=309 0 N=147 2 3 4 5 6 7 3 4 10 -1 N=13 6 5 2π/d( 111 200 9 2 π /d ( ) N=55 2 8 -1 ) ... N= 55 Pt ... N= 13 Pt 311 220 222 Intensity (a.u.) F.T. (a.u.) 55 Pt 13 Pt 2 3 4 5 6 q( -1 ) 0 2 4 6 8 10 R( ) 24 Nanotube (111) & Nanotube (001) 8 10 5 7 10 5 6 10 5 5 10 5 4 10 5 3 10 5 2 10 5 1 10 5 Nanotube Nanotube Nanotube Nanotube Nanotube Nanotube 111 (111) (111) (111) (111) (111) (111) N=45 N=69 N=95 N=121 N=235 N=335 222 333 0 2 3 4 5 6 3,5 10 7 8 9 10 2 π /d (-1) 5 Nanotube (001) N=31 Nanotube (001) N=95 Nanotube (001) N=147 Nanotube (001) N=191 Nanotube (001) N=243 200 3 10 2,5 10 2 10 5 5 5 5 Intensity 1,5 10 1 10 5 5 10 4 400 0 2 3 4 5 6 7 8 9 10 2 π /d (-1) 25 III.2 Comparison between Xas & Awaxs ... N= 55 Pt ... N= 13 Pt Awaxs is better than Xas for the determination of the local order after F.T. (a.u.) the first shell. The low k range of the absorption spectrum is dominated by multiple scattering which lead to a loss of information in the high R range. If X-ray scattering factors have 0 2 similar k dependence whatever the atomic number Z and this is true for wide as well as for small angle scattering, it is not the case of the Xas. Amplitude as well as phase scattering F. T. (a.u.) change significantly with Z for the latter and thus the chemical sensitivity of Xas is much higher. 0 1 4 6 8 10 R(0 ) Pt metal N= 55 Pt N = 13 Pt 2 3 4 5 6 R(0 ) 26 III.3 Size distribution & Morphology 27 W. VOGEL et al. J. Phys. Chem. 97,11611 (1993). III.4a Awaxs-Bimetallic Modeling Pt Mo Mo Pt Pt Pt Mo For cluster A (Pt155-Mo154), We consider a statistical distribution of Pt and Mo. MoMo Mo Pt Mo MoMo For cluster B (Pt147-Mo162), Pt atoms are in the core and Mo atoms are at the surface. 80 f 0 (Pt) 70 60 Equation de Debye I(q)=ΣiΣj fi(q) fj(q) sin (qRij)/qRij 50 40 30 20 Facteur de diff. atom. : f(q,E) = f0(q) + f'(E) + i f"(E) Relation entre f0(q) & la densité électronique ρ(r) ∞ f0(q) =4π∫0 ρ(r)sin(qr)/qr r2dr 10 0 5 10 20 q(0 15 -1 ) 20 Pt L III f"(Pt) 10 0 f'(Pt) -10 Relation de Kramers-Kronig ∞ f’(E)=2/πP∫0 ef”(e)/(e 2-E2 )de Théorème optique f"(q,E) = E σ(E)/2hcre -20 -30 11000 11500 12000 12500 13000 E(eV) Samant M. G. et al. J. Phys. Chem. 92,3542 (1988). Bazin D., Sayers D., Jpn J. Appl. P., Vol. 32, Suppl. 32-2 p 252 (1993). 28 III.4b Awaxs-Bimetallic Results Pt Mo Mo Pt Pt Pt Mo For cluster A (Pt155-Mo154), We consider a statistical distribution of Pt and Mo. 111 the WPSFs at the Pt and Mo edges are basically the same. The only difference is in the W.P.S.F. (a.u.) amplitude due to the fact that the atomic scattering factor is larger for Mo than for Pt. 2 MoMo Mo Pt Mo MoMo __ Pt L 200 3 III ... Mo K 4 5 6 k(0 -1 ) For cluster B (Pt147-Mo162), Pt atoms are in the core and Mo atoms are at the surface. 111 __ Pt L The splitting between the 111 and 200 peaks is less well defined for the WPSFs at the Pt edge than at the Mo edge. III 200 ... Mo K W.P.S.F. (a.u.) 2 3 4 5 6 k(0 -1 ) 29 III.5 The case of metal oxide AB2O4 - CoFe2O4 - decomposition of N2O Zn/Al2O3 -> ZnAl2O4/Al2O3 : Atomic comp. Zn 2% Al2O3 ZnAl2O4 440 220 311 400 511/333 IntensitŽ (u.a.) 111 331 422 1 1,5 2 2,5 3 3,5 4 4,5 k ( -1 ) 30 III.5a AWAXS at the Zn K edge Complex atomic factor f(q,E) = f (q) + f'(E) + i f"(E) IntensitŽ (u.a.) 0 E= 9660.eV E= 9200.eV 1 2 3 4 5 6 7 -1 8 k ( ) I(9660.)=I(ZnAlO)+I(Al2O3) I(9200.)=I(ZnAlO)+I(Al2O3) ∆I=∆I(ZnAlO) (220) (311) (511) (333) (440) (331) (422) IntensitŽ diffŽrentielle (u.a.)(400) (111) Debye equation I(q)=Σ Σ f (q)f (q)sin(qR )/qR i 1 2 3 4 5 6 7 k( -1 j i j ij ij 8 ) 31 III.5c Numerical simulation (220) (311) (511) (333) (440) (331) (422) IntensitŽ diffŽrentielle (u.a.)(400) (111) 1 2 3 4 5 6 7 k( -1 8 ) (220) (311) (511) (333) (422)(440) (331) (400) (111) x=0 x=0.1 x=0.2 x=0.3 x=0.4 x=0.5 IntensitŽ diffŽrentielle (u.a.) x=0.6 x=0.7 x=0.8 x=0.9 1 2 3 4 5 6 7 k( -1 =1/2 8 ) Simulations basées sur une substitution sur les sites tétrahédriques : (Zn1-xAlx)Td (Al2)Oh O4 32 III.5d Xas at the Zn K edge 2 1,5 ZnAl2O4 ZnAl2O4/Al2O3 /I) 0 log (I1 Zn 2+ Structure spinelle 0,5 0 9620 9640 9660 9680 9700 9720 9740 9760 E (eV) Zn-O T.F. (u.a) ZnAl2O4 : ZnO N=4. , R=1.95Å Zn-Al Zn-O Zn-Zn 0 1 2 3 4 5 6 R () 33 III.5e Quantitativ results Données structurales extraites de l’affinement de Zn/Al2O3 comparées à celles du modèle ZnAl2O4 Chemins N Zn-O Zn-Al Zn-O Zn-Zn Zn-Al-O Zn-O Zn-Al-O N 4 4 10.9 12 10.9 12 0.9 4 23.9 24 6.6 12 14.3 24 R(Å) R(Å) 1.95 3.34 3.38 3.49 3.60 4.27 4.35 1.95 3.36 3.40 3.50 3.62 4.27 4.36 T.F. (u.a) 0 1 2 3 4 5 6 R () 34 III.5f Zn K and L edge 2 2 ZnAl2O4 Zn, K edge Zn, LIII edge Sample 1 Sample 2 1,5 1,5 Absorption (a.u.) Absorption (a.u.) 1 1 0,5 0,5 E(eV) 0 9650 ZnAl2O4 Sample 1 Sample 2 9660 9670 9680 9690 E(eV) 0 9700 1020 1025 1030 1035 1040 1045 1050 35 III.5f Zn K and L edge 2 ZnAl2O4 Zn, K edge Sample 1 Sample 2 1,5 ZnAl2O4 Absorption1 (a.u.) Zn, L II edge 0,5 E(eV) 0 9650 9660 9670 9680 9690 9700 Zn K edge Different structural hypothesis 1. Multiple scattering processes of the photoelectron 2. Dimension of the ZnAl2O4 Absorbance (a.u.) K Zn edge L III Zn edge E (eV) -20 -10 0 10 20 30 40 50 3. Influence of Zn vacancies 4. Influence of oxygen vacancies 36 III.5f Structural proposition 37 IV-1. Monometallic clusters. N N N N N N N For fcc clusters, we can observe different modulations after the (0,0,0) peak and thus information regarding the crystallographic network as well as on the size on the particles is available through ab initio calculations. Nevertheless, we have to point out that physical parameters such the interatomic distance modify the position of these modulations. Also if other kinds of morphology like hemicuboctahedron are considered, no modulations exist. W. Vogel et al. J. Phys. Chem. 97, 11611 (1993). Intensity (a.u.) 0,3 0,4 0,5 0,6 0,7 = = = = = = = 13 55 147 309 561 923 1415 0,8 0,9 1 k(• -1 ) Central diffusion as calculated using the Debye equation near the (0,0,0) peak for cuboctahedra FCC Pt clusters of N atoms. N N N N N = = = = = 10 37 88 181 320 Intensity (a.u.) 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 k(• -1 ) Central diffusion as calculated using the Debye equation near the (0,0,0) peak for hemicuboctahedra FCC Pt clusters of N atoms 38 V. Diffraction Anomalous Fine Structure (Dafs) D. Sayers, H. Renevier, J. L. Hodeau, J. F. Berar, J. M. Tonnerre, D. Raoux, A. Chester, D. Bazin, C. Bouldin, ESRF NewsLetter, January 1997 N°27. Complex atomic factor f (q,E) = f (q) + f' (E) + i f" (E) site 0 site site Debye equation I(q) = Σ Σ f (q)f (q)sin(qR )/qR i j i j ij ij 39 V. Diffraction Anomalous Fine Structure (Dafs) 15 Pt 10 Complex atomic factor f (q,E) = f (q) + f' (E) + i site 0 site f" (E) site ◆The Debye equation is then used to get the diffraction diagram and finally by plotting the maximum of the intensity versus the photon energy it is easy to build the Dafs spectrum. 5 f" 0 -5 -10 -15 -20 f' 11550 11725 11900 12075 E(eV) f' & f" associated to the central atom (0,0,0). 1,3 1,2 55 Pt 1,1 1 13 Pt 0,9 This approach has been extended to clusters of Pt containing 13 and 55 atoms. It is clear that the Exafs oscillations are bigger than the Dafs ones. 0,8 0,7 E(eV) 11400 11600 11800 12000 12200 Dafs spectrum as calculated for a 13 Pt and a 55 Pt cluster. 40 First DAFS experimental results @ ESRF on NSMC 1,3 1,2 55 Pt 1,1 1 13 Pt 0,9 0,8 0,7 E(eV) 11400 11600 11800 12000 12200 Dafs spectrum as calculated for a 13 Pt and a 55 Pt cluster. D. Sayers, H. Renevier, J. L. Hodeau, J. F. Berar, J. M. Tonnerre, D. Raoux, A. Chester, D. Bazin, C. Bouldin, ESRF NewsLetter, January 1997 N°27. 41 VI. Summary - NSMC & RS Monometallic cluster Xas Awaxs Asaxs Network - + - Size + + + Size distribution - + + Morphology + + - Bimetallic cluster Xas Distribution of the + Awaxs - Asaxs - metals inside the cluster 42 II.5 N.S.M.C. & R.S. II 111 311 200 220 222 • Comparison between Xas & Awaxs applied to monometallic clusters. D. Bazin, D. Sayers, Jpn J. Appl. Phys. 32-2, 249, 1993. D. • Comparison between Xas & Awaxs applied to bimetallic clusters. Bazin, D. Sayers, Jpn J. Appl. Phys. 32-2, 252, 1993. D. • AWAXS in heterogeneous catalysis Bazin, L. Guczi, J. Lynch, App. Cat. A 226, 87, 2002. N=2057 N=1415 N=923 Intensity (a.u.) N=561 N=309 N=147 N=55 2 3 4 5 J.D. Grunwaldt et al., J. of Cat. 213,291 (2003) N=13 6 k(• -1 ) L. Drozdova et al. J. Phys. Chem. B 106,2240 (2002) • Real time in situ Xanes approach to characterise electronic state of nanometer scale entities D. Bazin, L. Guczi, J. Lynch, Rec. Res. Dev. Phys. Chem. 4, 259, 2000. • Soft X-ray absorption spectroscopy and heterogeneous catalysis. D. Bazin, L. Guczi, App. Cat. A 213/2, 147, 2001. • New opportunities to understand heterogeneous catalysis processes through S.R. studies and theoretical calculations of density of states : The case of nanometer scale bimetallic particles D. Bazin, C. Mottet, G. Treglia, Applied Catalysis A (1-2), 47-54, 2000. D. • New trends in heterogeneous catalysis processes on metallic clusters from S.R. & theoretical studies Bazin, C. Mottet, G. Treglia, J. Lynch, Applied Surf. Sci. 164, 140, 2000. 43 VII. Conclusion - Future Prospects Characterisation technics related to synchrotron radiation like Xas, Awaxs, Asaxs are in situ chemical tools. Follow the evolution of catalysts during the activation process. Structural Characterization @ the atomic scale Catalytic Activity 44