GENESIS OF THE METALLIC PHASE IN A HIGHLY DISPERSED CATALYST FORMED FROM Pt ACETYLACETONATE ON ALUMINA

HAL Id: jpa-00226171

https://hal.archives-ouvertes.fr/jpa-00226171

Submitted on 1 Jan 1986

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

GENESIS OF THE METALLIC PHASE IN A HIGHLY DISPERSED CATALYST FORMED FROM Pt

ACETYLACETONATE ON ALUMINA

J. Berdala, E. Freund, J. Lynch

To cite this version:

J. Berdala, E. Freund, J. Lynch. GENESIS OF THE METALLIC PHASE IN A HIGHLY DIS- PERSED CATALYST FORMED FROM Pt ACETYLACETONATE ON ALUMINA. Journal de Physique Colloques, 1986, 47 (C8), pp.C8-265-C8-268. �10.1051/jphyscol:1986849�. �jpa-00226171�

JOURNAL DE PHYSIQUE

Colloque C8, suppl6ment au n o 12, Tome 47, decembre 1986

GENESIS OF THE METALLIC PHASE IN A HIGHLY DISPERSED CATALYST FORMED FROM Pt ACETYLACETONATE ON ALUMINA

J. BERDALA, E . FREUND and J . P . LYNCH

Institut F r a n ~ a i s du Petrole, BP 311, F-92506 Rueil Malmaison Cedex, France

L'EXAFS et lointensit& des raies blanches nous ont permis d'Btudier les Btapes de preparation d'un catalyseur 0.7 % Pt (ex acetylacetonate de platine)/Al 0

2 3'

A des temperatures comprises entre 100°C et 200°C, sous air, nous observons le depart progressif des deux mol6cules d'ac6tylacetone. Le platine reste sous forme de complexes mononuclBaires isolBs. Entre 300°C et 500°C, on constate un frittage bidimensionnel sans formation de PtO

A faible tempCrature de reduction sous H2, les particules ne sont 2' pas parfaitement reduites, on observe des distances Pt-Pt trss courtes (2.71 A ) . A haute tempGrature, le frittage se poursuit avec formation de partlcules tridimensionnelles possedant des distances typiques de petits clusters (2.75 A).

Abstract

We have studied the preparation of an 0.7 wt % Pt/A1203 catalyst from platinum acetylacetonate.

Between 100°C and 200°C in air, we observe the progressive release of the two acetylacetone molecules, the ~latinum being fixed as isolated mononuclear complexes. Twa dimensional sintering, without oxide formation, occurs from 300°C to 500°C. Reduction qt low temperatures in hydrogen is incomplete and leads to very short Pt-Pt bonds (2.71 i). At higher temperature, sintering continues with formation of three odimensional aggregates whose bond lengths are typical of small clusters (2.75 A).

INTRODUCTION

The drying, calcination and reduction of an 0.7 wt % platinum acetylacetonate (Pt acac.) on alumina catalyst have been followed by EXAFS and white line studies. Pt acac. was chosen as a precursor to limit the type of bonds present since Pt-0 and Pt-C1 cannot clearly be resolved (1). All spectra were recorded in a reaction cell under gas flow at room temperature. Pt-Pt and Pt-0 bond numbers p n d , distances were determined after isolating the contributions using k welghted Fourier transforms over an energy range from 50 eV to 720 eV and using phases and backscattering amplitudes derived from Pt metal, Pt acac. and Pt02 references (Fig. 1). Errors were estimated by fitting EXAFS of one reference (Pt acac.) from an otheg (PtO ). We estimate a precision of 2 0.5 in coordination number and 2 0.02 A in distances for edge 2 energy differen~ga of less than 0.1 eV and Debye Waller factor differences of less than 0.04 A . The LIII near edge regions of selected references (Fig. 4) show the white lines whose intensity gives a qualitative evaluation of d electron deficiency.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1986849

JOURNAL DE PHYSIQUE

The catalyst was prepared by impregnation of alumina with a solution of Pt acac. whose molecule has the following structure _:

Fourier transforms are shown in figures 1 to 3 and the results of fitting back transformed EXAFS in table 1. The LIII edges of references and samples are shown in figures 4 to 6.

-

-

o . .

.

--

i ^{calcined ZZO-c} i ^Z.W+O.OZ i 2.4to.6 i i i

I 3w'C 1 2 . 0 4 + 0 . 0 2 1 2.5+0.7 1 2 . 7 8 + 0 . 0 3 l 0 . 7 t 0 . 5 1

I 5 W ' t 1 2.02t0.02 1 2.5i0.7 1 2.73t0.02 I 4.2tO.7 l

I I I I I

I ^Reduced 2 0 0 * t I ^2.02+0.02 I ^1.0+0.5 1 ^2.71+0.02

I (calcined 3W.C) 1

I I I I I

I ^{Reduced 4W.C I I}

I ( c a i c l n d 3w.C1 I I 1 ^{z.75+0.02 I}

I I - 1 I -11 ^5.4*1.0 I

DISCUSSION

The lack of Pt-Pt neighbours for the dried samwle shows that the -

metal stays in the form of isolated mononuclear complexes. We see only Pt-0 neighbours, at distances intermediate between Pt acac. and Pt02. It is known (2) that acetylacetone molecules cannot stay fixed to platinum at 500°C so the constant 2 or 3 oxygen neighbours seen between 200°C and 500°C can be assigned to the support. If this assignment is applied also to the dried specimen (it is known that one acetylacetone molecule is liberated on impregnation (2) ) we can say that the complex is here fixed by two to three support oxygens and retains one acetylacetone molecule (two oxygen bonds). We can see indications of this in the white line intensity (fig.5), which is intermediate between those of PtO (strong, ionic, bonding) and Pt acac. (weaker bonding), suggesting tha5 Pt acac. is not simply deposited on the support. We then see the loss of the second acetylacetone molecule at 200°C.

An other important point concerns particle coalescence during calcination in dry air, which appears at around 300°C. The first Pt-Pt distances detected are close to that of the metal (2.78 a) ^{with the}

formation of small clusters distances shorter than this (2.75 A ) and typical of small clusters appear (3,4). With increasing Pt-Pt coordination number, the number of Pt-0 bonds stays constant, suggesting a raft morphology. It seems therefore that air sintering at high temperature leads directly to two dimensional metallic particles without intermediate formation of oxides (PtO

C8-268 JOURNAL DE PHYSIQUE

or Pt02), which indicates strong inter.actions between complex and support. An heterogeneous mixture of reduced particles and oxide species is ruled out on grounds of the constant values of Pt-0 coordination and white line intensity whilst Pt-Pt coordination increases.

The reduced samples were beforehand calcined at 300°C to avoid air sintering (fig.3). For a temperature of 200°C in hydrogen we see a reduction in the number of Pt - 0 bonds. That some Pt - 0 bonds remain, in addition to the Pt - Pt bonds formed, shows that reduction is not complete. The Pt - ⁰

distance is slightly less than that observed in PtO and the Pt - Pt distance much shorter than in bulk metal. If the popula$ion is homogeneous these aggregates are very different from small metallic particles. It is possible that these are two dimensional aggregates well fixed to the support so that the number of Pt - 0 bonds is not negligeable compared with the number of Pt-Pt bonds. There is also the possibility that the specimen is heterogeneous with raft like particles such as were noted during calcination and completely reduced ones. In either case, the very short Pt-Pt distances are related to the very small nuclearity and positively charged platinum atoms (this is confirmed by the white line intensity) (Fig. 6).

At 400°C, reduction - i s complete. The small metaflic particles formed have Pt-Pt distances slightly shorter than the bulk value. Clearly for industrial preparation conditions for Pt acac. catalysts (drying at llO°C, calcination at 300°C and reduction at 400°C) it is at the reduction stage that particle formation occurs. Aggregation begins at 200°C under hydrogen but the particles become three dimensional only afterwards. Some Pt-0 bonds must be present but their number will be negligeable compared with the Pt-Pt bonds for three dimensional particles. In addition particle sizes deduced from first neighbour numbers and a spherical particle model are in good agreement with electron microscopy results (5).

CONCLUSION

EXAFS has shown that at the drying stage the metal stays as isolated mononuclear complexes. There is undoubtedly a strong su~oort - - - interaction with one acetylacetone molecule replaced by two or three oxygen atoms from the support. During calcination, the second acetylacetone molecule is released at 200°C, followed by two dimensional sintering between 300°C and 500°C, the bonds with two or three support oxygens remaining at all temperatures. Oxide formation is not observed. High temperature reduction (400°C in hydrogen) after calcination at 300°C transforms the isolated mononuclear complexes into small, three dimensional particles. The large number of surface atoms (high dispersion) leads to Pt-Pt distances ,shorter than the bulk metal. For low temperature reduction (200°C) we show very short Pt-Pt distances incompatible with the bulk structure and the presence of one Pt - ⁰ bond per platinum atom. At this stage, the particles are probably two dimensional.

( 1 ) J. BERDALA, E. FREUND and J.P. LYNCH

Environment of Platinum in a H PtCl /A1 0 catalysts : influence of metal loading and chlorine congent. ^{6 2 3}

EXAFS and Nead Edge structure IV.

( 2 ) S. VASUDEVAN Thesis IFP 1982

(3) G.M. VIA, J.H SINFELT and F.W. LYTLE J. Chem. Physi. 71, 690 (1979) ("4) R.B. GREEGOR and P.W. LYTLE

J. of Catalysis 63, 476 (1980) ( 5 ) J. BERDALA

Thesis IFP, 1986