fritz-haber-institut der max-planck-gesellschaft department...
TRANSCRIPT
MAX-PLANCK-GESELLSCHAFT
ACFHI
From Surfaces to Three Dimensions:Experimental Methods in Catalysis
Friederike C. Jentoft
Department of Inorganic ChemistryFritz-Haber-Institut der Max-Planck-Gesellschaft
Faradayweg 4-6, 14195 Berlin
IMPRS “Complex Surfaces in Materials Science”
Block Course WS 05/06
October 13, 2005
Outline Part I
1. IntroductionComplexity of Heterogeneous Catalysts
2. Role of Surface and Bulk Structure
3. Active Sites
4. Physisorption, Porosity
5. ChemisorptionIsothermsIR SpectroscopyThermal Desorption SpectroscopyAdsorption Calorimetry
Questions:Type? Number / density?„Strength“ (interaction with a certain molecule)?
Examples for Surface Sites
Metal sites
Brønsted-sites
Lewis-sites
Oxide Surfaces
v can be terminated by O or Mx+ or mixture (may depend on environment)
v can be terminated by foreign ions
v most frequently surfaces feature OH groups:
HO
Zr
HO
Zr….Zr
HO
Zr..Zr..Zr
type I type II type III
v zeolite OH groups inside the poresHO
Si Al
OH groups can beBrønsted acids = proton donorsBrønsted bases = proton acceptors
Detection of OH-Groups
v IR spectroscopy, NMR spectroscopy
v band position or chemical shift delivers information
4000 3500 3000 2500 2000 1500 1000
0
10
20
30
40
50
161216
32
1625
2119
2040
1379
1257
2118
2036
1934
H4PVMo
11O
40Cs
2H
2PVMo
11O
40Cs
3HPVMo
11O
40Cs
4PVMo
11O
40
Tran
smitt
ance
(%)
Wavenumbers / cm-1
CaF2window
absorption
Lewis-sites coordinatively unsaturated (cus) metal cations (acidic, electron pair acceptors)oxygen anions (basic, electron pair donors)
Lewis Sites
v note: on a surface, acidic and basic sites (Brønsted and/or Lewis) can coexist (would neutralize each other in solution)
v only detectable in an adsorption experiment
OH O
H NHH
HIR, XPS, NMR, UV/Vis...shift of bands(probe/surface)desorption T
TDS/TPD
adsorption ∆ Hcalorimetry
+ NH3
Probing Sites by Chemisorption
Different methods deliver different informationv isotherms: number of sitesv spectroscopies: type, strength (but not number unless extinction
coefficient known)v calorimetry, TDS: number and strength (but not type)
Metal Surface Sites
v specific adsorption of probe on only one type of site (e.g. on metal and not on support): depends on strength of interaction of probe with metal/support sites, conditions (T, p) can be optimized
Metal particles
Support (e.g. oxide) surface
CO molecules
Measuring Metal Surface Area with H2 or CO
v if adsorption is specific, number of sites can be derived from isotherm(no site heterogeneity, no spill-over)
IR Spectra of CO Adsorption on SAPO at 77K
10-3 < p < 10-2 torr
Onida et al. J. Phys. Chem. B 45 (1997) 9244-9249
Hydroxyl vibrations Carbonyl vibrations
OH
OH C O
v different OH groups will experience different shifts, shift is a measure of the acid strength of the OH group
Adsorption of CO on Pt/Mordenite
2300 2250 2200 2150 2100 2050 2000 19500.000
0.001
0.002
0.003
2131
p(CO): ca. 0.40 mbar dj6norm 250°C unter Vakuumdl7norm 450°C unter Vakuumdl71norm 375°C unter H2
2195
2171
2110
2088
2050
2228
Abs
orba
nce
(nor
mal
ized
)
Wavenumber / cm-1
v multiple sites!
v need reference data (zeolite without Pt, Pt single crystal data)
NH3 TPD
100 200 300 400 500 600 700 800
sulfated ZrO2M
S-In
tens
ity [a
.u.]
T [°C]
v NH3 is a strong base and a probe for acidic sites (Brønsted and Lewis)
v presence of sulfate on the zirconia surface generates strongly acidic sites
NH H
H
CO2 TPD
v CO2 is a probe for basic sites (oxygen anions)
v ZrO2 is a basic oxide, but the basic sites can be completely coveredwith sulfate
100 200 300 400 500 600 700
sulfated ZrO2
MS-
Inte
nsity
[a.u
.]
T [°C]
Integral Heat of Adsorption
v A probe may chemisorb on different sites under production of different heats of adsorption
v If all sites are covered at once, the evolved heat will be an integral heatv if the number of adsorbed molecules is known, an average / mean heat
of adsorption can be calculated
Differential Heats of Adsorption
v A general concept: example dissolution“first”, “last” heat of dissolution = differential heats
v Sites can be covered step by step, e.g. 1.2. 3.
v Differential heats of adsorption as a function of coverage can be determined
ATdiff nQq ,int )/( δδ=
Adsorption Calorimetry
v the sorptive must be introduced stepwise, i.e. at constant temperature, the pressure is increased slowly
v for each adsorption step, the adsorbed amount must be determined(isotherm)
v for each adsorption step, the evolved heat must be determined
v the differential heat can then be determined by division of evolved heat through number of molecules adsorbed in a particular step
20 21 22 23 24 25 26 27 28 29 300.2
0.4
0.6
0.8
1.0
1.2
1.4
0.0
0.5
1.0
1.5
2.0
2.5
3.0Th
erm
osig
nal /
V
Time / h
Equi
libriu
m p
ress
ure
/ hPa
vgeneration of adsorption isotherm
vdifferential heats of adsorption
Adsorption Calorimetry Raw Data: Equilibrium Pressure and Thermosignal
0 1 2 3 4 5 6 70.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Act. 473 KAct. 573 KAct. 723 KAct. 723 KA
dsor
bed
amou
nt /
mm
ol g-1
Isobutane partial pressure / hPa
Surface Sites: Effect of Activation
v with increasing activation temperature, the differential heats of adsorption increase
Isotherms313 K
0.00 0.01 0.02 0.03 0.04 0.050
20
40
60
80
100Act. 473 KAct. 573 KAct. 723 KAct. 723 K
Diff
. hea
t of a
dsor
ptio
n / k
J m
ol-1
Adsorbed amount / mmol g-1
Ammonia Adsorption on Heteropolyacids
v H3PW12O40 * x H2O reaction with ammonia
Diffuse Reflectance IR and UV-vis Spectroscopy
Outline Part II
1. Interaction of Light with a Sample2. Transmission & Lambert-Beer Law3. Scattering and Reflection4. Schuster-Kubelka-Munk Theory5. Experimental & Examples
White StandardsIntegrating SpheresMirror OpticsFiber Optics
UV- vis- NIR – MIR
v electronic transitions, vibrations (rotations)
v diffuse reflectance UV-vis spectroscopy (DR-UV-vis spectroscopy or DRS)
v diffuse reflectance Fourier-transform infrared spectroscopy (DRIFTS)
Type of transition
Spectral range
Molecular rotation Electronic excitation
X-ray radiation
InfraredRadio waves Micro waves
F M Nvisvis UV
6.2 to 1.5Energy /eV
200 bis 800(700-1000) bis 30003000 bis(25000-40000)
Wavelength / nm
50000 bis 1250012500 bis 33003300 bis 250Wavenumber/ cm-1
UV-visNear IR (NIR)Mid IR (MIR)
Molecular vibrations
Spectroscopy with Powder Samples
v how to measure spectra of catalyst (+ adsorbate)?v absorption as a function of wavelength, qualitatively and
quantitatively
ZrO2
Interaction of Light with Sample
v how to extract absorption properties from transmitted light?v how to deal with reflection and scattering?
incident light
reflection at phase boundaries
transmitted light
scattered light
absorption of light
(luminescence)
How to Deal with Reflection (1)
v fraction of reflected light can be eliminated through reference measurement with same materials (cuvette+ solvent)
incident light
reflection at phase boundaries
transmitted light
How to Deal with Reflection (2)
v fraction of reflected light can be eliminated through variation of sample thickness: number of phase boundaries remains the same
incident light
reflection at phase boundaries
transmitted light
Transmittance
if we manage to eliminate reflection:1 = α + τin this case, the absorption properties of the sample can be calculated from the transmitted light
I0 I
0II
=τ
Sample
if luminescence and scattering are negligible:1 = α + τ + ρα absorptance, absorption factor, Absorptionsgradτ(T) transmittance, transmission factor, Transmissionsgradρ reflectance, reflection factor, Reflexionsgrad
Transmitted Light and Sample Absorption Properties
separation of variables and integration
sample thickness: l ∫∫ −=lI
I
dlcI
dI
00
κ
decrease of I in an infinitesimally thin layer
lcII
κ−=0
ln
dlcIdI κ−=c: molar concentration of absorbing species [mol/m-3]κ: the molar napierian extinction coefficient [m2/mol]
I0 I
0II
=τ
Transmitted Light and Sample Absorption Properties
ατ κ −=== − 10
lceII
)1ln( ακ −−=== lcBAe
)1log(10 αε −−== lcA
extinction E (means absorbed + scattered light)absorbance A (A10 or Ae)
optical density O.D.
all these quantities are DIMENSIONLESS !!!!
Lambert-Beer Law
standard spectroscopy software uses A10!
napierian absorbanceNapier-Absorbanz
(decadic) absorbancedekadische Absorbanz
Scattering
v scattering is negligible in molecular disperse media (solutions)
v scattering is considerable for colloids and solids when the wavelength is in the order of magnitude of the particle size
Wavenumber WavelengthMid-IR (MIR) 3300 to 250 cm-1 3 to (25-40) µmNear-IR (NIR) 12500 to 3300 cm-1 (700-1000) to 3000 nmUV-vis 50000 to 12500 cm-1 200 to 800 nm
v scattering is reduced through embedding of the particles in media with similar refractive index: KBr wafer (clear!) technique, immersion in Nujol
1100 1000 900 800 7000
20
40
60
Wavenumbers (cm-1)
1083
10731059
983960
896862 788
(NaCl)(KBr)(CsCl)
Tran
smis
sion
(%)
But…….Reaction with Material Used for Embedding
H4PVMo11O40
v reaction with diluent possiblev dilution usually not suitable for experiments at high T/ with reactive gases
Limitations of Transmission Spectroscopy
v transmission can become very low! even in IR regime!
v catalysts are fine powders (high surface area)
v embedding (KBr) not desirable for in situ studies (heating, reaction)
5000 4000 3000 2000 10000.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07sulfated ZrO2 after activiton at 723 K
Tran
smitt
ance
Wavenumber / cm-1
Can We Use the Reflected Light?
v instead of measuring the transmitted light, we could measure thereflected light
v can we extract the absorption properties of our sample from the reflected light?
Detector
Specular Reflection (Non-Absorbing Media)
v fraction of reflected light increases with αv β depends on α and ratio of the refractive indices (Snell law)v insignificant for non-absorbing media, for air/glass about 4%
incident beam reflected beam
refracted beam
n0
n1
n1>n0
α
β
surface SIDE VIEW
Specular Reflection: Angular Distribution
v the intensity of the specularly reflected light is largest at an azimuth of 180°
incident beam reflected beam
surfaceTOP VIEW
azimuth180°
azimuth<180°
Light scatteringdeflection of electromagnetic or corpuscular
radiation from its original direction
Raman-scattering
Compton-scattering
Brillouin-scattering
elastic
Scattering
Rayleigh-scatteringλ > d
wavelength dependent: ∼1/ λ4
no preferred direction
Mie-scatteringλ ≤ d
wavelength independentpreferentially in forward (and
backward) direction
inelastic
Rayleigh- and Mie-Scattering
Quelle: RÖMPP on line
d >> λ: Mie-Theory approaches laws of geometric optics
h·υ
d < λ: Rayleigh-Scatteringisotropic distribution
d = λ: Mie-Scatteringin forward and backward directions
d > λ: Mie-Scatteringpredominantly in forward direction
Diffuse Reflection
v intensity of diffusely reflected light independent of angle of incidence (isotropic)
v Due to multiple reflection, refraction, and diffraction inside the sample
Reflection of Radiation
mirror-type (polished) surfaces mat (dull, scattering)
surfaces
mirror-type reflectionmirror reflectionsurface reflection
equally intense in all directions (isotropic)
images: Wolfgang Böhme Sphere Optics Hoffman GmbH
Typical Catalyst Particles
v need theory that treats light transfer in an absorbing and scattering medium
v want to extract absorption properties!
ZrO2
ca. 20 µm
scanning electron microscopy image
The Schuster-Kubelka-Munk Theory
Assumptions
v incident light diffuse
v isotropic distribution (Lambert cosine law valid) , i.e. no regular reflection
v particles randomly distributed
v particles much smaller than thickness of layer
A Simplified Derivation of the SKM Function
I
J
IJR =
dlSJdlSIdlKIdI +−−=
dlSIdlSJdlKJdJ +−−=with K, S: absorption and scattering coefficient [cm-1]
Divide equations by I or J, respectively, separate variables, introduce R=J/I
Integrate via expansion into partial fractions
∫ ∫∞
=++−
R
R
ldlS
SKRR
dR
0 02 1)1(2
2
A Simplified Derivation of the SKM Function
SK
RRRF =
−=
∞
∞∞ 2
)1()(2
2 constants are needed to describe the reflectance:absorption coefficient Kscattering coefficient S
Kubelka-Munk functionremission function
assume black background, so that R0 = 0make sample infinitely thick, i.e. no transmitted light (typical sample thickness in experiment ca. 3 mm)
)2( SKKSKSR
+++=∞
for K→0 (no absorption) R∞→1, i.e. all light reflectedfor S→0 (no scattering) R∞→0, i.e. all light transmitted or absorbed
Kubelka-Munk Function
v mostly we measure R’∞, i.e. not the absolute reflectance R∞but the reflectance relative to a standard
v depends on wavelength F(R’∞)λ
v corresponds to extinction in transmission spectroscopyv in case of a dilute species is proportional to the concentration of the
species (similar to the Lambert-Beer law):
scRF ε
∝′∞ )(
Spectroscopy in Transmission
reference „nothing“= void, empty cell, cuvette with solvent
Catalyst
LightSource Detector
spectrum: transmission of catalyst vs. transmission of reference
v in IR transmission is widely applied for solids v in UV-vis transmission is rarely used for solids
Reflectance Spectroscopy
reference„white standard“
Catalyst
LightSource
Detector
spectrum: reflectance of sample (catalyst) vs. reflectance of standard
v need element that collects diffusely reflected lightv need to avoid specularly reflected lightv need reference standard (white standard)
White Standards
Spectralon® thermoplastic resin, excellent reflectance in UV-vis region
v KBr: IR (43500-400 cm-1)v BaSO4: UV-visv MgO: UV-visv Spectralon: UV-vis-NIR
The Collection Elements
v integrating spheres (UV-vis-NIR)
v fiber optics (UV-vis-NIR)
v mirror optics (IR and UV-vis-NIR)
Diffuse Reflection & Macroscopic Surface Properties
v randomly oriented crystals in a powder: light diffusely reflected
v flattening of the surface or pressing of a pellet can cause orientation of the crystals, which are “elementary mirrors”
v causes “glossy peaks” if angle of observation corresponds to angle of incidence
v solution: roughen surface with (sand)paper or press between rough paper, or use different observation angle!
Integrating Sphere
Image: Römpp on-line
Sample
Source
Detector
Gloss trap
Screen
v the larger the sphere the smaller errors from the ports
v the larger the sphere the lower the intensity onto the detector
v typically 60-150 mm diameter
v coatings: BaSO4, Spectralon
400 500 600 700 8000.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5K
ubel
ka-M
unk
units
Wavelength / nm
Diffuse Reflectance Spectrum: Green Sheet of Paper
How to Avoid Specular Reflection
v specular reflection is strongest in forward direction
v collect light in off-axis configuration
sample surface
SIDE VIEW TOP VIEW
sample surface
90°
120°
Praying Mantis Optical Accessory (Harrick)
v first ellipsoidal mirror focuses beam on sample
v second ellipsoidal mirror collects reflected light
v 20% of the diffusely reflected light is collected
vcan be placed into the normal sample chamber (in line with beam), no rearrangement necessary
Source
Detector
Diffuse Reflectance IR (DRIFTS): Graseby Specac Selector Optics and Environmental Chamber
v ZnSe window (chemically resistant)
v up to 773 K
v no flow through be (rest in cup)
v large dead volume (est. 100 ml)
Comparison Transmission - Diffuse Reflectance
5000 4000 3000 2000 10000.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.0
0.2
0.4
0.6
0.8
1.0
1.2Tr
ansm
ittan
ce
Wavenumber / cm -1
Ref
lect
ance
sulfated ZrO2 after activation at 723 K Transmission: self-supporting waferReflectance: powder bed
v spectra can have completely different appearancev transmission decreases, reflectance increases with increasing wavenumber
Comparison Transmission - Diffuse Reflectance
v vibrations of surface species may be more evident in DR spectra
5000 4000 3000 20003.03.23.43.63.84.04.24.44.64.85.0
0.0
0.2
0.4
0.6
0.8
1.0
Abs
orba
nce
A Nap
ier
Wavenumber / cm -1
Kub
elka
-Mun
k Fu
nctio
n
sulfated ZrO2 after activation at 723 K
Transmission: self-supporting waferReflectance: powder bed
OH vibrations
SO vibrations
Fiber Optics
v light conducted through total reflectancev fiber bundle with 6 around 1 configuration: illumination through 6 (45°), signal
through 1v avoids collection of specularly reflected light
Bilder: Hellma (http://www.hellma-worldwide.de) and CICP (http://www.cicp.com/home.html)
Literature
v W. WM. Wendlandt, H.G. Hecht, “Reflectance Spectroscopy”, Interscience Publishers/John Wiley 1966, FHI: 22 E 13
v G. Kortüm, “Reflectance Spectroscopy” / “Reflexionsspektroskopie” Springer 1969, FHI: missing