Temperature-programmed methods 1 1

Temperature-programmed
Methods in Catalysis Research
Procedures and pitfalls of a very
commonly used tool

Temperature-programmed methods 2 2
 Definitions
• TPD: temperature-programmed desorption
• TPR: temperature-programmed reduction or
– temperature programmed reaction
• TPO: temperature-programmed oxidation
• TPRS: temperature-programmed reaction
spectroscopy

Temperature-programmed methods 3 3
 Methods
• TPD: a solid is first exposed to an adsorbate gas
under well-defined conditions (wide range of
pressure and temperature) and then heated under
inert conditions with a temperature program:
– parental method: Flash desorption from metal wires in
UHV : G. Ehrlich, Adv. Catal., 1963
• All other methods keep solid and reactants in
contact during temperatre-programmed processing
• TPD – TPRS most suitable designations
Temperature-programmed methods 4 4
 Atmospheres, Reactors and
Detectors

Temperature-programmed methods 5 5
 Atmospheres, Reactors and
Detectors

Temperature-programmed methods 6 6
 Atmospheres, Reactors and
Detectors

Temperature-programmed methods 7 7
 Experiment

High-pressure variant
with inert carrier G1 and
reactant G2

R.N. Rogers,
Anal. Chem., 32, 672, (1960)

Temperature-programmed methods 8 8
FHI-AC variable pressure Filament
TDS-reactor set-up
Metal clips
z-
transfer
rot

Line of sight mass

spectrometer nozzle Electric contacts

Temperature-programmed methods 9 9
 Influence of Vacuum Treatment

Sample K47

4∗10-9 m/e=18 m/e=32

m/e=18 m/e=32

MS-Response (I. C.)

MS-Response (I. C.)

3∗10-9 1∗10-10

“as is” “as is”

2∗10-9
5∗10-11

1∗10-9

after 12h of vacuum

0 0
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Temperature / °C Temperature / °C

Strong
Strong influence
influence of
of vacuum
vacuum treatment
treatment both
both for
for water
water and
and oxygen
oxygen
desorption
desorption traces
traces Solid
Solid state
state dynamics!
dynamics!

Temperature-programmed methods 10 10
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Assignment of relevant MS traces

Reaction study of sample “Partie 5010” (hemihydrate) after TDS treatment:

Molecule m/z fragments

Butane 43, 27, 42, 41, 39

Butene 41, 39, 55, 27, 26

Butadiene 39, 54, 27

2,5-dihydro-Furan 41, 70, 39, 42, 27

Furan 68, 39

MSA 26, 54, 98

Temperature-programmed methods 11 11
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Aqueous vs Alcoholic preparation

Hemihydrate:

m/e=18
m/e=18
MS-Response (norm. u.)

alcoholic
3∗10 -9

2∗10 -9
aqueous

1∗10 -9

0
0 100 200 300 400 500 600
Temperature /°C
Temperature-programmed methods 12 12
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Aqueous vs Alcoholic preparation

Hemihydrate:

1∗10 -10 alcoholic m/e=32

m/e=32
MS-Response (norm. u.)

8∗10 -11

6∗10 -11
aqueous

4∗10 -11

2∗10 -11

0 100 200 300 400 500 600

Temperature /°C
Temperature-programmed methods 13 13
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Aqueous vs Alcoholic preparation

Hemihydrate:

2.5∗10 -10
m/e=44
m/e=44
MS-Response (norm. u.)

2∗10 -10

1.5∗10 -10 alcoholic

(“Partie 5010”)
1∗10 -10

5∗10 -11
aqueous
0
0 100 200 300 400 500 600
Temperature /°C
Temperature-programmed methods 14 14
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Reactor Study

Reaction study of sample

1.4 MSA
MSA K 50/16 (fresh VPO) after
2 times TDS treatment:
1.3 m/e=26
m/e=26
•flow of 0.7vol% n-C4H10 /
MS-response (norm. u.)

m/e=44
m/e=44 20vol% O2 / 79.3vol% He
1.2
m/e=18 •25 ml/min @ 1bar
m/e=18
•T=400 °C
1.1 ??(MSA?) m/e=54
m/e=54 •30 mg pellet
(MSA?)

1
n-butane
n-butaneconversion:
conversion:18%
18%
0.9
m/e=41,
m/e=41,39
39
0.8 m/e=43
m/e=43
heating up to 400 °C
m/e=32
m/e=32(x10)
(x10)
0 10 20 30 40 50

Temperature-Time / min
programmed methods 15 15
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
 Comparison sample F8 “fresh” vs equilibrated

6∗10-13
fresh 4.5∗10-13
m/z: 54 m/z: 68

MS-response (I. C.)

MS-response (I. C.)

4∗ 10-13
5∗ 10-13

3.5∗ 10-13

4∗ 10-13
equilibrated 3∗10-13

3∗ 10-13 2.5∗10-13

0 100 200 300 400 500 0 100 200 300 400 500
Temperature / °C Temperature / °C

Desorption
Desorptionpeak
peakaround
around300
300°C°Ctypical
typicalfor
fornon-
non-conditioned
conditionedVPO
VPO
(compare
(comparetotoprevious
previousinvestigations);
investigations);organics
organicsfrom
fromsynthesis
synthesis

Temperature-programmed methods 16 16
M. Hävecker, Electronic Structure, Dept. Inorganic Chemistry, Fritz-Haber-Institut (MPG), Berlin, Germany
m/e=26
m/e=26
2∗10-12

MS-response (I. C.) 1.8∗10-12

after reaction

1.6∗10-12

1.4∗10-12

1.2∗ 10-12
after 1st TDS
1∗10-12

100 200 300 400 500

Temperature / °C

Desorption
Desorption of
of organic
organic fragments
fragments stemming
stemming not
not from
from synthesis
synthesis
after
after reaction
reaction (m/e=26:
(m/e=26: MSA?)
MSA?)

Temperature-programmed methods 17 17
 2∗10-12 3∗10-10 m/e=18
1.8∗10-12
2.5∗10-10
1.6∗10-12
m/e=26

MS-response (I. C.)

m/e=26
MS-response (I. C.)

2∗10-10
1.4∗10-12

m/e=39
m/e=39
1.2∗10-12
1.5∗10-10

1∗10-12
m/e=41
m/e=41 1∗10-10
8∗10-13
m/e=43
m/e=43 m/e=32
6∗10-13 5∗10-11
m/e=55
m/e=55 m/e=44
4∗10-13
m/e=68
m/e=68 100 200 300 400 500
100 200 300 400 500 Temperature / °C
Temperature / °C

2 types of organic species: product, intermediate(s),

no catalyst organics

Temperature-programmed methods 18 18
 gaps
• temperature-programmed methods are a
prototype area for the existence of gaps
between surface science and catalysis
characterisation.
• often same methodology and data
interpretation.
• but fundamental differences in boundary
conditions.

Temperature-programmed methods 19 19
 TPD: a complex process
• The net rate of TPD is considered as the ration
between adsorption and desorption
• in static condition: equilibrium
• in TPD: disturbance by pumping or gas flow
• boundary cases:
– re-adsortion possible (thermodynamic control
– re-adsortion suppresed (kinetic control) case of UHV
TDS analysis

Temperature-programmed methods 20 20
 TPD and TDS: relation between surface
science and catalysis charaterisation
• TDS low pressure (probe situation)
• TDS: kinetic control, no re-adsorption, usually
(tacitly) first order
• no clear distinction between sorptiopn and reaction
• TPD high pressure, reaction situation
• TPD: transition from kinetic to thermodynamic
regime explicitely studied
• changeover in reaction process from sorption to
reaction common

Temperature-programmed methods 21 21
 TDS: The method
• A pre-adsorbed species is removed from a
well-defined surface by rapid heating (10
K/s) in a well-pumped UHV environment
(no equilibration and re-adsorption) (caveat
TMP!). (Langmuir unit!)
• Care must be taken to see only desorption
from the surface!
Feulner and Menzel, J. Vac. Sci. Technol., 17, (1980), 662

Temperature-programmed methods 22 22
Temperature-programmed methods 23 23
 Surface composition
during reaction:
kinetics

Temperature-programmed methods 24 24
 TDS: The observation
• rate of desorption:

r = ν (Θ) Θn exp [-Edes (Θ)/RT]

r = desorption rate
n = desorption order
Θ= coverage in monolayers
ν = pre-exponential factor
ν can vary between
1013 and 1018 according
to transition state

Temperature-programmed methods 25 25
 TDS: Complete analysis

Integration to chosen
low desorption

Conventional analysis
for activation energy

Temperature-programmed methods 26 26
 TDS: The cheap analysis

Temperature-programmed methods 27 27
TDS: The “elaborate analysis”

Use width and peak desorption temperatures as

two parameters for finding activation energy and pre-factor

Temperature-programmed methods 28 28
 TDS: Quality of analysis

Temperature-programmed methods 29 29
 The basis of TPD
• Simplest case: A solid with adsorption sites S*
gets in contact with a gas G of concentration C
and populates in a first order process the vacant
adsorption sites.
• Langmuir boundary conditions:
– fixed number N of sites S* (F/cm²)
– constant adsorption enthalpy dHa (and desorption
enthalpy)
– all parmetres are temperature-independent and
coverage-independent

Temperature-programmed methods 30 30
 The basis of TPD
dN/dT = p k na (N*-N) – knd N
Langmuir model assuming the balance between competitive
adsortion and desorption kinetics
as net effect: the ratio of the two kinetic constants and the
adsorbate partial pressure p are the parameters.
k = σ (2πMRT)1/2
The kinetic constant relate to the nature of the adsorbed species by
the molecular mass M (g/mol),
the specific molecular surface area (cm-²),
and the gas constant (J/Kmol)
Temperature-programmed methods 31 31
 The basis of TPD
p=CRT
The partial pressure relates to the gas concentration.

The TPD experiment requires the temperature dependencies of the

sorption process: it is assumed that the process is thermally activated
(not spontaneously occuring); only a fraction na will be adsorbed:

na = Aa exp (-Ea/RT)

kd = Ad exp (-Ea/RT) desorption process

Temperature-programmed methods 32 32
 The basis of TPD

The experiment produces a signal proportional

to the change in gas concentration C with temperature

C = -S/F dN/dT
S denotes surface area in (cm²/g), F the flow rate (cm³(STP)/sg)

The experiment is time-programmed;

one obtains C as function of time t as initial observation data:

Temperature-programmed methods 33 33
 The basis of TPD
S N kd
C(t) =
F + σ (RT/2πM)1/2 na (N*-N)
It is important to run the experiment strictly linear in temperature:

T = T0 + βT with β being the heating rate in (K/s)

na is the fraction of adsorbing molecules from the
stream of species A and
Θ is the site occupancy: Θ = N/N*

Temperature-programmed methods 34 34
 Reaction rate: first order
S + N* Θ Ad exp(-Ed/RT)
C(T) =
F + S N* (1- Θ) σ (RT/2πM)1/2 Aa exp (Ea/RT)

F
dΘ/dT = - C(T)
S β N*

One observes the concentration of the adsorbate CT as function of

sample temperature
and can relate this to the physically relevant change in surface coverage
with temperature dΘ/dT

Temperature-programmed methods 35 35