Jackson S.D., Hargreaves J.S.J. Metal Oxide Catalysis

References 93

56 Thompson , T.L. and Yates , T. ( 2006 )

Chemical Reviews , 106 , 4428 .

57 Brus , L. ( 1986 ) Journal of Physical

Chemistry , 90 , 2555 .

58 Alivisatos , A.P. ( 1996 ) Science , 271 ,

933 .

59 Liu , L. and Bard , A. ( 1989 ) Journal of

Physical Chemistry , 93 , 3232 .

60 Fournier , M. , Louis , C. , Che , M. ,

Chaquin , P. and Masure , D. ( 1989 )

Journal of Catalysis , 119 , 400 .

61 Masure , D. , Chaquin , P. , Louis , C. , Che ,

M. and Fournier , M. ( 1989 ) Journal of

Catalysis , 119 , 415 .

62 Barton , D.G. , Shtein , M. , Wilson , R.D. ,

Soled , S.L. and Iglesia , E. ( 1999 ) Journal

of Physical Chemistry B , 103 , 630 .

63 Wachs , I.E. , Kim , T. and Ross , E.I.

( 2006 ) Catalysis Today , 116 , 162 .

64 Tian , H. , Ross , E.I. and Wachs , I.E.

( 2006 ) Journal of Physical Chemistry B ,

110 , 9593 .

65 Morey , M. , Davidson , A. , Eckert , H.

and Stuckey , G. ( 1996 ) Chemistry of

Materials , 8 , 486 .

66 Weckhuysen , B.M. , Wachs , I.E. and

Schoonheydt , R.A. ( 1996 ) Chemical

Reviews , 96 , 486 .

67 Groppo , E. , Lamberti , C. , Bordiga , S. ,

Spoto , G. and Zecchina , A. ( 2005 )

Chemical Reviews , 105

, 115 .

68 Centi , G. , Perathoner , S. , Triﬁ r ò , F. ,

Aboukais , A. , Sissi , C.F. and Guelton ,

M. ( 1992 ) Journal of Physical Chemistry ,

96 , 2617 .

69 Le Noc , N. , Trong On , D. , Solomykina ,

S. , Echchahed , B. , B é land , F. , Cartier ,

C. and Bonneviot , L. ( 1996 ) Studies in

Surface Science and Catalysis , 101 ,

611 .

70 Bordiga , S. , Buzzoni , R. , Geobaldo , F. ,

Lamberti , C. , Giamello , E. , Zecchina ,

A. , Leofanti , G. , Petrini , G. , Tozzola , G.

and Vlaic , G. ( 1996 ) Journal of Catalysis ,

158 , 486 .

71 Tabacchi , G. , Gianotti , E. , Fois , E. ,

Martra , G. , Marchese , L. , Coluccia , S.

and Gamba , A. ( 2007 ) Journal of Physics

C – Solid State Physics , 111 , 4946 .

72 Busca , G. , Martra , G. and Zecchina , A.

( 2000 ) Catalysis Today , 56 , 361 .

73 Boujday , S. , Lambert , J. - F. and Che , M.

( 2003 ) The Journal of Physical Chemistry

B , 107 , 651 .

74 Bonneviot , L. , Legendre , O. , Kermarec ,

M. , Oliver , D. and Che , M. ( 1990 ) Journal

of Colloid and Interface Science , 134 , 534 .

75 Lambert , J. - F. , Hoogland , M. and Che , M.

( 1997 ) The Journal of Physical Chemistry B ,

101 , 10347 .

76 Spielbauer , D. , Zeilinger , H. and

Kn ö zinger , H. ( 1993 ) Langmuir , 9 , 460 .

77 Schoonheydt , R.A. , De Vos , R. , Pegrims ,

J. and Leeman , H. ( 1989 ) Studies in

Surface Science and Catalysis , 49A , 559 .

78 Lakowicz , J.R. ( 1999 ) Principles of

Fluorescence Spectroscopy , 2nd edn, Kluwer

Academic/Plenum Publishers .

79 Anpo , M. and Kubokawa , Y. ( 1980 )

Journal of the Chemical Society – Faraday

Transactions I , 76 , 1014 .

80 Coluccia , S. , Deane , M. and Tench , A.J.

( 1977 ) Proceedings of the 6th

International Congress on Catalysis ,

London 1976 - Ed. The Chem Soc.,

London, 1 , 171 .

81 Coluccia , S. , Deane , M. and Tench , A.J.

( 1978 ) Journal of the Chemical Society –

Faraday Transactions I , 74 , 2913 .

82 Coluccia , S. and Borello , E. ( 1984 )

Proceedings of the 8th International

Congress on Catalysis , Berlin 191984 - Ed

DECHEMA, 3 , 69 .

83 Coluccia S. ( 1985 ) Adsorption and Catalysis

on Oxides Surfaces (eds M. Che and G.C.

Bond ), Elsevier , Amsterdam , p. 59 .

84 Kibblewhite , J.F.J. and Tench , A.J. ( 1974 )

Journal of the Chemical Society – Faraday

Transactions I , 70 , 72 .

85 Anpo , M. , Yamada , Y. , Kubokawa , Y. ,

Coluccia , S. , Zecchina , A. and Che , M.

( 1988 ) Journal of the Chemical Society –

Faraday Transactions I , 84 , 751 .

86 Anpo , M. , Yamada , Y. , Coluccia , S. ,

Zecchina , A. and Che , M. ( 1989 ) Journal

of the Chemical Society – Faraday

Transactions I , 85 , 609 .

87 Chizallet , C. , Costentin , G. , Krafft , J. ,

Lauron - Pernot , H. and Che , M. ( 2006 )

ChemPhysChem , 7 , 904 .

88 Coluccia , S. , Hemidy , J.F. and Tench , A.J.

( 1978 ) Journal of the Chemical Society –

Faraday Transactions I , 74 , 2763 .

89 Matsuoka , M. and Anpo , M. ( 2003 )

Journal of Photochemistry and Photobiology

A: Chemistry , 3 , 225 .

90 Dzwigaj , S. , Matsuoka , M. , Anpo , M. and

Che , M. ( 2001 ) Catalysis Letters , 72 , 211 .

94 2 The Application of UV-Visible-NIR Spectroscopy to Oxides

91 Dzwigaj , S. , Matsuoka , M. , Anpo , M.

and Che , M. ( 2006 ) Microporous and

Mesoporous Materials , 93 , 248 .

92 Shiraishi , Y. , Morishita , M. , Teshima ,

Y. and Hirai , T. ( 2006 ) The Journal of

Physical Chemistry B , 110 , 6587 .

93 Dzwigaj , S. , Matsuoka , M. , Anpo , M.

and Che , M. ( 2000 ) The Journal of

Physical Chemistry B , 104 , 6012 .

94 Anpo , M. , Kondo , M. , Coluccia , S. ,

Louis , C. and Che , M. ( 1989 ) Journal of

the American Chemical Society , 111 ,

8791 .

95 Higashimoto , S. , Tsumura , R. , Zhang ,

S.G. , Matsuoka , M. , Yamashita , H. ,

Louis , C. , Che , M. and Anpo , M. ( 2000 )

Chemistry Letters , 29 , 408 .

96 Anpo , M. , Takahashi , I. and Kubokawa ,

Y. ( 1982 ) Journal of Physical Chemistry ,

86 , 1 .

97 Hazenkamp , M.F. and Blasse , G. ( 1992 )

Journal of Physical Chemistry , 96 , 3442 .

98 Zhang , S.G. , Ichihashi , Y. , Yamashita ,

H. , Tatsumi , T. and Anpo , M. ( 1996 )

Chemistry Letters , 10 , 895 .

99 Lamberti , C. , Bordiga , S. , Arduino , D. ,

Zecchina , A. , Geobaldo , F. , Span ò , G. ,

Genoni , F. , Petrini , G. , Carati , A. ,

Villain , F. and Vlaic , G. ( 1998 ) Journal

of Physical Chemistry B , 102 , 6382 .

100 Marchese , L. , Maschmeyer , T. , Gianotti ,

E. , Coluccia , S. and Thomas , J.M.

( 1997 ) Journal of Physical Chemistry B ,

101 , 8836 .

101 Marchese , L. , Gianotti , E. , Dellarocca ,

V. , Maschmeyer , T. , Rey , F. , Coluccia ,

S. and Thomas , J.M. ( 1999 ) Physical

Chemistry Chemical Physics , 1 , 585 .

102 Dedecek , J. , Sobalik , Z. , Tvaruzkova , Z. ,

Kaucky , D. and Wichterlova , B. ( 1995 )

Journal of Physical Chemistry , 99 , 16327 .

103 Dedecek , J. and Wichterlova , B. ( 1994 )

Journal of Physical Chemistry , 98 , 5721 .

104 Barrie , J.D. , Dunn , B. , Hollingsworth , G.

and Zink , J.I. ( 1989 ) Journal of Physical

Chemistry , 93 , 3958 .

105 Yamashita , H. , Matsuoka , M. , Tsuji , K. ,

Shioya , Y. , Anpo , M. and Che , M. ( 1996 )

Journal of Physical Chemistry , 100 ,

397 .

106 Anpo , M. , Sunamoto , M. and Che , M.

( 1989 ) Journal of Physical Chemistry , 93 ,

1187 .

107 Patterson , H.H. , Cheng , J. , Despres , S. ,

Sunamoto , M. and Anpo , M. ( 1991 )

Journal of Physical Chemistry , 95 , 8813 .

108 Iwamoto , M. , Furukawa , H. , Matsukami ,

K. , Takenaka , T. and Kagawa , S. ( 1983 )

Journal of the American Chemical Society ,

105 , 3719 .

109 Centi , G. ( 1996 ) Applied Catalysis A:

General , 147 , 267 .

110 Blasco , T. and Lopez Nieto , J.M. ( 1997 )

Applied Catalysis A: General , 157 , 117 .

111 Garcia , H. , Lopez Nieto , J.M. , Palomares ,

E. and Solsona , B. ( 2000 ) Catalysis Letters ,

69 , 217 .

112 Gil - Llambias , F.J. , Escudey - Castro , A.M.

and Santos - Blanco , J. ( 1983 ) Journal of

Catalysis , 83 , 225 .

113 Dzwigaj , S. , Matsuoka , M. , Franck , R. ,

Anpo , M. and Che , M. ( 1998 ) Journal of

Physical Chemistry B , 102 , 6309 .

114 Gillespie , R.J. ( 1972 ) Molecular Geometry ,

Van Nostrand - Reinhold , London .

115 Hazenkamp , M.F. , Strijbosch , A.W.P.M.

and Blasse , G. ( 1992 ) Journal of Solid

State Chemistry , 97 , 115 .

116 Marion , M.C. , Garbowski , E. and Primet ,

M. ( 1990 ) Journal of the Chemical

Society – Faraday Transactions , 86 ,

3027 .

The Use of Infrared Spectroscopic Methods in the Field of

Heterogeneous Catalysis by Metal Oxides

Guido Busca

95

3

3.1

Introduction

3.1.1

The Evolution of Vibrational Spectroscopies

The fundamentals of transmission/absorption IR spectroscopy were developed in

the last decades of the nineteenth century and in the very ﬁ rst years of the twenti-

eth century [1] , and, as a consequence, IR spectroscopy rapidly became one of the

most widely used techniques for chemical analysis.

Subsequently, it has been recognized that other IR techniques (reﬂ ection, emis-

sion and photothermal techniques) can also be applied, and that IR spectroscopy

only gives partial information on the vibrational structure of most polyatomic

species. In fact, selection rules apply to IR light absorption phenomena, so that

only vibrational modes that are associated with changes of the molecular dipolar

moment can be directly excited.

Another way to obtain part or all the vibrational spectrum of a chemical species

is to look at the inelastic scattering of beams other than IR radiation. In other

words, when such beams are scattered by a polyatomic species, part of the photons

(or particles) lose (or gain) energy by interaction with the vibrational states of the

chemical species. The vibrational spectrum is obtained by measuring the differ-

ence between the energies of the incident beam and of its small fraction that is

inelastically scattered. If the incident radiation is relatively high energy such as

UV, visible or near - infrared ( NIR ) photons we are dealing with Raman spectros-

copy, based on the so - called Raman effect, recognized by C.V. Raman in 1928 [2] .

Owing to the different selection rules for the Raman effect, the Raman - active

modes do not correspond to IR - active modes in centrosymmetric chemical species,

and (generally) only partly correspond for non - centrosymmetric species. Thus,

Raman spectroscopy is in most cases almost complementary to IR spectroscopy.

Alternatively the entire, or part of the, vibrational spectra of polyatomic chemical

species can be obtained by looking at the inelastic scattering of particles other than

Metal Oxide Catalysis. Edited by S. David Jackson and Justin S. J. Hargreaves

ISBN: 978-3-527-31815-5

96 3 The Use of Infrared Spectroscopic Methods

photons: they can be electrons, and the corresponding techniques are Electron

Energy Loss Spectroscopy ( EELS ) or Inelastic Electron Tunnelling Spectroscopy

( IETS ), following two different experimental setups. The incident particles can also

be neutrons, which gives rise to Inelastic Neutron Scattering Spectroscopy ( INS ),

or atoms, giving rise to Atom Scattering Spectroscopy ( ASS ). These techniques

were developed by physicists, particularly in the 1960s, and have found application

mainly in the ﬁ eld of surface chemistry and physics. A further new technique,

vibrational Sum Frequency Generation Spectroscopy ( SFG ) has been recently

developed and applied to gas – solid and liquid – solid interfaces on metal catalysts

[3] and on oxide surfaces [4] .

3.1.2

Application of IR Spectroscopy to the Surface Chemistry of Oxide - Based Materials:

a Historical Perspective

It seems that the very ﬁ rst application of IR spectroscopy to the surfaces of oxide

materials was reported in 1937 by Buswell and coworkers [5] , concerning water

adsorbed in montmorillonite. In the 1940s and 1950s, Terenin and coworkers at

Leningrad University showed that this technique can be applied widely to mole-

cules adsorbed on solid materials, including metal oxides [6] . In fact most metal

oxides only absorb radiation in the low - energy IR region (below 1000 cm

− 1

) while

most gaseous or volatile molecules also absorb strongly in the higher energy IR

region (4000 – 1000 cm

− 1

). However, adsorption on solids causes vibrational pertur-

bation of the adsorbate (if adsorbed as such) or can produce a chemical reaction,

and the spectroscopic detection and assignment of these phenomena gives infor-

mation on the adsorption or interaction modes. Ultimately, data are obtained both

on the chemical structure of the metal oxide surface and on the perturbation mode

of the adsorbate.

The application of IR spectroscopy to catalysis and surface chemistry was later

developed in the ﬁ fties by Eischens and coworkers at Texaco laboratories (Beacon,

New York) in the USA [7] and, almost simultaneously, by Sheppard and Yates at

Cambridge University in the UK [8] . Mapes and Eischens published the spectra

of ammonia chemisorbed on a silica - alumina cracking catalyst in 1954 [6] , showing

the presence of Lewis acid sites and also the likely presence of Br ø nsted acid sites.

Eischens, Francis and Pliskin published the IR spectra of carbon monoxide

adsorbed on nickel and its oxide in 1956 [9] . Later they presented the results of an

IR study of the catalyzed oxidation of CO on nickel at the First International Con-

gress on Catalysis, held in Philadelphia in 1956 [10] . Eischens and Pliskin also

published a quite extensive review on the subject of “ Infrared spectra of adsorbed

molecules ” in Advances in Catalysis in 1958, where data on hydrocarbons, CO,

ammonia and water adsorbed on metals, oxides and minerals were reviewed [11] .

These papers evidence clearly the two tendencies observed in subsequent spectro-

scopic research in the ﬁ eld of catalysis. They are the use of probes to test the

surface chemistry of solids and the use of spectroscopy to reveal the mechanism

of the surface reactions. They used an in situ cell where the catalyst sample was

deposited on a ﬂ uorite disk, with ﬂ uorite windows and a tungsten - wound quartz

furnace, connected to a gas - manipulation apparatus.

The industrial interest in IR spectroscopy in catalysis is demonstrated by the

work of Peri and coworkers at Amoco. Peri ﬁ rst published IR studies on the

surface properties of alumina in 1960 [12] and performed several studies on

alumina, silica and silica - alumina. Peri also published a useful summary of the

use of IR spectroscopy in catalysis research in 1984 [13] .

In the 1960s and 1970s several different groups investigated the IR spectra of

adsorbed probe molecules for the characterization of the surface of metals and

oxide catalysts. However, relatively few studies used IR spectra for the study of

reaction mechanisms. Among the best results, Kokes and coworkers used IR

spectroscopy to investigate the conversion and hydrogenation of oleﬁ ns on zinc

oxide. In particular these authors identiﬁ ed an anti π - allyl species as the intermedi-

ate in the double - bond isomerization of 1 - butene to 2 - butene [14, 15] . In the 1980s,

IR techniques underwent a great improvement with the application of Fourier

transform instruments [16] , ﬁ rst commercialized at the end of the 1970s.

Since the end of the 1970s, IR cells have been developed for in situ experiments

in ﬂ ow conditions, with the simultaneous detection of the reaction products and

measures of conversion and selectivities [17] . The concept of experiments under

dynamic conditions , developed by Tamaru [18] , has also been very interesting,

although it has not found widespread application. More recently this approach has

evolved in the concept of Operando Spectroscopy [19, 20], with the possibility of

applying several spectroscopic techniques simultaneously to the same catalyst

sample in working conditions being realized [21] . Among other recent develop-

ments, the coupling of IR spectroscopy with different theoretical calculations, the

use of IR spectroscopic techniques at liquid helium temperature, or at very low

temperature [22] , and the development of the reﬂ ection technique on oxide layers

generated over metal monocrystals [23] have attracted particular interest.

3.2

Experimental Techniques

3.2.1

The Detection of the Vibrational Spectrum of a Polyatomic Chemical Species:

IR and Raman Spectroscopies

Any chemical species containing more than one atom, namely polyatomic mole-

cules, molecular ions, molecular radicals and any crystal, can be described in terms

of atomic positions, bond lengths and bond angles. However, it was recognized

long ago that, in fact, any atom vibrates around its equilibrium position in any

condition, giving rise to more or less complex vibrational modes. At absolute zero

temperature, all polyatomic chemical species lie at the vibrational ground state.

However, they can undergo vibrational excitation and so occupy higher energy

vibrational states. The energy difference between nearest vibrational states lie in

3.2 Experimental Techniques 97

98 3 The Use of Infrared Spectroscopic Methods

the medium and far IR ranges, generally deﬁ ned by the wavenumber limits

4000 – 400 cm

− 1

(0.496 > E > 0.0496 eV, medium IR) and 400 – 10 cm

− 1

(0.0496 > E

> 1.24 10

− 3

eV, far - infrared , FIR ).

The simplest way to cause vibrational excitation is to allow the chemical species

to absorb an energy quantum from an electromagnetic radiation of an appropriate

energy, so giving rise to a vibrational transition from the ground state to an excited

vibrational state. When electromagnetic radiation with intensity I

0

is incident on

a sample, the light may (essentially) be absorbed, reﬂ ected or transmitted. From

the conservation of energy it follows that

IIII

0

=++

RTA

(3.1)

where I

0

, I

R

, I

T

and I

A

are the intensities of the incident, reﬂ ected, transmitted and

absorbed radiation, respectively. By dividing Equation 3.1 by I

0

, we obtain:

RT++=α 1

(3.2)

where R = I

R

/ I

0

, T = I

T

/ I

0

and α = I

A

/ I

0

are the apparent reﬂ ectance, transmittance

and absorptance, respectively, while

AII=−log

10 0T

(3.3)

is the absorbance. All these units depend, obviously, on the particular radiation

used, that is, its energy, E , its frequency, ν , its wavelength, λ , or its wavenumber,

ν :

E h hc hc== =νλν

(3.4)

where h is the Planck constant and c is the velocity of light in a vacuum.

The analysis of the quanta that are actually absorbed by a polyatomic chemical

species and those that are not absorbed (so are transmitted) gives information on

the vibrational structure of these species and, consequently, on its chemical and

geometric structure. This is the transmission/absorption IR technique. However,

selection rules apply to such a phenomenon. They are simpliﬁ ed as follows:

∆ν = ±1

(3.5)

∂

⎛

⎝

⎜

⎞

⎠

⎟

≠

µ

Q

0

(3.6)

On the other hand, the absorbance at a particular wavelength also depends upon

the population of the corresponding vibrational states. According to Boltzmann ’ s

law:

NN g g

ii

EkT

00

=

−

e

∆

(3.7)

where N

i

is the state population, g

i

is the state multiplicity, ∆ E is the energy dif-

ference between the states 0 and i, T the absolute temperature and k the Boltzmann

constant. ∆ E values for fundamental vibrational transitions fall in the IR region.

The population of the ﬁ rst excited state is very small at room temperature and

much lower than the population of the ground state even at 1000 K. This means

that only transitions originating from the ground state can be ordinarily excited at

ambient, or relatively low, temperature.

Equation 3.5 , where ν is the vibrational quantum number, means that only

transitions between nearest vibrational states can directly occur in the case of the

harmonic oscillator. This means that the IR spectrum is generally mostly consti-

tuted by fundamental transitions , that is, those associated with excitation from the

fundamental state to the ﬁ rst excited state. This condition, however, is relaxed in

the case of anharmonic oscillators, so that not only fundamental transitions but

also overtone and combination modes (also called the harmonics , i.e. modes associ-

ated with the excitation from the fundamental state to a second or third excited

state ) can be sometimes observed, although they are usually weak.

Equation 3.6 indicates that only vibrational modes that are associated with a

change in the dipole moment µ of the molecule between the extremes of the

atomic displacements ( Q is the normal coordinate) different from zero (i.e. the

“ polar ” modes) can be directly excited.

In the solid state, the polar phonons (those that are IR active) split into two

components, the transverse optical mode ( TO ) and the longitudinal optical

mode ( LO ). This TO/LO splitting occurs because the electric ﬁ eld associated

with the transverse wave = 0 while that associated with the longitudinal wave is

≠ 0. The coupling of these modes with the electric ﬁ elds associated with the vibra-

tion gives rise to ν

LO

> ν

TO

. This factor is relevant in relation to the shape and

interpretation of the IR spectra of solid materials and will be further considered

below.

According to the Beer – Lambert law:

Acl=ε

(3.8)

the absorbance linearly depends on the molar concentration of the absorbing

species c , through the molar absorption coefﬁ cient ε and the sample pathlength l .

This relation (rigorously valid only in non - scattering media) is the basis for quan-

titative analysis performed through IR spectroscopy. The range of the Beer – Lambert

law in FTIR spectrometry may be extended [24] .

As is well known, the selection rules for Raman spectroscopy may be simpliﬁ ed

as follows:

∆ν = ±1

(3.9)

∂

⎛

⎝

⎜

⎞

⎠

⎟

≠

α

Q

0

(3.10)

3.2 Experimental Techniques 99

100 3 The Use of Infrared Spectroscopic Methods

which means that in Raman spectroscopy also, only transitions between nearest

levels and associated with changes in the polarizability α upon motion are allowed

for a harmonic oscillator. The ﬁ rst selection rule is relaxed for anharmonic oscil-

lators. Group theory shows that for centrosymmetric chemical species, Raman -

active modes are IR inactive, and vice versa. This is the so - called mutual exclusion

rule . Other modes can be both IR and Raman inactive. For non - centrosymmetric

molecular species, modes also occur that are both IR and Raman active. Thus,

Raman spectroscopy is in most cases almost complementary to infrared

spectroscopy.

Some chemical species are vibrationally silent , e.g. monatomic noble gases or IR

inactive , e.g. homomolecular diatomic molecules (H

2

, N

2

, O

2

, F

2

, Cl

2

, etc.)

3.2.2

The Transmission/Absorption IR Technique

This technique was originally applied by the pioneers of IR spectroscopy and is

still perhaps the most widely used today. However, the application of this tech-

nique to solids involves problems of sample form and preparation. This technique

can be successfully applied to powdered insulating or semiconducting solids,

although problems arise from radiation scattering.

When radiation is incident on a non - absorbing layer consisting of N particles

per unit volume, its intensity decreases exponentially with the sample thickness

x following the law:

II I

xNx

T

ee==

−−

00

ϕσ

(3.11)

so that an apparent absorbance, A ′ , can be deﬁ ned for an absorbing and scattering

medium:

′

=− = =

′

+

()

=+ =+

′

AIINxNsxANsxASlog . . .

10 0

043 043 043

T

σε

(3.12)

where ϕ is the linear attenuation coefﬁ cient and σ is the linear attenuation coefﬁ -

cient of one particle. σ contains a component due to absorption, ε ′ , and a compo-

nent due to scattering, s . So the apparent absorbance is composed of the true

absorbance A superimposed upon the component due to the light scattering S ′ .

This simpliﬁ ed approach is similar to that given by Henry [25] who proposed the

following expression for the component dependent on scattering:

′

=−

()

Skxdn1

2

λ

(3.13)

where k is a constant independent of λ but dependent on the particle arrangement,

d is the particle diameter and n is the refractive index of the material. On the other

hand, according to the Rayleigh theory, S ′ depends on the third power of the par-

ticle diameter (and so linearly on the particle volume) and on the fourth power of

the radiation frequency, ν . The effect of scattering on the IR transmission spectra

of powder samples has been studied by Duykaerts [26] who showed that if the

particle diameter is sufﬁ ciently small (considerably smaller than the IR wave-

length), the scattering of the IR radiation is very small. Thus, ﬁ nally, the apparent

absorbance of a layer of a powder depends on its own absorptivity, on the disk

thickness and density, on the particle size and on the wavelength.

In practice, to obtain a good spectrum of the fundamental vibrations of an

insulating material with the transmission/absorption IR technique, it is necessary

to prepare a layer appropriately diluted and sufﬁ ciently thin. To do this, the most

widely practiced technique involves KBr pressed - discs. KBr is a readily available

powdered material which does not absorb in the medium IR region (down to near

400 cm

− 1

, but cuts out the FIR). It can be easily mixed homogeneously with the

powder to be investigated, and pressed, thus obtaining “ diluted ” self - supporting

disks which are very useful for IR transmission. The spectra so obtained give rise

to a baseline due to scattering, that gives very high transmission in the low -

frequency region (down to the cut - off near 400 cm

− 1

due to absorption by KBr

itself), but generally still sufﬁ cient transmission in the higher frequency medium

IR region, owing to the wavelength dependence of scattering. Other materials (for

example CsI) can be used to produce similar pressed discs, with cut - off limits at

even lower frequencies. Alternatively, the powders can be simply deposited in the

form of a thin layer on a disk of a transmitting medium.

Because KBr and the other alkali halides absorb in the FIR region, in this region

other binding materials must be used. The preparation of disks with polyethylene

(which is transparent in most of the FIR region) is the most commonly applied

technique. However, in this range the scattering is very low because of the long

wavelength involved, so that it is sometimes preferable to simply deposit the

powder on a disk of a transmitting material using a solvent that later evaporates.

Another technique, used mostly in the past, is to mull the powder with liquids,

such as liquid parafﬁ n (nujol) and perhalogenated organic liquids. The powder is

mixed to make a paste from which a layer is formed. Obviously, absorptions from

the mulling liquid (e.g. in the C

–

H stretching and deformation regions for parafﬁ n

oils, or C

–

X modes for halocarbons) also occur and obscure some spectral regions.

3.2.3

The Reﬂ ection Techniques

When incident radiation interacts (from the vacuum) with a ﬂ at surface, the

“ reﬂ ectance ” of the parallel and perpendicular polarized light (with respect to the

incidence plane), R

||

and R

⊥

, are given by the relations below:

R



=

−−

+−

εθε θ

cos sin

2

(3.14)

R

⊥

=

−−

+−

cos sin

θε θ

2

(3.15)

3.2 Experimental Techniques 101

102 3 The Use of Infrared Spectroscopic Methods

where ε is the complex dielectric constant of the material (assumed to be isotropic)

and θ is the incidence angle. From these relations it follows that the reﬂ ected light

is mainly polarized perpendicular (in fact R

⊥

is large while R

||

has a minimum for

0 < θ < 90). When θ = 0, converting ε = ( n − i k )

2

, we obtain:

R

nk

=

−

()

+

()

+

1

2

(3.16)

where n and k are the real and imaginary parts of the absolute index of refraction.

For conducting materials, such as metals, the refraction index is totally imaginary,

that means that n = 0 so that R = 1 at any frequency, and the radiation is totally

reﬂ ected. This is a case of “ total ” or “ specular ” reﬂ ection.

For insulating or weakly semiconducting materials, like most inorganic com-

pounds, k = 0 (i.e. the index of refraction is real) and n varies little in the IR spec-

trum except between the absorption frequencies, that is, for ν > ν

LO

and ν < ν

TO

.

This means that outside of the latter skeletal region, monocrystal surfaces of

inorganic compounds only weakly reﬂ ect the radiation, and the smaller the value

of R the greater the value of n . So all light is essentially transmitted, or, better,

refracted. However, for ν

LO

> ν > ν

TO

, the index of refraction is imaginary and k >

0 goes through a maximum. This means that near the skeletal absorption region,

R again becomes ≅ 1, that is, the skeletal vibrations of inorganic compounds give

rise not only to absorbed radiation in transmission experiments but also to reﬂ ected

radiation in the reﬂ ection experiments (reststrahlen effect). Thus, the specular

reﬂ ectance for insulating materials, both in the form of monocrystals and in the

form of sintered pellets, is frequently the basis for the best determination of the

skeletal spectrum, as far as the IR active modes are concerned.

In Figure 3.1 , the reﬂ ectance spectrum of a MgO monocrystal is reported and

is compared with the transmission/ absorption spectrum of a MgO powder. MgO

(periclase) has the rock - salt structure: as the other rock - salt type solids, MgO is

Raman - silent (as far as the fundamentals are concerned, see below) and only

shows one IR - active fundamental mode. The reﬂ ection by the MgO monocrystal,

is strong between two limits. The upper limit is the so - called LO while the lower

limit is the TO.

The TO quoted for MgO at 401 cm

− 1

(see Figure 3.1 a) corresponds to the lower

frequency limit of the reﬂ ectance band while the LO corresponds to the higher

energy limit of the reﬂ ectance, and is quoted at 718 cm

− 1

for MgO (see Figure 3.1 a).

For ν

LO

> ν > ν

TO

the reﬂ ectance is strong and almost constant, while out of these

limits the reﬂ ectance is very small. The transmission/absorption IR spectrum of

the MgO powder is inﬂ uenced by the TO/LO splitting. In fact it shows the

maximum slightly above ν

TO

(407 cm

− 1

in our spectrum of a polyethylene disk) and

a shoulder near ν

LO

.

Total reﬂ ection of IR incident light at any wavelength from a surface can occur

in two cases: (i) if the beam arises from a transmitting medium and is incident

on the surface of a conducting material; (ii) if the beam arises and is incident

within a transmitting medium, but the medium from which the beam arises has

Jackson S.D., Hargreaves J.S.J. Metal Oxide Catalysis

Подождите немного. Документ загружается.