CATALYSTS BY USING SINCHROTRON RADIATION

advertisement

THE COMPLETE LOCAL AND GLOBAL STRUCTURE OF THE NICKEL

CATALYSTS CHARACTERIZED BY SYNCHROTRON RADIATION

N. Aldea 1 Andreea Gluhoi 1 , B. Barz 1 , P.Marginean

1 , Xie Yaning 2 ,

Hu Tiandou 2 , Liu Tao 2 , Zonghua Wu 2 and Ziyu Wu 2

1 National Institute for Research and Development of Isotopic and

Molecular Technologies, R-3400 Cluj-Napoca, P.O. 5, Box 700,

Romania

2 Beijing Synchrotron Radiation Facilities of Beijing Electron Positron

Collider National Laboratory, Beijing, People's Republic of China

Abstract

The supported Ni catalysts were analyzed by Extended X -ray Absorption Fine

Structure (EXAFS) and X-Ray Diffraction (XRD) in order to determine their local and global structure. The present study evidenced by EXAFS method a strong deformation of the local structure of the metal due to its interaction with oxide support. The average particle size, the mean square of the microstrain the probability of the faults and the particle distribution function of the supported Ni catalysts were determined by XRD method. The method is based on Fourier analysis of the experimental X-ray line profiles (XRLP) (111),

(200) and (220). The global structure is obtained with a fitting method based on the generalized Fermi function (GFF) facilities for approximation. The local and global microstructures of the catalysts are modified by the strong interaction between the metal clusters and oxidic supports. As a result the chemisorption and catalytic processes which occur at the Ni-support interface are also changed.

Introduction

The theoretical background of EXAFS and XRD analysis was presented in previous papers [1,2]. The main mathematical properties of the GFF used for approximation and deconvolution of the experimental XRLP were described in Ref. [1,3]. A completed description of the methods for preparation of the supported nickel catalysts, experimental details using synchrotron radiation and data processing were presented in

Ref. [1,3-5]. The present paper offers the complete local and global structure characterization by analyzing the first to forth coordination shells and (111), (200) and

(220) XRLP using X-ray absorption spectroscopy method and X-ray diffraction, respectively. We have investigated the following systems: 60 at. % Ni/ZnO, 89.6 at. %

Ni/UO

2

, 88.4 at. % Ni/ThO

2

, 60 at. % Ni/MgO, 80 at. % Ni/SiO

2

, 70 at. % Ni/ZrO

2 and

85 at. % Ni/Al

2

O

3

. The transmission EXAFS and XRD measurements were carried out

in 4W1B and 4W1C beamlines in Beijing Synchrotron Radiation Facilities (BSRF) operating at 30-50 mA and 2.2 GeV at room temperature [6].

Results and discussion

EXAFS results

Fig. 1-(a) shows the absorption coefficients for investigated catalysts. The extraction of the EXAFS signal is based on E

0

determination, followed by background removal before pre-edge and after edge baseline fitting with different possible modeling

( a )

Ni foil

1

2

2

4

5

6

7

-200 0 200 400 600 800 1000

E-E

0

(eV)

,

,

6

5

1

2

3

4

( b )

1 2 3 4 5 6

-1 Distance (Å)

1

6

7

2

3

4,5 energy E

0

= 8332 eV, lower box - Ni foil as standard sample; (b) the Fourier Transforms of the EXAFS spectra (  ) black Ni powder; (1)- 60 at % Ni/ZnO, (2) - 89.6 at % Ni/UO

2

,

(3) - 88.4 at % Ni/ThO

2

, (4) - 60 at % Ni/MgO, (5) - 80 at % Ni/SiO

2

, (6) - 70 at %

Ni/ZrO

2

, (7) - 85 at % Ni/Al

2

O

3 functions. The local structure of these samples corresponding to each coordination shell have been determined by Fourier transform technique of the EXAFS function defined in the range 20-150 nm -1 . The local structure of the standard Ni black powder for each coordination shell is described by the following parameters: N

1

=12, R

1

=0.249,

(a)

Ni black

0.018

0.016

0.014

0.012

(b)

0.010

0.008

1

2

0.006

0.004

.

.

20 25 30 35 40 45 50

7

Diffraction angle (

O

)

0.002

0.000

40 60 80 100 120 140 160 180 200 220 240

Distance (A)

Fig.2. (a) The XRLP (111), (200) and (220):, (1) - 60 at % Ni/ZnO, (2) - 89.6 at %

Ni/UO

2

, (3) - 88.4 at % Ni/ThO

2

, (4) - 60 at % Ni/MgO, (5) - 80 at % Ni/SiO

2

, (6) - 70 at

% Ni/ZrO

2

, (7) - 85 at % Ni/Al

2

O

3

; (b), the crystallite distribution function: ■ (111), ●

(200) and (―) - (220) X-ray line profiles for 60 at. % Ni/ZnO.

N

2

=6, R

2

=0.352, N

3

=24, R

3

=0.431, N

4

=12 atoms and R

4

=0.497 nm, respectively.

Table 1.The local structural parameters of the investigated samples

Sample

60 at % Ni/ZnO

89.6 at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

60 at % Ni/ZnO

89.6 at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

60 at % Ni/ZnO

89.6 at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

60 at % Ni/ZnO

89.6 at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

Shell

1

2

3

4

R±ΔR

Shell radius(nm)

0.25±0.01

0.25± 0.01

0.25±0.01

0.25±0.005

0.25±0.006

0.25±0.003

0.256±0.001

0.35±0.01

0.351±0.03

0.349±0.02

0.352±0.01

0.35±0.02

0.433±0.008

0.433±0.05

0.432±0.007

0.433±0.01

0.499±0.001

0.498±0.001

0.5±0.003

0.5±0.001

0.502±0.003

N±ΔN

Atomic numbers

1 0±0.1

9.01±0.1

10.15±0.12

7.17±0.1

9.32±0.11

9.16±0.2

6.08±0.12

5.53±0.1

4.32±0.05

4.46±0.02

3.73±0.04

4.46±0.02

25.05±0.03

17.58±0.022

12.92±0.03

18.64±0.04

9.1±0.1

10.10±0.05

10.34±0.02

7.09±0.01

10.10±0.05

The diminution of the Fourier transform magnitude is a result of the reduced average coordination number. Each peak from |

(r)| is shifted from the true distance due to the phase shift function which is included in the EXAFS signal. By assuming that each radial distribution function |

(r)| can be a Gaussin distribution and based on Lavenberg-Marquard [1] fit, the total radial distribution function

|

(r)| is split into four contributions. Thus, the broadening and center of each coordination shell is represented in Fig. 1-(b). By taking the inverse Fourier transform of the Gaussin distribution given in Fig. 1-(b), we obtained the

experimental EXAFS function attached to each coordination shell. Then, the theoretical amplitude envelope and the phase shift functions were extracted from a black nickel powder as a standard sample. Table 1 contains the best values of the local structural parameters for all investigated coordination shells. In according with our previous data reported in [1-3], we found that the average interatomic distances of the first to the forth coordination shells are closely related with standard sample. The atom number of the first shell decreases with 15% (88.4 at.

% Ni/ThO

2

) up to 49% (85 at. % Ni/Al

2

O

3

) relative to the standard sample. By analyzing the atom number of the second coordination shell, we also obtained a diminution between 8 % (60 at. % Ni/ZnO) and 38 % (60 at. % Ni/MgO). The structural model of the third coordination shell gives an increase of the atom mumber with 4 % for 60 at. % Ni/ZnO and a decrease between 22 % (70 at. %

Ni/ZrO

2

) and 46 % (60 at. % Ni/MgO). The last coordination shell shows a decrease between 14 % (88.4 at. % Ni/ThO

2

) and 41 % (60 at. % Ni/MgO).

Therefore, we consider that this diminution of the atom number is due to a strong interaction between the Ni atoms and the support surface. The metal-support interface can have influence upon the electronic properties of the metal clusters, and as a consequence their reactivity for adsorption of the molecules are modified.

Also, the nature of the interaction of the metal clusters with the support was expected to influence both the electronic properties and morphology of the clusters and thus their catalytic properties.

XRD results

The XRLP (111), (200) and (220) were processed in manner of the Ref. [1-2].

Their relative intensities with respect to the diffraction angle for standard sample as well as for investigated samples are shown in Fig. 2-(a). Nobody has ever used the

GFF for the XRLP approximation and its Fourier Transform before. In former section we have shown that the coordination shell radius of investigated sample have similar values as Ni standard sample. An additional proof is given by the position and the shape of each Ni K edge around to 8333 eV; it means that all the samples contain only metallic nickel. These important results are strongly correlated with the positions of (111), (200) and (220) XRLP from the experimental spectra presented in Fig. 2-(a). Therefore, these results explain the metallic features of the investigated clusters despite the strong deformation of the crystalline structure. Fig. 2-(b) shows the particle size distribution for 60 at. %

Ni/ZnO sample .

The main results regarding the investigated supported nickel catalyst microstructure are summarised in Table 2. As a measure of the discrepancy between the XRLP approximated by the GFF and pseudo-Voigt or Pearson V we defined a residual index [1-2]. By analysing the values of the residual index for each of the distributions, we demonstrated the validity and reliable of the GFF

approximation. The values of the crystallite size can be correlated with the mean value of the grain diameter [7], if we take into consideration the physical meaning of the crystallite size and the grain one. It is possible that a grain of nickel cluster is built up from more nickel crystallites. In this case, the adsorption probability of the hydrogen on the interface of the crystallites, inside the metal cluster, is much reduced.

Table 2. The global structural parameters of the investigated samples

Name of sample

(111),

 g

=0.0029 nm -1

60 at % Ni/ZnO

89.6at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

(200),

 g

=0.0031 nm -1

60 at % Ni/ZnO

89.6at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

(220),

 g

=0.0028 nm -1

60 at % Ni/ZnO

89.6at%Ni/UO

2

88.4 at % Ni/ThO

2

60 at % Ni/MgO

80 at % Ni/SiO

2

70 at % Ni/ZrO

2

85 at % Ni/Al

2

O

3

 f

(

 h

,

[nm] -1 g

)

0.093

0.104

0.083

0.104

0.164

0.241

0.214

0.075

0.963

0.728

0.07

0.134

0.141

0.228

0.112

0.104

0.083

0.096

0.163

0.179

0.228

6.8

6.9

8.8

6.7

4.7

2.9

3.4

D

( hkl ) eff

7.9

6.9

8.8

6.7

4.7

2.9

3.4

[nm]

9.5

7.3

10.4

10.2

5.9

5.1

3.2

2 hkl

10 -4

0.2

0.1

0.3

0.1

0.6

0.8

0.7

0.2

0.07

0.4

0.3

1.4

1.3

0.04

0.2

1.5

0.19

0.13

0.42

0.89

1.5

0.14

1.8

0.12

0.19

0.37

1.01

0.73

0.14

0.22

0.12

0.13

0.36

0.41

1.04

0.1

0.34

0.4

0.06

0.20

0.24

0.4

SFP (hkl)

[%]

 g

(

 g

) - integral width for standard black Ni powder,

 f

(

 h

,

 g

) - integral width for true samples,

D

( hkl )

- effective particle size, eff

2 hkl

- mean square of the microstrain, SFP (hkl)

- stacking fault probability (Ref. [1] contains the meaningful of each parameters).

Conclusions

The information presented in this paper describes the whole local and global structure of supported nickel catalysts determined by EXAFS and XRD analysis.

The additional results give a complete image of the strong deformation of the crystalline structure. The conclusions that can be drawn from these studies are:

(i) The local structure of the investigated systems shows a strong morphological deformation of the nickel clusters caused by the metal-support interaction. These morphological modifications can be accompanied by modification of the electronic structure that has a large effect on the reactant adsorption. A result of these modifications induced by the supports is the enhancement of the catalytic activity for H/D exchange reaction between hydrogen and water vapor.

(ii) The numerical results obtained from (111), (200) and (220) XRLP show that by using the GFF distribution we can also obtain reliable microstructural parameters.

References

[1].

N. Aldea, Andreea Gluhoi, P. Marginean, C.. Cosma and Xie Yaning, Extended X-ray

Absorption Fine Structure and X-ray Diffraction studies on supported Ni catalysts ,

Spectrochim. Acta Part B 55, 997 (2000).

[2].

N. Aldea, Andreea Gluhoi, P. Marginean, C. Cosma, Xie Yaning, Hu Tiandou,

Whongua Wu and Baozhong Dong, Investigation of supported nickel catalysts by Xray absorbtion spectrometry and X-ray diffraction using synchrotron radiation ,

Spectrochim. Acta Part B 57, 1453 (2002).

[3].

B. Barz, N. Aldea, Andreea Gluhoi, P. Marginean, Xie Yaning, Hu Tiandou, Liu Tao and ZonhuaWu and Ziyu Wu The analysis of the interaction metal-support in Nicatalysts by extended x-ray absorption fine structure and x-ray diffraction using synchrotron radiation, Colloquium Spectroscopicum Internationale CSI XXXIII, 7-12

Sept. 2003, Granada, Spain.

[4].

A. Gluhoi, Barz, P. Marginean, N. Aldea, X-ray diffraction analysis on supported nickel catalysts , National Conference of Physics, Targu-Mures, Romania 2002.

[5].

B.Barz, Andreea Gluhoi, P. Marginean, N. Aldea, Xie Yaning and Baozhong Dong, X-

Ray diffraction analysis on supported nickel catalysts and their physical-chemical properties , CERES Programme, Annual Scientific Session, Dec. 2-3 2002, Bucharest.

[6].

B.S.R.F. Activity Report 2001, Beijing Electron Positron Collider Laboratory.

[7].

P. Marginean and A. Olariu, Metal /oxide support effects in H

2

-H

2

O deuterium exchange reaction catalyzed by nickel, , J. Catal. 95 (1985) 1-12.

Download