1

Hélène Provendier Hélène Provendier Institut de Recherches sur la CatalyseInstitut de Recherches sur la Catalyse

Université Claude Bernard Lyon 1Université Claude Bernard Lyon 1

École FRANCO-VIETNAMIENNE 2005 École FRANCO-VIETNAMIENNE 2005 de CATALYSE de CATALYSE

CINETIQUE et RAFFINAGECINETIQUE et RAFFINAGE

Institut de ChimieIndustrielle

Hanoi, 18-22 Avril 2005

2

Physical and chemical techniques for catalyst characterization

Physical and chemical techniques for catalyst characterization

Centre National de la Recherche Scientifique

IRCInstitut de Recherches sur la Catalyse

Hanoi, 18-22 Avril 2005

ICI

Dr. Hélène Provendier

Institut de Recherches sur la CatalyseVilleurbanne

France

Ecole Franco-Vietnamienne 2005 de Catalyse Cinétique et Raffinage

3

Part 1 : Adsorption techniques

Part 2 : Diffraction techniques

Part 3 : Spectroscopies

Contents Contents

4

Part 1 :Characterization of

porous solidsusing adsorption

techniques

Part 1 : Adsorption techniques Part 1 : Adsorption techniques

5

Outline I. Introduction to solid surface and porosity1) Porous solids2) Some definitions3) Qualitative description of a porous solid4) Texture of a solid 5) Classification of pores6) Characterization methods

II. Nature of Adsorption1) Definition of adsorption2) Two types of adsorption3) Transition physisorption-chemisorption4) Activation energy

III. Physisorption1) General2) Theoretical approach3) Porous volume measurements4) Pore size distribution

IV. Chemisorption1) General2) Chemisorption isotherms

V. Experimental part

6

Most materials are to some extent porous : they contain empty cavities.

Physical properties :- density, - thermal conductivity - strength

The control of porosity is of great industrial importance for example in the design of catalysts, industrial adsorbents, membranes and ceramics.

Porosity influences :

I. Introduction to solid surface and porosity

In catalysis, porosity determines the accessible surface to the reactant (gas or liquid)It is thus important to characterize porous solids in catalysis

depend on the pore structure of a solid

• the chemical reactivity (activity and selectivity) of solids• the physical interaction of solids with gases and liquids• mass and heat transfers in the solid

1) Porous solids

7

Porous solid : a solid with pores, i.e. cavities, channels or interstices,which are deeper than they are wide.

Pore volume Vp : volume of the pores, as measured by a given method which must be stated, (together, for instance, with the nature of the probe-molecule, the wavelength of the radiation used or the ultimate intrusion pressure ...).

Pore size (generally, pore width) : the distance between two opposite walls of the pore (diameter of cylindrical pores, width of slit-shaped pores).

Porosity : ratio of the total pore volume Vp to the apparent volume V of the particle or powder (excluding interparticle voids). = VP / V

Roughness (or rugosity) factor : ratio of the external surface area to the area of the geometrical envelope of the particles.

Surface area : extent of the total surface as determined by a given method under stated conditions. It is essential to state the method used.

2) Some definitions

8

* Closed pores : (a). They influence such macroscopic properties as bulk density,mechanical strength and thermal conductivity, but are inactive in such processes as fluid flow and adsorption of gases.

* Open pores : (b) (c) (d) (e) and ( f ) . They have a continuous channel of communication with the external surface.

- blind (i.e. dead-end, or saccafe) pores : (b) (f)opened only at one end - through pores : (e)opened at two ends

* Classification according to the shape : - cylindrical (either open (c) or blind ( f ) ) ,- ink-bottle shaped (b), - funnel shaped (d) or slit-shaped.

Close to, but different from porosity is the - roughness of the external surface (g).

Schematic cross-section of a porous solid

3) Qualitative description of a porous solid

f

9

4) Texture of a solid

Texture : detailled geometry of empty spaces in particles. It includes :- intergranular spaces in agglomerates- intragranular pore ditribution- particle shape and external surface- pore shape and porous volume- accessibility of gases to internal surface

Parameters describing the texture of a catalyst- specific surface area (accessible per gram of solid)- porosity

• pore shape• pore size distribution• mean pore size

- granulometry• particle size distribution• shape and size of agglomerates

10

5) Classification of pores

1) By size - macroporous samples  > 50 nm

- mesoporous samples 2 nm <  < 50 nm - micro-porous samples  < 2 nm

2) By shape- cylindrical

- slit-shaped (parallel or not)

- funnel (entonnoir)

- spherical

- ink-bottle shaped

- waved (vague)

11

6) Characterization methods

Specific surface area Adsorption (BET method)

Porosity Adsorption (capillary condensation)Hg porosimetryThermoporosimetryDensity measurementsPermeability (for membranes)Transmission microscopy

Particle size SievingSedimentationLight diffusionX-ray diffractionElectronic Microscopies (Scanning and Transmission)

12

• Solid surfaces show strong affinity towardsgas molecules that it comes in contact withand some of them are trapped on the surface

• the process of trapping or binding ofmolecules to the surface is called adsorption

• Desorption is removal of these gasmolecules from the surface

2) Two types of adsorption

– Physical adsorption :* Van der Waals forces * bond energy is less than 50 kJ/mole

– Chemical adsorption :* bond energy is more than 50 kJ /mole* direct chemical bond

1) Definition of adsorption

Solid

Gas

II. Nature of Adsorption

13

Adsorption and Reaction at Surfaces

14

Rapid equilibration, transport limited

Activated desorption in original form

Exothermic like condensation -5.

15

Energy similar to chemical reaction

Depends on reactivity of adsorbent and adsorptive

-50…..

16

3) Transition physisorption-chemisorption

When P increases, the volume of adsorbed gas Vads increases.When T increases, the volume of physisorbed or chemisorbed gas globaly decreases.

Physisorption takes place at lower temperature than chemisorption and allows the molecule to approach the surface without high energy requirement. Physisorption predominates at low temperature and chemisorption at elevated temperature.

Physisorption

Chemisorption

P = cte

T

Vads Ex : O2 adsorption on Ni At P=1 atm

T = - 200°C physisorptionT = 25 °C chemisorptionT = 800°C oxidation (NiO)

If the molecule is chemically transformed before desorption, there is contact catalysis

17

2) Then chemisorption must be activated to form the transition state (energy required)

Example of H2 adsorption on Ni catalyst

H2 + 2 Ni 2 Ni-H HC = 120 kJ.mol-1)

1) Physisorption is the first step of chemisorptionDistance from surface : r (Ni…H) = 3.2Å

Distance from surface : r (Ni-H) = 1.6Å

18

4) Activation energy

EP = Energy evolved during physisorption (exothermic)

EC = Energy evolved during chemisorption (globaly exothermic)

Ea = Activation energy (transition state)

Ed = Energy required for desorption

Molecule A-B approaches the surface S of an adsorbent along the distance axis r.The first interaction process is physisorption at equilibrium position rP : exothermic process EP is evolved. Next is the endothermic stage where activation Ea is input to form the transition state. This displaces the molecule toward equilibrium position rC and Ec is evolved during chemisorption. Chemisorption is always an activated process.

19

III. Physisorption1) General

-Physisorption is not adsorbent or adsorptive specific. This is a process similar to condensation of gas on a surface.

-The quantity of physisorbed molecules depends on the accessible surface area and not on its chemical nature.

-Physisorbed molecules progressively form successive layers when the gas pressure P increases. When P reaches the vapor pressure P0, there is condensation on the surface.

-However there can be capillary condensation in the solid pores for P < P0 depending on the pore size.

20

For molecules in contact with a solid surface at a fixed temperature, the Langmuir Isotherm, developed by Irving Langmuir in 1916, describes the partitioning between gas phase and adsorbed species as a function of applied pressure.

The adsorption process between gas phase molecules, A, vacant surface sites, S, and occupied surface sites, SA, can be represented by the equation,

assuming that there are a fixed number of surface sites present on the surface.

An equilibrium constant, K, can be written :

2) Theoretical approacha) Langmuir theory (monolayer adsorption)

1) Thermodynamic Derivation

21

Thus it is possible to define the equilibrium constant, b:

              Rearranging gives the expression for surface coverage:

             = Fraction of surface sites occupied (0 1) = V/Vmwith V = adsorbed volume of gas ; Vm = adsorbed volume at saturation ( = 1)

•[SA] is proportional to the surface coverage of adsorbed molecules, or proportional to

•[S] is proportional to the number of vacant sites, (1 - )

•[A] is proportional to the pressure of gas, P

This simplest theory is based on three assumptions:• 1. Adsorption cannot proceed beyond monolayer coverage.• 2.All surface sites are equivalent and can accommodate, at most, one adsorbed atom.• 3.The ability of a molecule to adsorb at a given site is independent of the occupation of neighboring sites.

22

2) Kinetic Derivation

The rate of adsorption will be proportional to the pressure of the gas and the number of vacant sites for adsorption. If the total number of sites on the surface is N, then the rate of change of the surface coverage due to adsorption is:                                                     

The rate of change of the coverage due to the adsorbate leaving the surface (desorption) is proportional to the number of adsorbed species:

In these equations, ka and kd are the rate constants for adsorption and desorption

repectively, and p is the pressure of the adsorbate gas.

> 0

< 0

a

d

23

At equilibrium, the coverage is independent of time and thus the adorption and desorption rates are opposite (equal in absolute value).

where b = ka/kd

The solution to this condition gives us a relation for :

ka P (1 – ) = kd This leads to :

I. Langmuir, J. Amer. Chem. Soc., 40, 1361 (1918); I. Langmuir, J. Amer. Chem. Soc., 54, 2798 (1932);I. Langmuir, Nobel Lecture, 1932

Ref :

a d= -

24

3) Dependence of b on external parameters:

b is only a constant if the enthalpy of adsorption is independent of coverage.

As with all chemical equilibria, the position of equilibrium (determined by the value of b) will depend upon a number of factors:

1.The relative stabilities of the adsorbed and gas phase species involved. 2.The temperature of the system (gas and surface normally the same). 3.The pressure of the gas above the surface.

In general, factors (2) and (3) exert opposite effects on the concentration of adsorbed species : the surface coverage may be increased by raising the gas pressure but will be reduced if the surface temperature is raised.

25

4) Heat of adsorption

b = ka / kd is a T-dependent equilibrium constant

the values of b determined from the Langmuir isotherms at various

temperatures allow for the evaluation of the enthalpy of adsorption, Hads, through the van't Hoff equation :

b

The temperature dependance of the adsorption isotherm gives the heat of adsorption

26

5) Graphical form of the Langmuir Isotherm

where b3 > b2 > b1

Isotherm is an equation which relates the amt of substance attached to the surface (or surface coverage ) to its concentration in the gas phase (or gas pressure P) at a fixed temperature.

For small P values : bP

For high P values : 1

27

6) Linearisation of Langmuir isotherm

Can be linearised as

1/bVM

)(PfV

V=V mbP

Graphical determination of b and Vm

bP

bP

V

V

m

1

mm V

P

bVV

P

1

)(PfV

P

28

First theory : Langmuir for monolayer adsorptionExperimental reality : adsorption is higher than monolayerSecond theory : BET for multilayer adsorption

7) Specific surface area determination using Langmuir theory

Some sets of data obtained using nitrogen as the adsorptive gas and adsorbing it at N2

condensation temperature (77K at 1 bar) plot a straight line only in limited regions, allowing b and Vm to be evaluated from the slope (1/Vm) and the intercept of the line (1/bVm).

Slope = 1/Vm

Intercept = 1/bVm

b = Slope / InterceptVm = 1/ Slope

The specific surface area SL of the adsorbent evaluated from Langmuir theory can be calculated from Vm by :

0mV

NVS Am

L

Where : is the area of the surface occupied by a single physisorbed gas molecule (=16.2Ų for N2)NA the Avogadro constant (NA = 6.02 1023 mol-1)m the mass of the adsorbing solid sampleV0 the molar volume of the gas (22414 cm3)

29

b) Derivation of the BET Isotherm (multilayer adsorption)BET : Brunauer, Emmet, Teller

Consider a surface:

When the pressure system is increased, the adsorbate adsorbs in multilayers to the surface.

30

Brunauer, Emmet, Teller (BET) theory :

Generalization of the Langmuir theory to multilayers adsorption, with the following hypotheses :

•The evaporation (desorption) rate of adsorbed molecules in a layer is equal to the condensation (adsorption) rate on the layer under it.

•The adsorption heat in the layers (except the first one) is equal to the liquefaction heat of the gas.

•At saturation the number of layers is considered as infinite.

1) Assumptions

31

2) Definition :

0, 1, ..., n = Surface area (cm-2) covered by 0, 1, ..., n layers of adsorbed

molecules.

At Equilibrium:

0 must remain constant. (First layer formation and destruction rates are equal)

Rate of Evaporation from First Layer = Rate of Condensation onto Bare Surface

(Eq.1)

3) Kinetic derivation theory :

For the first layer :

32

Similarly, at equilibrium 1 must remain constant. (Formation and destruction rates

of monolayer are equal)

k1P0 + k-22 = k2P1 + k-11

Substituting into (Eq.1) gives for the second layer : k-22 = k2P1

Extending this argument to other layers,

firstfirst

Monolayer formation : Monolayer disappearence :

33

Definitions:

Total surface area of the catalyst,

Total volume of gas adsorbed on surface V

where v0 is the volume of gas adsorbed on one square centimeter of surface when it is

covered with a complete layer.

where vm is the volume of gas adsorbed when the entire surface is covered with a

complete monolayer.

(Eq.2)

34

From Eq.1 with

Assuming that the properties of the 2nd, 3rd ... layers are equivalent, then,

Similarly, with

3 =x2 =x21

Generally, i =xi-1 =xi-11 =xi-1y0 =cxi0 with c=x/y

Then

35

Substituting into Eq.2

                                                                              

                                                                            

36

At saturation pressure of gas, P0, an infinite number of adsorbate layers must

build up on the surface. From previous equation, for this to be possible, V/Vm must be infinite. This means that at P0, x must equal 1.

So x = P/P0

This leads to BET isotherm equation

This can be rearranged to give,

(Eq 3)

37

Graphical form of the BET isothermV = f(P)

Replacing x by P/P0 in Eq 3

4) Graphical form of the BET Isotherm

00

0

)1(1)1(PP

cP

PP

cV

Vm

38

5) Dependance of the isotherm form with the BET constant c

x = P/P0

E1 is the adsorption heat of the first layer and EL is the liquefaction heat. C is an

indication of the affinity of the adsorbed molecule for the solid : if this affinity is high, E1 >> EL and c is high

39

From the adsorption isotherms VM and c can be deduced by plotting P/(V(P0-P)) = f(P/P0),

and the experimental results for the slope (generally determined for P/P0<0.35) and the

extrapolated value at P=0.

6) Linearisation of BET isotherm

Intercept = 1/cVM

Slope = (c-1)/cVM

InterceptSlopeVM

1

Slope

InterceptSlopec

00

)1(1

)( P

P

cV

c

cVPPV

P

mm

BET equation can be linearised as

40

7) Specific surface area determination using BET theory

From VM value obtained using BET linearised isotherm, the specific surface area called SBET is calculated using the same formula as in Langmuir theory :

0mV

NVS Am

BET

Ex : 2 linearised isotherms

BET theory is valuable in the range

0.05 < P/P0 < 0.35

At low P : heterogeneous surfaceAt high P : capillary condensation

Where : is the area of the surface occupied by a single physisorbed gas molecule (=16.2Ų for N2)NA the Avogadro constant (NA = 6.02 1023 mol-1)m the mass of the adsorbing solid sampleV0 the molar volume of the gas (22414 cm3)

41

c) Other theories proposed

Depending on the hypotheses made (mono or multilayer adsorption, interactions between the adsorbed molecules or not), several theories have been developped.

However, the Langmuir and the BET theories are the mostly used.

42

The desorption measurement is performed after adsorption to see if desorption is exactly the reverse of adsorption or if there is any hysterisis or non-equilibrium effects in the adsorption/desorption cycle. This is basically performed in reverse of the adsorption procedure.

Classification following the shape of the adsorption/desporption isotherm

d) Desorption

For mesoporous solids, the adsorbate condenses in the pores at P<P0 and desorbes from the pores at P’<P<P0 due to capillary effects. The hysteresis loop is due to capillary pressure,which retains the condensed gas from desorbing and induces desorption at lower pressure.

Capillary condensation in mesopores

43

Six types of adsorption isotherms following IUPAC classification :Type I : non porous or micro-porous samples ( < 2 nm)Types II and III : macroporous samples ( > 50 nm) Types IV and V : mesoporous samples (2 nm <  < 50 nm) Type VI : step isotherm (rare)

Types III and V correspond to a low enthalpy of adsorption (exponential)

e) Types of isotherms

Ref : K.S.W. Sing et al., Pure and Appl. Chem., 57, 603, 1985

44

* Nitrogen (at 77 K) is the recommended adsorptive for determining the surface area and mesopore size distribution, but it is necessary to employ a range of probe molecules to obtain a reliable assessment of the micropore size distribution.

* An alternative technique to gas adsorption (e.g. mercury porosimetry) must be used for macropore size analysis.

* Krypton adsorption (at 77 K) is usually adopted for the determination of relatively low specific surface areas (< 2 m²g-l), but this technique cannot be employed for the study of porosity.

f) Choice of adsorptive :

Depending on the adsorptive and on temperature and pressure, the area of surface occupied by one molecule () varies. A rough mean value derived from assumption of close packing at the surface is given by :

3/2

2/1)2(*4*866.0*4

AN

M

With M the molecular weightAnd the density of liquid adsorbate

Ex:

N2 = 16.2 ŲAr = 14.2 ŲKr = 21.0 Ų

45

3) Porous volume measurements

a) Density measurements using Hg-He volumetry

The pore volume (Vp) is the total volume of the pores of 1g of solid. It results from the difference between the apparent volume (Va) of 1g of solid and the real volume (Vr) occupied by the molecules contained in 1g of solid.

Vp = Va – Vr = (1/a – 1/r)

cm3.g-1

With a the apparent densityAnd r the real density

1) He expansion in a closed chamber of volume V (cm3) containing a mass W (g) of solid

He occupies the intergranular volume (Vg) and penetrates in the pores (Vp) but doesn’t adsorb.The real density is given by :

)( Pgr VVV

W

2) Hg filling. After He evacuation, the chamber is filled with Hg at 1 atm.

Hg occupies the intergranular volume (Vg) but doesn’t penetrate in the pores at 1 atm.The apparent density is given by :

ga VV

W

46

4) Pore size distributiona) Capillary condensation methods

1) Kelvin’s law 

•This law considers only the capillary condensation phenomenon (0.35 < P/P0 < 0.95).•A gas condensation on the surface of a cylindrical pore follows the law

is the superficial tension of the liquid adsorbate, rK is the pore radius (meniscus),P is the gas pressure and P0 the saturation pressure of the gas,VM is the molar volume of the condensed adsorptive and is the wetting angle. If the liquid perfectly wets the surface cos=1.

Hence at a given P1 there is a pore radius r1 such that all the pores with r < r1 are filled

and all the pores with r > r1 are empty.

cos

2ln

0 RTr

V

P

P

K

M

However, due to different meniscus shapes during adsorption and desorption, the pressure for emptying the pore is lower than condensation pressure, whichexplains the hysteresis loop on isotherms of mesoporous solids.

47

2) Barrett, Joyner and Halenda (BJH) model

This model completes the Kelvin approach by considering also the variation of the number of adsorbed layers. After evaporation of condensed liquid, adsorbed molecules remain as a film.

The pore radius r is the sum of the Kelvin radius rK plus the multilayer thickness t of the

adsorbed molecules forming a condensed film: r = t + rK

The law t = f(P/P0) was the object of many studies, but generally the law used for N2 is the

following Hasley equation :

 

3/1

0

ln

554.3)(

PP

nmt

Ref : G.D. Hasley, J. Chem. Phys., 16, 931 (1948)

t

rK

r

Cylindrical porefilled with a film (t)of condensed liquid

From an elementary variation of P (dP), the elementary pore radius (dr) and the elementary pore volume (dVP) are calculated. The pore size distribution is given by (dVp/dr) = f(r)

48

However, the shape of the pore is not always cylindrical and the shape of the hysteresis loop on the isotherms gives some information about the shape of the pores.

Isotherms are classified following their shape by J.H. de Boer

Ref : J.H. De Boer, ‘The structure and Properties of Porous Materials’, D.H. Everett et F.S. Stone Ed., London, (1958) 68

3) Influence of the pore shape

49

4) Hg porosimetry

A non-wetting liquid (2 < < ) requires a positive excess hydrostatic pressure P to be applied to enable it to enter pores of radius r ; P will vary inversely with r.

This principle is the basis of mercury porosimetry :A weighed sample is enclosed in a stout metal bomb, and evacuated to remove air from the pores. Mercury is then admitted to fill the cell and surround the sample, and is subjected to progressively increasing pressures, applied hydraulically. At each pressure P, the corresponding volume V of mercury contained in the cell is measured.

It is assumed that as the pressure is increased, mercury enters pores in decreasing order of size. Thus, if V is the volume intruded between P and P+ P, it will equal the volume of pores with radii between r and r - r, with

In this way, a volumetric distribution of pore sizes is obtained (dV/dr = f(r)).

Pr

cos2 : superficial tension of Hg (484 mN.m-1)

: contact angle (141°)

50

Example of pore size distribution

P/P0

Vads

Treatment of mercury intrusion data

Nitrogen adsorption-desorption isotherm and corresponding pore size distribution

51

IV. Chemisorption1) General

•Very exothermic (heat of adsorption up to 500 kJ/mol)

•There is creation of a chemical bond between adsorptive and adsorbent(electron transfert)

•Activated phenomenon(Ea required)

•Chemisorption can be dissociative :

H2 + 2 Pt 2 Pt-H

H2 + ZnO-Zn-H

-O-H

EaEd

H

52

2) Chemisorption isotherms

a) Langmuir theory

Assumptions :

•The surface of the solid contains a fixed number (Ns) of chemisorption sites•Only one adsorptive molecule at a time can occupy a site•The heats of adsorption of each site are supposed equal and independant of the coverage •There is no interaction between adsorbed molecules

Similarly to the equation obtained for physisorption, the fraction of the sites which are occupied is given by Langmuir equation

With : N = number of sites currently occupiedNs = total number of accessible sitesb = coefficient depending on H and TVa = chemisorbed volumeVm = volume of chemisorbed monolayer

bP

bP

V

V

N

N

m

a

s

1

Vm is obtained for = 1 (all accessible chemisorption sites are occupied)

53

b) Real isothermAs seen before, chemisorption may yield Type I (Langmuir) adsorption isotherm :

However in practice, a true Type I isotherm is rarely obtained since Va usually continues to increase as P increases after saturation of the active sites.

This method involves a combination of physisorption (reversible) and chemisorption (irreversible)

To differenciate the chemisorption from the physisorption contribution, the sample is evacuated after a first saturation, which removes only the physisorbed gas.

Then the analysis is repeated but this time the sample is already saturated with chemisorbed molecules : only physisorption is measured.

The difference A-B corresponds to chemisorptionand follows Langmuir model.

54

c) Number of accessible active sites (Ns)

From the Langmuir equation : •Vm is obtained by plotting (P/Va) = f(P)•Ns could be obtained by plotting (P/N) = f(P)

bP

bP

V

V

N

N

m

a

s

1

Knowing : • the monolayer volume of chemisorbed molecules (Vm)• the stoichiometric factor Fs (number of surface atoms interacting with one adsorptive molecule)The number of accessible active sites Ns is obtained by :

0V

FNVN sAm

s With NA = Avogadro’s number

V0 = molar volume of adsorptive

Fs depends on the adsorptive and on the adsorbent ; it can often be found in literature

Ex : For H2 Fs = 2 on Co, Ni, Pt…For CO Fs = 1,15 on Ni

Ni=COand

Ni

NiCO(mean value)

because are formed

Fs = 2

Fs = 1

55

d) Metal dispersion

Usual catalysts are composed of an active metal deposited on an oxide (inactive) support.The total amount of active metals added to the support is known from the preparation method.

Due to high calcination (or pretreatment or reaction) temperatures, some of the active metal may become inaccessible to or inactyive toward the reactant molecules due to migration into the support bulk, formation of a new product, sintering toward formation of larger particles…All these processes lead to a decrease in the active surface area.

It is extremely important for catalysis to determine the accessible quantity of active species since it is related directly to the overall performance of the catalyst and the efficiency of the activation procedure.

The fraction of the total active metal sites (NT) which are accessible to the reactant is called dispersion.

Dispersion D is given as a percentage :

100*(%)T

S

N

ND

56

V. Experimental part

1) Experimental procedure

a) Sample preparation

The sample should be finely crushed (higher specific surface area) and degased in order to evacuate the gases initially adsorbed on the sample. This surface cleaning (degassing) is most often carried out by placing (about 50 mg) of the solid in a glass cell and heating it to 350°C (at least 3h) under vacuum or flowing inert gas.

If this degassing is performed on another apparatus than the adsorption measurements, the sample should be kept under vacuum in a special cell until adsorption measurements.

The quantification of the adsorbed gas can be performed using various techniques :•Gravimetry•Volumetry •Gas Chromatography•Thermic conductivity

b) Adsorptive analysis

57

2) Apparatus used for adsorption

a) Scheme

58

b) Picture Once clean, the sample is brought to a constant temperature by means of an external bath. Then adsorptive pressure is slowly increased and adsorption takes place.

59

In conclusion :

-physisorption is performed mainly with N2 at 77K (Tcondensation at 1atm)

determination of specific surface area S (support and metal) accessible to the gas reactant.

Supports with high S values permit to prepare catalysts with higher metallic dispersion

-chemisorption is performed with a suitable gas (H2, CO) at higher temperature (around room temperature) depending on the affinity of the gas for the metal and of the dissociation stoechiometry (Fs). Chemisorption takes place on metal

determination of the number of active metallic sites (and dispersion) High metallic dispersion permits higher activity of the catalyst

60

1.S. Brunauer, "Physical Adsorption" (Princeton University Press, Princeton, N. J., 1945) 2.P. Atkins, "Physical Chemistry" (Freeman, New York, 1978) 3.G. A. Somorjai, "Principles of Surface Chemistry (Prentice-Hall, Englewood Cliffs, N. J. 1972) 4.R.S. Drago, C.E. Webster, and J.M. McGilvray, J. Am. Chem. Soc., 1998, 120, 5385.Nelson FM and FT Eggertsen : Determination of surface area, Adsorption measurements by a

continuous flow method. Anal Chem, 30:1387-1958.6.Qadeer R, S Akhtar, J Hanif and A Majeed : Nitrogen adsorption on La2O3 (La=Pr, Nd, Gd, Dy, Er)

powder by continuous flow method. Sci Intl (LHR), 6:215-217, 1994.7.Brunauer S, LS Dening, WS Dening and E Teller : On a theory of the van der Waals adsorption of

gases. J Am Chem Soc, 62:1723-1732, 1940.8.Qadeer R, J Hanif, M Saleem and M Afzal : Characterization of activated charcoal. J Chem Soc Pak, 16:229-235, 1994.9.Stoeckli HF, JP Houriet, A Perret and U Huber : In Characterization of Porous Solids, Eds SJ Gregg, KSW Sing, HF Stoeckli, London Society of Chemical Industry, P 31, 1978.

References

61

Part 2 : Diffraction techniques Part 2 : Diffraction techniques

X-Ray Diffraction

62

OutlineI. Basics in crystallography1) Nature of solids2) Crystal structures3) Bravais space lattice4) Indexing of planes and directions

II. X-Ray Diffraction (XRD)1) What are X-rays?2) Production of X-rays3) Principle of X-ray diffraction by a particle4) Principle of X-ray diffraction by solids (crystals)5) Reciprocal lattice6) Diffraction pattern7) Ewald construction

III. Methods used to obtain a XRD pattern1) Rotating Crystal Method2) The Laue Method3) The Powder Method

IV. Experimental part1) Apparatus2) Preparation of the sample3) Standard acquisition4) Determination of crystalline phases5) Determination of mean crystal size

V. Applications of XRD

63

I. Basics in crystallography

1) Nature of solidsSolids can be described as :

• amorphous : the atoms are arrangedin a random way similar to The disorder we find in a liquidEx : glasses are amorphous materials

• crystalline : the atoms are arranged in a regular patternThere is as smallest volume element, called unit cell,that by repetition in three dimensions describes the crystal

The unit cell is described by - three axes (base vectors) a, b and c - three angles , and

Ex : quartz, NaCl, perovskitesAbout 95% of the solids can be described as crystalline

Quartz crystal

64

2) Crystal structures

7 crystal structures (primitive cell)

Examples of Bravais lattice for cubic structures

Addition of atoms

14 Bravais space lattice

65

http://epswww.unm.edu/xrd/symmetry.pdf

3) Bravais space lattice

14 Bravais space lattice

•P - Primitive: simple unit cell •F - Face-centered: additional point in the center of each face •I - Body-centered: additional point in the center of the cell •C - Centered: additional point in the center of each end •R - Rhombohedral: Hexagonal class only

66

14 Bravais space lattice

Cubic

Tetragonal

Hexagonal

Orthorhombic

Monoclinic Triclinic

http://www.uwgb.edu/dutchs/symmetry/bravais.htm

P

Rhombohedral

67

In the lattice of a crystal, there are many different planes with different spacings. The interplanar distance d(hkl) is the minimal spacing between two consecutive planes of the same family (hkl).

The plane (hkl) is indexed using the coordinates of the vector starting at point (ooo) and ending at point (hkl), which is in direction [hkl] and perpendicular to the plane (hkl).

4)

h, k, l are called Miller indexes

68

5) Aims of XRD characterization of catalysts

Heterogeneous catalysts are often composed of one of several metalsdeposited on a support (oxide) like Ni / Al2O3 or Ni-Rh / Al2O3 (reforming catalysts)

The amount of each element is known from elemental analysis, but thecrystalline structures formed during preparation must be known in orderto understand the activity of the catalyst.

For example, the same elements could create other phases like spinels NiAl2O4,much less active in reforming catalysis than metallic Ni. If we do not know that Ni isinserted in a spinel structure, we could conclude that Ni is a bad active phase.

It is also important to know the environement of the Ni in Ni-Rh / Al2O3 catalyst :Question of Ni-Rh alloy formation or not? What is the active phase?

The crystalline structure of the catalyst is important because it influencesthe catalytic activity.

69

II. X-Ray Diffraction (XRD)

X-rays are electromagnetic radiation (like light rays) but with much higher frequency and thus a much higher energy and much lower wavelength

Wavelength : 0.02 < < 0.2 nmEnergy : E = h.c /

X-rays were discovered in 1895 by Röntgen

1) What are X-rays?

70

2) Production of X-rays

X-rays are emitted when electrons from L layer relax to K layer

71

3) Principle of X-ray diffraction by a particle

An electron in an alternating electromagnetic field will oscillate with the same frequency as the field.

When an X-ray beam hits an atom, the electrons around the atom start to oscillate with the same frequency as the incoming beam. The particle scatters the incident beam uniformly in all directions.

72

In a solid, in almost all directions there will be destructive interference, that is, the incoming waves are out of phase and there is no resultant energy leaving the solid sample.

However the atoms in a crystal are arranged in a regular pattern, and in a few directions there will be constructive interference. The waves will be in phase and there will be well defined X-ray beams composed of a large number of scattered rays mutually reinforcing one another.

4) Principle of X-ray diffraction by solids (crystals)a) Introduction

73

• Consider first a single plane of regularly spaced atoms:• Imagine a beam of coherent light is incident on the atoms at an angle IN.

Some of the rays interact with the atoms and are scattered in all directions. (Most of the rays are transmitted.)  

• Consider the two scattered waves, A and B. They are in phase, reinforcing each other to give a diffracted beam, only when they travel the same distance, i.e. when x = y. This only occurs for scattered waves with an outgoing angle of

OUT = IN

• Thus a diffracted beam from a single row of atoms is made up of all the waves which are scattered with an outgoing angle equal to the incoming angle of the incident waves. This is true for incident waves of any wavelength.

b) Diffraction by a single layer of atoms

74

Consider many layers of regularly spaced atoms, such as it might be encountered in a crystalline material

It is interesting to know what spacing of the layers of atoms will give rise to scattered waves being in phase, when interacting with many layers of atoms.

• Consider two waves C and D, scattered from particles in adjacent planes separated by a distance d. They are only in phase if the extra path length of wave D over C (= x + y) equals a whole number of wavelengths.  

• The equation for this path difference gives the Bragg law: x + y = 2dsin = n

c) Diffraction by many layers of atoms

75

Bragg’s law

76

Consider a beam of waves to be incident on a crystal. The measurement, r, made from the diffraction pattern, tells us about 2B. From this we can deduce 1/d, since

we know that for diffraction of radiation of wavelength, , the Bragg angle, B, increases with 1/d.

6) Diffraction pattern

77

The relation by which diffraction occurs is known as the Bragg law or equation.

Because each crystalline material has a characteristic atomic structure, it will diffract X-rays in a unique characteristic pattern.

Possible characterization of crystals and crystalline phases in solid catalystusing X-ray diffraction

n. = 2 d(hkl) sin

78

Diffraction can occur whenever Bragg’s law is satisfied. With monochromatic radiation, an arbitrary setting of a single crystal in an x-ray beam will not generally produce any diffracted beams. There would therefore be very little information in a single crystal diffraction pattern from using monochromatic radiation.

This problem can be overcome by continuously varying or   over a range of values, to satisfy Bragg's law. Practically this is done by: • using a range of x-ray wavelengths (i.e. white radiation), or• by rotating the crystal or, using a powder or polycrystalline specimen.                           

3 methods•Rotating crystal method•Laue method•Powder method

III. Methods used to obtain a XRD pattern

79

In the rotating crystal method, a single crystal is mounted with an axis normal to a monochromatic x-ray beam. A cylindrical film is placed around it and the crystal is rotated about the chosen axis. 

As the crystal rotates, sets of lattice planes will at some point make the correct Bragg angle for  the monochromatic incident beam, and at that point a diffracted beam will be formed.  

The reflected beams are located on the surface of imaginary cones. When the film is laid out flat, the diffraction spots lie on horizontal lines.

1) Rotating Crystal Method

80

2) The Laue Method

The Laue method is mainly used to determine the orientation of large single crystals. White radiation is reflected from, or transmitted through, a fixed crystal.

The diffracted beams form arrays of spots, that lie on curves on the film. The Bragg angle is fixed for every set of planes in the crystal.

Each set of planes picks out and diffracts the particular wavelength from the white radiation that satisfies the Bragg law for the values of d and involved. Each curve therefore corresponds to a different wavelength.

The spots lying on any one curve are reflections from planes belonging to one zone. Laue reflections from planes of the same zone all lie on the surface of an imaginary cone whose axis is the zone axis.

81

Back-reflection Laue

In the back-reflection method, the film is placed between the x-ray source and the crystal. The beams which are diffracted in a backward direction are recorded. One side of the cone of Laue reflections is defined by the transmitted beam. The film intersects the cone, with the diffraction spots generally lying on an hyperbola.

Transmission Laue

In the transmission Laue method, the film is placed behind the crystal to record beams which are transmitted through the crystal. One side of the cone of Laue reflections is defined by the transmitted beam. The film intersects the cone, with the diffraction spots generally lying on an ellipse.

82

Crystal orientation is determined from the position of the spots. Each spot can be indexed, i.e. attributed to a particular plane, using special charts. The Greninger chart is used for back-reflection patterns and the Leonhardt chart for transmission patterns.

The Laue technique can also be used to assess crystal perfection from the size and shape of the spots. If the crystal has been bent or twisted in anyway, the spots become distorted and smeared out.

83

The powder method is used to determine the value of the lattice parameters accurately. Lattice parameters are the magnitudes of the unit vectors a, b and c which define the unit cell for the crystal.

3) The Powder Method

If a monochromatic x-ray beam is directed at a single crystal, then only one or two diffracted beams may result.

If the sample consists of some tens of randomly orientated single crystals, the diffracted beams are seen to lie on the surface of several cones. The cones may emerge in all directions, forwards and backwards.

A sample of some hundreds of crystals (i.e. a powdered sample) show that the diffracted beams form continuous cones. A circle of film is used to record the diffraction pattern as shown. Each cone intersects the film giving diffraction lines. The lines are seen as arcs on the film.

84

Example : powder diagram of MgO (cubic centerd faces) obtained with Cu-K in a Debye-Scherrer chamber of 180 nm (fotographic film)

2S=4r

85

IV. Experimental part

1) Apparatus

A XRD apparatus is basically composed of three parts :

- the X-ray source

- the sample

- the detector

fixed in a fotographic chamber

located on a goniometer

a) General

86

b) Source of X-rays :- X-ray tube

high speed e- are slown down by an anode and emit X-rays(not polarised)

- synchrotronproduction of e- with parallel spins (polarised X-rays)

87

•X-rays are electromagnetic radiation similar to light, but with a much shorter wavelength.

•They are produced when electrically charged particles of sufficient energy are decelerated. In an X-ray tube, the high voltage maintained across the electrodes draws electrons toward a metal target (the anode).

•X-rays are produced at the point of impact, and radiate in all directions.

•Tubes with copper targets, which produce their strongest characteristic radiation (K1) at a wavelength of about 1.5 angstroms, are commonly used for geological or solid state chemistry applications.

88

c) X-ray detectors

• Qualitative detectors : fluorescent screen or fotographic film• Quantitative detectors :

1. Gas ionisation detectors2. Solid detectors (scintillation counter, semi-conductor detector…)

scintillation counter semi-conductor detector

89

d) Sample part

The simpliest apparatus is a fotographic chamber.

Debye-Scherrer chamber Guinier chamber

Polycrystalline sample Monocrystal sample

1) Fotographic chambers

X-ray source

Sample

Diffraction coneFotographic film

Collimator

X-ray source

Crystal

Sample

90

•The apparatus which is nowadays the mostly used is the powder diffractometer with Bragg-Brentano focuser.

•The basic geometry of an X-Ray Diffractometer involves a source of monochromatic radiation and an X-ray detector situated on the circumference of a graduated circle centered on the powder specimen.

•Divergent slits, located between the X-ray source and the specimen, and divergent slits, located between the specimen and the detector, limit scattered (non-diffracted) radiation, reduce background noise, and collimate the radiation.

•The detector and specimen holder are mechanically coupled with a goniometer so that a rotation of the detector through 2 degrees occurs in conjunction with the rotation of the specimen through degrees, a fixed 2:1 ratio.

2) X-Ray Powder Diffractometer

91

92

Diffractometer D8-advance Bruker

93

A curved-crystal monochromator containing a graphite crystal is normally used to ensure that the detected radiation is monochromatic.

When positioned properly just in front of the detector, only the K radiation is directed into the detector, and the Kß radiation, because it is diffracted at a slightly different angle, is directed away.

The signals from the detector are filtered by pulse-height analysis, scaled to measurable proportions, and sent to a linear ratemeter for conversion into a continuous current. Common output devices include strip-chart recorders, printers, and computer monitors.

94

2) Preparation of the sample

The sample is ground into fine powder and deposited on a glass supportFew drops of ethanol are spread above the powder and mixed to obtain a homogeneous suspension.Ethanol evaporates rapidly and the powder is then well spacially distributed on the support.

A varnish can be used to protect the sample from oxidation or the sample can be placed under an inert atmosphere in a chamber equipped with X-ray transparent windows.

3) Standard acquisition

Continuous scanning with 2 variation from 10 to 90° with 1s step every 0.02°

95

Examples of diffractogram obtained by powder method

96

4) Determination of crystalline phasesComparison of the diffractogram of the sample with those of JCPDS(Joint Comittee on Powder Diffraction Standards) Databank.

97

5) Determination of mean crystal size

Width of the peaks varies inversely proportional to cristal size

High crystral syze lead to very fine diffraction peaks

98

Diffraction patterns can be used in order to :

•Identify the crystalline compounds and their phases in a powder by refinement of the complete XRD pattern according to Rietveld(comparison with JCPDS files containing lots of compounds diffraction patterns)

•Measure the average spacing between layers or rows of atoms

•Determine the precise crystallographic lattice constants in particular for solid solutions (incorporation of atomic species into a host lattice)

•Determine the orientation of a single crystal or grain

•Find the crystal structure of an unknown material

•Measure the size, shape and internal stress of small crystalline regions, degree of cristallinity of rapid solified materials

V. Applications of XRD

99

Examples of XRD applied to catalyst characterization1) Preferential orientation

Polycrystalline gold film with statistical orientation

Mica muscovite with preferential orientation of the cystals in (00l) direction

100

2) Resolution between two phases

Profile analysis showing the resolution of the tetragonal and monoclinic phases in a Zr(Y)O2 ceramic

101

0

2000

4000

6000

8000

10000

20 30 40 50 60 70 80 90

Inte

ns

ité *

*

** *

* * *

* **

** * * a

b

2

* LaFeO3

(f. JCPDS 37-1493)

Example of perovskite solid solution LaNixFe1-xO3

XRD of 2 perovskite catalysts : LaNi0,3Fe0,7O3 (a) et de LaNi0,7Fe0,3O3 (b)

Formation of a unic perovskite structure

3) Solid solution

Ref : H. Provendier et al., Applied Catalysis I, 180 (1999) 163-173

102

0

500

1000

1500

2000

2500

3000

3500

4000

32 32,2 32,4 32,6 32,8 33 33,2 33,4

Inte

ns

ité

* °°

a

b

Zoom into the maximum diffraction peak of perovskite structure

XRD of LaNi0,3Fe0,7O3 (a) et LaNi0,7Fe0,3O3 (b)

Deviation of the peaks with increasing Ni content

103

XRD of LaNixFe1-xO3 family (0 ≤ x ≤ 1)

* LaFeO3 (f. JCPDS 37-1493) et ° LaNiO3 (f. JCPDS 33-0711)

 

 XRD of LaNixFe1-xO3 for 32,0°  2  33,5° :

(a) LaFeO3 , (b) LaNi0,3Fe0,7O3 , (c) LaNi0,4Fe0,6O3 , (d) LaNi0,5Fe0,5O3

(e) LaNi0,7Fe0,3O3 , (f) LaNi0,8Fe0,2O3 , (g) LaNiO3 .

 Progressive deviation of the peaks with x

20

1000

2000

3000

4000

5000

6000

32 32,2 32,4 32,6 32,8 33 33,2 33,4

Inte

ns

ité

* °°

x=1

x=0,8

x=0,7

x=0,5x=0,4x=0,3

x=0

Inte

nsit

y

104

aj

ac i

i=1

j

1

ai2 = (hi

2 + ki2 + li

2) * di2

The six most intensive diffration peaks indexed in a cubic system : (100), (110), (111), (200), (211) et (220)

Average lattice parameter in cubic system

3,833,843,853,863,873,883,893,903,913,923,93

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

x

a (

A°)

Evolution of the average lattice parameter ac versus Ni content x

105

References

•J.P. Sibilia, A guide to materials characterization and chemical analysis, Second Edition, VCH Publishers Inc., 1996

•B. Imelik and J.C. Vedrine, Catalyst characterization, Physical techniques for solid materials, Plenum ¨Press, New York, 1994

•J.-P. Eberhart, Analyse structurale et chimique des matériaux, Ed Dunod, Bordas, Paris 1989

• B. Imelik and J.C. Vedrine, Les techniques physisques d’étude des catalyseurs, Ed Technip, Paris 1988

• X-Ray Methods, C. Whiston, 1987.

•Modern Powder Diffraction, D.L. Bish and J.E. Post Eds., Reviews in Mineralogy Vol. 20,1989.

• Methods and Pratices in X-Ray Powder Diffraction, JCPDS-International Centre forDiffraction Data, 1989.

•Introduction to X-ray Powder Diffraction Diffractometry, R. Jenkins and R.L. Snyder, 1996.

106

Part 3 : Spectroscopies Part 3 : Spectroscopies

IR and UV spectroscopies applied to catalysis

107

Introduction to spectroscopy

The absorption of light is familiar to eveyone. The absorption of visible light is what makes things coloured. For example a blue dye used in a pair of jeans appears blue because the light at the red end of the spectrum is absorbed. This leaves the blue light to be reflected to the observer’s eye.

108

I. Interaction of electromagnetic radiation (light) with molecules

1) Molecule energy is quantizied

Molecules have different states characterized by differences in energy

Ground state

Excited state

109

2) Light spectrum

Light has energy determined by its wavelength,, or frequency,

c / and E = h (c = 3 x 108 m/s, the speed of light; h = Planck's constant)Thus, there is an inverse relationship between wavelength and Energy. In the diagram below, gamma rays have the largest energy and radiowaves the smallest.

<-----Energy

110

If light striking a molecule in State 1 has an energy (h.) equal to the difference in energy between molecular State 1 (E1) and State 2 (E2), there is a probability that a molecule will absorb a photon and jump to the higher energy state, State 2.

The molecule can jump back to state 1 by: 1) reemitting a photon of the same energy (or lower energy in the case of fluorescence) or2) transferring energy to another molecule

E2

E1

E2

E1

E = E2 – E1 = h.

3) Light absorption by molecules

111

Light waves are described by sinusoidally oscillating electric vectors, E, and perpendicular magnetic vectors, H. Hence, they are called Electromagnetic radiation

             The probability of absorption is related to the orientations of E (and also H) vectors with respect to the molecule. For now, assume that light waves and molecules are randomly oriented.

4) Light : electromagnetic radiation

112

5) What magnitudes of energies are involved?

113

OutlineI. Introduction to IR-spectroscopy1) Brief historical background2) Principle of the techniques3) Uses of Infrared spectroscopy4) Nature of the samples

II. Theory about molecular vibrations1) Introduction2) First approach 3) Real diatomic molecule4) The vibrations of polyatomic molecules

III. Interaction of light with a sample1) Beer-Lambert’s law2) Nature of interactions

IV. Experimental part1) The apparatus2) Mathematical treatment : Fourier Transform3) IR spectrum : Example of CO2

4) Methods used for IR spectroscopy

V. Applications of IR spectroscopy

IR-spectroscopy

114

I. Introduction to IR-spectroscopy

1) Brief historical background

1905 Coblenz obtained the first IR spectrum1940 Terenin and Kasparov made the first attempt to employ IR in adsorption studies1952 KBr was used for solid discs preparations for powder analysis1960s the era of Fourier transfom IR (FT-IR) began

2) Principle of the technique

The principle of IR spectroscopy is similar to UV spectroscopy but the energy scale is much lower (2-40 kJ/mol) and the wavenumber (200-4000 cm-1).The energies correspond to transition energy between different vibrationnal (and rotational) states of molecular bond.

J. Ryczkowski, Catalysis Today 68 (2001) 263-381

115

3) Uses of Infrared spectroscopy

IR spectroscopy can be used to :

•Identify materials•Determine the composition of mixtures•Provide information helpful in deducing molecular structure

(functional groups…)•Give information about superficial properties of solid catalysts

(surface hydroxyl groups, acidity…)•Monitor the course and extent of reactions (in situ IR)

(reactional intermediates, changes in surface properties…)

4) Nature of the samples

Materials in solid, liquid or gaseous state may be studied by IR spectroscopy. A convenient sample size is several milligrams, but spectra can be obtained from some picograms with special techniques.

116

Atoms in molecules and solids do not remain in fixed relative positions but vibrate about some mean position. This vibrational motion is quantized and at room temperature, most of the molecules are in their lowest vibrational state.

II. Theory about molecular vibrations

Absorption of electromagnetic radiation with the appropriate energy (infrared light) allows the molecules to become excited to a higher vibrational level. The absorption of infrared light as a function of wavelength give rise to an infrared spectrum with specific fingerprints, assignable to particular molecular entities.

1) Introduction

h

117

2) First approach : model of harmonic oscillator for diatomic moleculea) classical mechanics

m1m2

Schematic representation of a chemical bond considered as a spring, wich exerts a force F on an atom

F = - k (r – req)

The potential energy close to the equilibrium position is V(r)

The application of the fundamental dynamic principle leads to

With the reduced mass of the molecule21

21 *

mm

mm

With k the force constant (N.m-1)r : interatomic distancereq : interatomic distance at equilibrium

r

118

The resulting vibrational energy is

)²(2

1eqvib rrkE

And the wavenumber

= =c

The wavenumber depends on two parameters :

•Reduced mass mWavenumber decreases with heavier atoms

Ex :

•Force constant kWavenumber increases with stronger bonds

Ex:

119

b) Quantum mechanics applied to diatomic molecule

The harmonic vibration of a diatomic molecule is treated by solving the Schrödinger equation :

The solution of this equation is that vibrational energy is quantized according to :

k

2

1

and n an integer numbern = 0, 1, 2…called vibrational quantum number

At any temperature T, the molecular vibrations are distributed over the energy levels according to Boltzmann’s law :

kT

hnNN Tn

21

exp. Nn is the population of vibrational level n

)2

1( nhEn

120

3) Real diatomic molecule : anharmonic oscillator

The deviations of the real potential energy curve from that of the harmonic oscillator is called anharmonicity and is due :

- to electronic repulsion for r < req

- to electronic rearrangements as the atoms move further and further apart for r > req

The treatment of anharmonic ocillators leads to the quantum mechanical expression of energy :

hxnnhE an )²2

1()

2

1( Xa is called anharmonicity constant

harmonic oscillator

anharmonic oscillator

121

How is it possible to extract information about the composition of a compound Using IR spectroscopy?

When a IR radiation is sended to a sample, each molecule will absorb part of the radiations having convenient wavelength so that it can be excited to a higher vibrational and rotational state.

To each functional group of a molecule correspond defined vibrations and rotations, which depend on the geometry of the molecule, number of atoms, nature of atoms, environment…and which are characteristic of the functional group.The energies involved for these vibrations can be known by calculation or experiments.

If we can determine the energy absorbed by the sample in a wide enough range of IR wavelengths, we should be able to identify the functional group responsible for each defined absorption and determine the composition of the sample.

This is the aim of IR spectroscopy

122

4) The vibrations of polyatomic moleculesWhich are the characteristic vibrations of molecules?

a) number of normal vibration modes

When N independently moving atoms are combined in a molecule, in 3D spacethe 3N degrees of freedom are transformed into :

- 3 degrees of translational freedom- 3 degrees for rotation (or only two for linear molecules)- the rest for vibrations

Molecule with N atoms

Linear : 3N – 5 vibration modes

Non-linear : 3N – 6 vibration modes

123

b) name of the vibration modes

stretching

1) Bond deformation

2) Angle deformation* in the plane

symmetricalstretching

antisymmetricalstretching

* out of the plane

wagging

twisting

3) Ring deformation

breathing

bendingor scissors

s

as

rocking

Except for rings,molecules N – 1 bondswith N – 1 stretching

N atoms 2N – 5 (or 4) angle vibrations

124

Example 1 : H2O vibration modes

Non-linear molecule with 3 atoms 3*3-6=3 normal vibration modes

(3657 cm-1) (1595 cm-1) (3756 cm-1)

2 stretchingand 1 bendingvibrations

125

Example 2 : CO2 vibration modes

Linear molecule with 3 atoms 3*3-5=4 normal vibration modes

2 stretchingand 2 bendingvibrations

126

d) Criteria for bond assignement in IR spectroscopy

1) Table of group frequencies2) Strength of the bonds 3) Stretching versus bending vibrations4) Mass of the atoms

c) Rules in IR spectroscopy

1) The force constant of a stretching vibration is larger than that of a bending vibration

frequency of stretching > frequency of bending for a group2) Frequency increases if the mass of the atoms decreases

3) Frequency increases if the bond strength increases4) Normal vibrations may have the same frequency

they are degenerate5) Some vibrations are forbidden according to Laporte’s selection rule :

vibrations which induce a change in dipole moment are allowed in IR(homonuclear diatomic molecules like O2, N2… are inactive in IR)

127

128

Examples :

129

Possible identification of functional groups using IR

Applications of IR spectroscopy :

•Identify materials•Determine the composition of mixtures•Provide information helpful in deducing molecular structure

(functional groups…)•Give information about superficial properties of solid catalysts

(surface hydroxyl groups, acidity…)•Monitor the course and extent of reactions (in situ IR)

(reactional intermediates, changes in surface properties…)

How can it be made practically?

130

III. Interaction of light with a sample

Sample

DetectorThe relation between I and I0

is given by the Beer-Lambert LawDefining A absorbance

log (I0/I) = A = c l

When light with an incident intensity I0 is directed to a sample, a part is absorbed by the sample I0-I and the intensity of the transmitted light I is a function of the sample thickness l, of the concentration c of absorbing molecules in the sample and of the exctinction coefficient

1) Beer-Lambert’s law (case of diluted liquid)

Measurements of I and IO permit to access to the absorbed part

131

• It is possible to extract absorbing properties of the sample from the transmitted light

• However reflection and scattering should not always be neglicted

2) Nature of interactions between light and sample

132

fraction of reflected light can be eliminated using a referencemeasurement with same materials (cuvette + solvent or diluting species)

a) How to limit reflection?

Reference

Sample

133

• scattering is negligible in molecular disperse media (solutions)

• scattering is considerable for colloids and solids when thewavelength is in the order of magnitude of the particle size

• scattering is reduced through embedding of the particles in media with similar refractive index: KBr wafer…

b) What about scattering ?

Scattering is important for solids (3-40 mm) : Can we use it ?

134

Material transparent in some areas of IR wavelengths

135

IV. Experimental part

1) The apparatus2 types of spectrometers :

a) Dispersive IR instrument

For many decades, the workhorse of the IR laboratory was the light dispersive spectrometer.While still used for routine experiments, it has been superseded by the Fourier TransformInfrared (FT-IR) spectrometer, which employs an interferometer in place of a monochromator.

IR source(SiC)

136http://spectroscopy.lbl.gov/FTIR-Martin/FTIR-Martin_files/frame.htm

b) FT-IR spectrometer with Michelson interferometer

137

2) Mathematical treatment : Fourier Transform

In continuous scan mode the moving mirror is repetitively scanned as fast as 1/20e secondto generate the Fourier transform of the IR spectrum. After coaddition of many scans (meanly 32)To improve the signal-to-noise ratio, the Fourier transform is converted to the IR spectrum by a dedicated computer.

dxxxGg )2cos()()( The Fourier transfom of G(x) is

138

3) IR spectrum : Example of CO2

as CO2349cm-1

s CO1537cm-1

CO667cm-1

as CO2349cm-1

s CO1537cm-1

CO667cm-1

absent

139

4) Methods used for IR spectroscopy

a) Transmission spectroscopy

Transmitted light is analyzed

140

Uses of transmission spectroscopy

This technique can be used for gas, liquid or transparent solid samples

For liquids, 1cm long cells with 2 transparent faces are used (CaF2)For gases, several centimeter long cells with 2 windows are used.

In the particular case of solid catalysts, the sample is mostly as a powder and has to besieved and pressed together to form a thin pellet (diameter about 18mm, thickness 10-2 – 5.10-2 mm)

The pellet can be placed on a quartz support in a special cell, which allows to keep the sample under controlled atmosphere and controlled temperature

Part A is equipped with IR transparent windows, where the pellet is maintained during spectrum recordPart B (in quartz or pyrex) is used for thermic treatment

141

Problem : not transparent enough samples can not be analysed using transmission !

Possible dilution :it can be diluted in a transparent and inert compound like KBr, SiO2…

Other techniques : use of reflectionIncident light should not be perpendicular to the sample surfaceI and Io not in the same direction

142

b) Reflection spectroscopy

Instead of measuring the transmitted light, it is possible to measure thereflected light. This is more convenient for dark samples.

Can we extract the absorption properties of our sample from thereflected light?

143

particle diameter d >

•light partially reflected on elementary mirrors (crystal facets) inclined statistically at all possible angles to the macroscopic surface

•light penetrates into sample, undergoes numerous reflections, refractions and diffraction and emerges finally diffusely at the surface

particle diameter d ≤

•scattering occurs•multiple scattering should produce isotropic distribution

1) Particle size effect :

Possible analysis by specular reflection

Possible analysis by diffuse reflection

144

2) Diffuse Reflectance (DRIFT)

Diffuse reflectance is the radiation, which penetrates into the sample and then emerges.

This technique needs :•element that collects diffusely reflected light•to avoid specularly reflected light•reference standard (white standard)

Used forLow reflecting samples

Powder particle diameter <

145

Cooling water (in) Cooling water (out)

Pictures of DRIFT cell

Scheme of DRIFT cell

Well adapted to in situ characterisation in heterogeneous catalysis

146

Diffuse Reflectance (DRIFT)

This system is designed so that diffuse reflectance is optimized and specular minimized.

147

3) Specular reflection

This is a mirror-type reflection

• fraction of reflected light increases with • depends on and ratio of the refractive indices (Snell law)

This technique is used for the study of thin metallic films, monocrystals,Surfaces with low rugosity (<<)…

n0 sin = n1 sin

148

4) Attenuated Total Reflectance (ATR)

This method permits to study non transparent samples (fine powder or liquids).The sample is held in optical contact with a prism (transparent monocrystal with high refraction index (N>24) like Ge, ZnSe, CaF2) and the light entering the prism undergoes internal reflection at the prism surfaces and penetrates into the film. So the exiting light is attenuated at specific frequencies corresponding to absorption by the film sample.

5) Photoacoustic Spectroscopy (PAS)

In PAS the IR energy absorbed by the sample is converted into heat. Thermal conduction to the sampleSurface induces expansion of gas in contact with it and produces an acoustic signal (proportional to theabsorption) when the incident beam intensity is modulated at a frequancy in the acoustic range.

Applications include analysis of surface coatings, alumina-based catalysts…, as well as obtaining theIR spectrum of opaque materials.

149

6) Recent development : optic fibers

This system allows to follow a reaction in liquid phase in situ

light conducted through total reflectance

150

Large number of applications of IR spectroscopy in catalysis and surface science

V. Applications of IR spectroscopy

(coupling)

151

IR spectra of CH4 adsorbed on modified MgO samples at 0.08 kPa and 88K

Example 1 : CH4 adsorption on modified MgO samples

Deviation of wavenumber in comparison to gas phase due to interaction with modified MgO :the interaction betweenmethane and the Cs-containing MgO catalyst is the strongest.

J. Ryczkowski / Catalysis Today 68 (2001) 263–381 281

as CH3

(gas 3020 cm-1)

Effect of dopingmaterials :

s CH3

(gas 2914 cm-1)

152

FT-IR spectra of pyridine adsorbed on zeolites (A) USY (a) and MCM-41 (b)And (B) amorphous silica-alumina (a) and MCM (b) degassed at (1) 200°C, (2) 300°C and (3) 400°C

(A) : USY has a higher amount of Brönsted acid sites than MCM-41 zeolite(B) : MCM-41 has a higher amount of Lewis sites than amorphous Si-Al

A B

Example 2 : evaluation of the acidity of Si-Al catalysts

Ref : A. Corma et al., J. Catal., 1994, 148, 569

153

Influence of the covering rate on CO vibration frequency

Influence of the composition of the 12CO-13CO on CO frequency

CO deviates from 2063 to 2100 cm-1 when the CO covering rate over Pt [111]increases.

1)Metal – CO bond strength decreasesand CO bond strength increases2)CO-CO dipolar interactions increases

CO deviates from 2063 to 2100 cm-1 when the 12CO content increases andthe 13CO content decreases

Due to CO-CO dipolar interactions

Ref : Crossley A.,King D.A., Surf.Sci., 68, 1977, 528

2 possible explanations :

Isotopic exchange should only influence CO-CO dipolar Interactions and not Me-CO bond strength

Here :

Example 3 : Isotopic dilution

154

Example 4 :Transition between physisorbed and chemisorbed CO on ZrO2 with increasing T

IR band at 2185 cm-1 observed at Ta = 300K corresponds to linear CO. At 300 K a purge leads to the disappearance of the IR band indicating that CO is reversibly adsorbed.The increase in T (spectra b–h) under CO, leads to a progressive decreasein intensity of the IR band alongside a slight shift to higher wavenumbers (2187 cm-1 at 340K and 2190 cm-1 at 378 K) due to chemisorption.

J. Ryczkowski / Catalysis Today 68 (2001) 263–381 281

155

Example of in situ IR measurementsJ. Ryczkowski / Catalysis Today 68 (2001) 263–381 281

156

Study of metallic alloy M1-M2

Geometric effect : Electronic effect :M2 addition leads only to an increase in the distances between CO-M1 vibrations

M2 addition modifies only the electronic density of M1

Ref : Toolenaar et al., J. Catal., 82, 1983, 1

Some alloys show different catalytic properties than pure metals. It can be due to :

x = 12CO/(12CO + 13CO)Shape of the curve CO versus x gives the informationTest : CO adsorption on M1 and alloy with isotopic dilution

157

Comparison : Transmission - Diffuse Reflectance

•spectra can have completely different appearance•transmission decreases, reflectance increases with increasing wavenumber

158

vibrations of surface species may be more evident in DRIFT spectra

Comparison : Transmission - Diffuse Reflectance

absolute reflectance Rabsorption coefficient kscattering coefficient s

Kubelka-Munk Function

159

In conclusion : What can we learn from IR?

160

W. WM. Wendlandt, H.G. Hecht, “Reflectance Spectroscopy”,Interscience Publishers/John Wiley 1966

B.M. Weckhuysen, P. van der Voort, G. Catana, “Spectroscopy of transition metal ions on surfaces”, Leuven University Press 2000

J.P. Sibilia, “A guide to materials characterization and chemical analysis”, Second Edition, VCH Publishers Inc., 1996

References

P. W. Atkins, « Physical Chemistry, Oxford University Press »

The science of spectroscopy : http ://www.scienceofspectroscopy.info.

Spectroscopy Now : http ://www.spectroscopynow.com.

161

Outline

I. Introduction to UV-spectroscopy1) General2) Shape of the peaks 3) Why using UV-spectroscopy ?4) Form of the samples :

II. Measurements1) Instrumentation2) Techniques3) Quantitatively

III. What Electronic properties lead to absorption of UV-Light?1) Example2) Nature of the electronic transitions

IV. Examples of UV spectroscopy applications in catalysis

UV-spectroscopy

162

1) General

1. UV (ultraviolet) radiation: = 190 - 400 nm Visible: = 400 - 800 nm

                                                                 2. This corresponds to approximately 400 kJ/mole of photons, the magnitude of energy in a covalent bond. 3. Energies of these magnitudes correspond to electronic transitions ==> promotion of an electron from one molecular orbital to an unoccupied molecular orbital of higher energy.

Absorption of a photon of light has a high probability of occuring when the orientation of the electric vector of the light wave, E, is oriented in a direction which will induce a movement of an electron from one molecular orbital to another oribital of higher energy.

Another way of stating this is to say that E is oriented along a Transition Dipole of the molecule.

E = h. = h.c / = 6.626.10-34*3.108 / 300.10-9 = 6.626 .10-19 J / photon

I. Introduction to UV-spectroscopy

163

For a single type of transition between two well-defined states, the Absorbance spectrum would be a very sharp peak.

In practice, there are a range of possible transitions from various vibrational energy levels of the ground state to various vibrational energy levels in the first excited state

==> a series of narrow absorption peaks of similar energies ; these cannot be resolved resulting in a single broad and sometimes complex peak for each type of electonic transition.

2) Shape of the peaks

164

3) Why using UV-spectroscopy ?

Ultraviolet/visisble spectroscopy is a useful analytical technique.It can be used :- to identify some functionnal groups in molecules- to assay (determine the content and strength of a substance :

trace metal content in alloys…)- to provide information about the electronic structure (especially for aromatic and transition metal compound)

165

4) Form of the samples :

Liquids and solids are analyed as neat materials or in dilute solutions, where a wide variety of solvents can be used

Vapors are examined directly

The sample for UV spectroscopy are prepared like those for IR spectroscopy

The sample size required is typically of the order of 1mg, but in special applications in the UV region, spectral information can be obtained from nanogram amounts

166

II. Measurements

1) Instrumentation

a) Single Beam Spectrophotometer

167

b) Split Beam Spectrophotometer

This system should have 2 detectors. It is possible to use only 1 detector and read I and Io successively

168

Two lamps are used :- a hydrogen or deuterium lamp for the ultraviolet region- a tungsten/halogen lamp for the visible region

169

A reference cell containing only reference material (gas, solvent…) is used.

For double beam instruments, light is passed simultaneously through the sample cell and the reference cell. The spectrometer compares the intensities of light passing through the sample and of light passing through the reference cell.

The transmitted or reflected radiation is detected and the spectrometer records the absorption spectrum by scanning the wavelength of light passing through the cells.

For single beam instruments, the principles are the same, but data on the reference are taken first, followed by the sample.

170

Fotograph of a UV-Vis spectrometer

171

It is possible to make measurements by :

1) Transmission :This technics is available for liquids, gases or transparent solidsLight is passing through the sample and the transmitted light is analysed.

Absorbance id defined as : A = log (Io / I)with I = intensity of transmitted beamAnd Io = intensity of incident beam

2) ReflectionThis method is used for solid samples (catalysts), surfaces (painting) or trouble suspensions, which are not transparent enough to be analysed by transmission. Two kinds of reflection exist :- specular reflection (mirror effect of the crystals)- diffusive reflection (diffusion of the light in all space)

Reflectance is defined : R = I / Io

2) Techniques

172

3) Quantitatively

For light passing through a solution, the rate at which photons are absorbed as a function of distance, l, through the sample cuvette can be thought of as a 1st order reaction and is proportional to the concentration, c, of absorbing molecules with a rate constant called '.Thus:

This is the Beer-Lambert Law

change to log10  

A = c l where A and are for a specific wavelength,

173

III. What Electronic properties lead to absorption of UV-Light?

1) Example : a simple molecule : Formaldehyde => H2C=O

a) Electronic Configuration:

C -- 1s2 2s2 2py1 2px

1 2pz0 ----->1s2 (3sp2)3 2pz

1

O -- 1s2 2s2 2py2 2px

1 2pz1

C uses 3 sp2 hybrid orbitals to form 3 -bonds with O and the 2 H's ; the remaining 2pz orbital forms a bond with O. O has the 2py atomic orbital

which is not involved in bonding, and it contains a non-bonding pair of electrons.

174

b) Molecular Orbital Diagram only higher energy orbitals are shown

n = non-bonding orbital (O 2py)

= bonding orbital (C 2pz + O 2pz)

* = anti-bonding orbital (C 2pz - O 2pz )

c) Quantum Mechanics:

---> * transition is "allowed" and will occur when E is parallel to the x axis n ---> * transition is "symmetry forbidden" ==> transition does not induce a dipole change in the molecule. This transition does occur because of limitations in theory used to predict transitions but with very low probability, <1% of ---> *.

175

2) Nature of the electronic transitions

The electronic transitions are found in organic or inorganic chemistry and can be classified in different groups

1) d-d transitions :

They are found in transition metal ions, when they are complexed by coordinates (Energy of the d orbitals depends on the nature, structure of the complex) and when d level is not full

2) charge transferts

Transitions between molecular orbitals centered on different atoms : electron transfer between metal and ligand

3) Transitions – * and n – *

Electronic transfer between molecular orbitals in organic molecules

176

IV. Examples of UV spectroscopy applications in catalysis

Catalyst Deactivation: in situ UV-vis Spectroscopy

• formation of additional bands during n-butane isomerization• band at 310 nm, allylic cations? Analyzed using apparent absorption A=1-R• spectroscopic and catalytic data can be correlated

177

W. WM. Wendlandt, H.G. Hecht, “Reflectance Spectroscopy”,Interscience Publishers/John Wiley 1966

B.M. Weckhuysen, P. van der Voort, G. Catana, “Spectroscopy of transition metal ions on surfaces”, Leuven University Press 2000

J.P. Sibilia, “A guide to materials characterization and chemical analysis”, Second Edition, VCH Publishers Inc., 1996

References

P. W. Atkins, « Physical Chemistry, Oxford University Press »

The science of spectroscopy : http ://www.scienceofspectroscopy.info.

Spectroscopy Now : http ://www.spectroscopynow.com.