Operando thermal analysis
Andrey Tarasov 8.12.2017
Andrey Tarasov, FHI MPG 1
Definition of TA Group of physical-chemical methods which deal with studying materials and processes under conditions of programmed changing's of the surrounding temperature.
Thermogravimetry Differential Scanning Calorimetry
∆M
t / T
Heat flow
t / T
d
dTCm
d
dQp
∆Q=m∙Cp∙∆T ∆m=M∙∆N
d
dm
Md
dNr
1
Enthalpy of formation/transformation (reaction, phase transition)
Course of the reaction, yield of the reaction
Key parameters:
Andrey Tarasov, FHI MPG 2
Terms and Definitions
A(s)↔ B(g) A(s)↔B(s)
A(s)↔B(g)
Definition of TA Group of physical-chemical methods which deal with studying materials and processes under conditions of programmed changing's of the surrounding temperature.
Differential Scanning Calorimetry
∆M
t / T
Heat flow
t / T
d
dTCm
d
dQp
∆Q=m∙Cp∙∆T ∆m=M∙∆N
d
dm
Md
dNr
1
Andrey Tarasov, FHI MPG 3
operando≈in-situ – „on site“→in heterogeneous catalysis „in reaction mixture under operation conditions“
T, P, μ→ conversion/actitvity, mass – transfer conditions
Terms and Definitions
Thermogravimetry
SMP - Measurement signal FB - Buoyancy force , f(T) mA - Mass of adsorbed gas, f(T) mSC - Mass of sample container mS - Mass of sample , f(T) VA - Volume of adsorbed gas , f(T) VSC - Volume of sample container VS - Volume of sample, f(T)
ρgas - Density of gas, f(T)
SMP • g = ((mSC + mS + mA) - (VSC + VS + VA) • ρgas) • g
Buoyancy Force System mass Signal
Andrey Tarasov, FHI MPG 4
Thermogravimetry: Basic Principle
Ionisation chamber
ampermeter
mirror
battery
Piezoelectric balance
88Ra, 84Po 1898, Marie Curie “weighted” radioactivity
Is it sensitive enough?
Mr(Ra)=225
Marie and Pierre Curies diary, 1902. Curie Museum
Andrey Tarasov, FHI MPG 5
1 g earth crust≈10-13g Ra
SMP - Measurement signal FB - Buoyancy force , f(T) mA - Mass of adsorbed gas, f(T) mSC - Mass of sample container mS - Mass of sample , f(T) VA - Volume of adsorbed gas , f(T) VSC - Volume of sample container VS - Volume of sample, f(T)
ρgas - Density of gas, f(T)
SMP • g = ((mSC + mS + mA) - (VSC + VS + VA) • ρgas) • g
Buoyancy Force System mass Signal
Andrey Tarasov, FHI MPG 6
Thermogravimetry: Basic Principle
T
T=const ?
P
P=const
Schematic diagram of the Cahn Electrobalnce
Balance
Commercially available thermobalances employ one of the Cahn-type electromagnetic balance
Andrey Tarasov, FHI MPG 7
Temperature detection
No contact Close proximity With contact
Andrey Tarasov, FHI MPG 8
Sample holders
Andrey Tarasov, FHI MPG 9
Sample holders
Andrey Tarasov, FHI MPG 10
Thermobalance construction geometries
Horizontal, parallel-guided Vertical, free suspended
Vertical, top-loader
Mettler Toledo
Netzsch, Setaram
TA instruments
Rubotherm
Andrey Tarasov, FHI MPG 11
Furnace Convection Currents and Turbulence
Andrey Tarasov, FHI MPG 12
Effect of furnace top opening on apparent mass-change, 5Kpm
200 400 600 800 1000 1200 1400Temperature /°C
-0.5
0.0
0.5
1.0
TG /mg
[13.1]
[12]
• Reproducible in different gas atmospheres and by different heating rates
5Kpm
Normal baseline
Anomaly baseline
100 Nml/min
15 Nml/min
gas flow decreases during the heating, starting from 600°C
Andrey Tarasov, FHI MPG, 14
Reassembling the oven Top part of the tube was plugged with solid material The material was analyzed by various techniques
Andrey Tarasov, FHI MPG, 15
Main crystalline phase is AlF3
XRD
Andrey Tarasov, FHI MPG, 16
Formula Concentration, %
AlF3 92.4 Bi2O3 0.45
CaO 0.62 Cr2O3 0.54
CuO 1.17 Fe2O3 1.69
K2O 0.55
MoO3 2.95 NiO 0.16
P2O5 0.33 Ru 0.642
SiO2 1.3
SO3 0.45
WO3 0.33
ZnO 2.21
XRF
Elemental analysis
Andrey Tarasov, FHI MPG, 17
SEM_EDX
Surface microstructure
Andrey Tarasov, FHI MPG, 18
HV 7kV fig -001a
a
b c
HV 20kV fig -002a
a
SEM_EDX
Andrey Tarasov, FHI MPG, 19
DSC
85 90 95 100 105 110 115 120Time /min
0.05
0.10
0.15
0.20
DSC /(mW/mg)
350
400
450
500
550
Temp. /°C
[14]
[14]
exo
37 38 39 40 41 42 43 44 45 46Time /min
0.60
0.65
0.70
0.75
DSC /(mW/mg)
400
420
440
460
480
Temp. /°C
[14]
[14]
exo
Tonset:450°C
• Reversible melting point (explains dynamic behaviour of the flow)
Ammount of the phase undergoes melting 1.5-3%
Q=-3.3 J/g(sample)
K1-3AlF4-6, CsAlF4 (NOCOLOK ®) Tm=560-570°C
eutectic LiF-NaF-KF (46.5-11.5-42) Tm=454°C AgF Tm=435°C
Zn Tm= 419°C
200 400 600 800 1000 1200 1400Temperature /°C
-0.5
0.0
0.5
1.0
TG /mg
[13.1]
[12]
Normal baseline
Anomaly baseline
Melting and blocking the opening
Andrey Tarasov, FHI MPG, 20
Nearing the plug flow conditions
Andrey Tarasov, FHI MPG 21
Andrey Tarasov, FHI MPG 22
Nearing the plug flow conditions
Case Study: Methanol Synthesis
Cu
ZnO
CO2 + 3H2 CH3OH + H2O
molkJHOHOHCHHCO /533 2322
molkJHOHCOHCO /41222
molkJHOHCHHCO /952 32
250°C, 30bar
Andrey Tarasov, FHI MPG 23
What is the surface composition during the operation?
High Preasure Thermobalance
Andrey Tarasov, FHI MPG 24
Case Study: Methanol Synthesis
OH H
Water
OH
HydroxyOO
OC
Carbonate
OO
H
C
Formate (bidentate)
O
CH3
Methoxy
H
C
Formyl
O O
O HC
Formate (monodentate)
OOC
Carboxyl (cis)
H
Case Study: Methanol Synthesis
Reaction intermediates and spectators:
Andrey Tarasov, FHI MPG 25
Gas Density (kg/m3)
CO/CO2/H2/He(6/8/59/27) 0.33
CO/Ar/H2/He(14/4/59/23) 0.34
CO2/He/H2 (12.4/28.8/58.8) 0.35
Density alignment
3 4 5 6 7 8 9 10
96
97
99
100
102
103
105
time / hrs
0.0
5.0x10-11
1.0x10-10
1.5x10-10
CO2
I / AM / %
CO2/COCO
400mg SiO2, 30 bar, 250°C, feed gas flow 120 ml/min
Andrey Tarasov, FHI MPG 26
Case Study: Methanol Synthesis
Case Study: Methanol Synthesis
10 15 20 25 30 35 40
98
99
100
101
102
103
104
CO
+1.2%
CO2/H
2
-0.7%
+1.9%
CO2
CO/H2
CO2/H
2
t / h
M
/ %
0
5
10
15
20
25
30
35
40
45
50
Co
nce
ntr
atio
n /
%
10 15 20 25 30 35 40 4597
98
99
100
101
102
M
/ %
85%H2/Ar
85%H2/Ar/1.5%H
2O
-0.8%
t / h
+1.3%
85%H2/Ar
m/z 18
0.5%
0
1x10-11
2x10-11
3x10-11
I /
A
Addition of 1.5% H2O (250°C, 1bar) TG Methanol Synthesis (250°C, 30bar)
Andrey Tarasov, FHI MPG 27
Group/M, gmol-1 Surface stoichiometry
∆M, % (Cu sites)
∆M, % (ZnO sites) Full ∆M,% (Cu/ZnOx sites)
H2O/18 water
1 0.72 0.49 1.21
OH/17 hydroxy
1 0.68 0.46 1.14
OCH3/31 methoxy
1 1.24 0.85 2.09
CHOO/45 formate
2/1 1.8/0.9 1.2/0.6 3/1.5
Andrey Tarasov, FHI MPG 28
Theoretical mass increase of the catalyst under estimation of full covered surface with one type of species
Case Study: Methanol Synthesis
Cu sites – 400 µmol g(cat)-1 ZnOx sites – 273 µmol g(cat)-1
Schumann et al. ACS Catalysis, 2015, 5 Schumann et al. ChemCatChem, 2014, 6
10 15 20 25 30 35 40
0.00
0.25
0.50
0.75
1.00
1.25
1.50
CO/H2
CO2/H
2
Su
rface
Cove
rageM
L
t / h
water derived species
carbon based species
CO2/H
2
10 15 20 25 30 35 40 45
0.0
0.5
1.0
1.5
2.0
85%H2/Ar
85%H2/Ar
85%H2/Ar/1.5%H
2O
Covera
ge / M
L
t / h
ZnOx
Cu
experimental curve
Case Study: Methanol Synthesis
Andrey Tarasov, FHI MPG 29
TG Methanol Synthesis (250°C, 30bar) Addition of 1.5% H2O (250°C, 1bar)
Target Reaction Dry reforming (DRM): CO2 + CH4 ⇌ 2 CO + 2 H2 ΔH0= 247 kJ/mol
Side reactions (⇨ catalyst deactivation because of coking):
Boudouard reaction: 2 CO ⇌ C + CO2 ΔH0= -171 kJ/mol
Methane pyrolysis: CH4 ⇌ C + 2H2 ΔH0= 75 kJ/mol
Suppression of side reactions:
high reaction temperature (900 °C)
⇨ high temperature stable, long term active, non coking materials required
⇨ ⇨ Ni/MgAl2O4 catalyst
CO2 + CH4 CO + H2
Ni
MgAl2O4-x
Andrey Tarasov, FHI MPG 30
Case Study: Dry Reforming of methane
CO2 + CH4 ⇌ 2 CO + 2 H2 ΔH0= 247 kJ/mol
• Continuous coke formation • Better performance at higher temperatures
9%wt/h
3%wt/h
2%wt/h
4.2mmol (CH4)/g(cat)/s
3.2mmol (CH4)/g(cat)/s
Activity Weight change
Andrey Tarasov, FHI MPG 31
Case Study: Dry Reforming of methane
• Continues coking process • Less coking at lower Ni concentrations
Andrey Tarasov, FHI MPG
Tarasov et al., Chem. Ing. Tech., 2014, 86, 1916-1924
32
Activity in DRM
Carbon formation
Case Study: Dry Reforming of methane
Definition of TA Group of physical-chemical methods which deal with studying materials and processes under conditions of programmed changing's of the surrounding temperature.
Thermogravimetry Differential Scanning Calorimetry
∆M
t / T
Heat flow
t / T
d
dTCm
d
dQp
∆Q=m∙Cp∙∆T ∆m=M∙∆N
d
dm
Md
dNr
1
Enthalpy of formation/transformation (reaction, phase transition)
Course of the reaction, yield of the reaction
Key parameters:
Andrey Tarasov, FHI MPG 33
Terms and Definitions
A(s)↔ B(g) A(s)↔B(s)
A(s)↔B(g)
Terms and Definitions
Differential Scanning Calorimetry (DSC) –
experimental methods for measuring thermal effects
(heat evolution and consumption)
- in extended range of temperatures;
- with improved productivity;
- with higher temporal resolution
Andrey Tarasov, FHI MPG 34
Differential – permanently compared to reference
Principle of combined thermocouple Registration the temperature of the object and temperature difference between sample and reference
Terms and Definitions
Differential Scanning Calorimetry (DSC) –
experimental methods for measuring thermal effects
(heat evolution and consumption)
- in extended range of temperatures;
- with improved productivity;
- with higher temporal resolution
Andrey Tarasov, FHI MPG 35
Differential – permanently compared to reference
Scanning – varied (programmed) temperature (permanently and/or stepwise)
Calorimetry – `near calorimetric´ (although reduced) precision (typically, 1-2% of
measured thermal value) – as a rule satisactory for a wide range of applications
Based on different technical principles and schemes but fundamentally same physical background: heat transfer and balance
TA – (Thermal analysis ) TS vs. t TG – (Thermogravimetric analysis) Δm vs. T
DSC - (Differential Scanning Calorimetry): Voltage to keep ΔT = TS-TR= 0 vs. T
DTA - (Differential Thermal Analysis) ΔT = TS-TR vs. T
DDTA - (Derivative DTA) dΔT/dt vs. T
RDTA (Reverse Differential Thermal Analysis) dt/dT vs. T
Calvet TA
DSC-TG
TG
DTA-TG
Basic Principles and Terminology
Calvet-DSC
Andrey Tarasov, FHI MPG 36
Andrey Tarasov, FHI MPG 37
q = kT + C(dT/dt) – heat removal and stortage
Based on different technical principles and schemes but fundamentally same physical background: heat transfer and balance
Basic Principles and Terminology
Disk shaped, ´Quantitative TA´ Calvet cell
38 Andrey Tarasov, FHI MPG
DSC Heat flux
Methods and Instruments
1. ‘Quantitative TA’
– multiple sensors
– tight contact between sensors and holders
– rapid heat removal from the sample
Methods and Instruments
C(dT/dt) 0 and
q kT
Andrey Tarasov, FHI MPG 39
- relatively simple design;
- wide range of T’s;
- compatible with other methods
(e.g. TG)
- calibration is critical and
can vary from sample to sample;
- gas-solid diffusion
1. ‘Quantitative TA’
– multiple sensors
– tight contact between sensors and holders
– rapid heat removal from the sample
Andrey Tarasov, FHI MPG 40
Methods and Instruments
AuAuPd, one layer 56 thermocouples
DSC sensors
Cu/Constantan
disk sensor on
Si(X) waffer.
CuNi disk sensor
on Ag plate.
Pt-PtRh(10)
diskshaped
thermocouple
Pt-PtRh(10) point
thermocouple
Mettler
Mettler Toledo
NETZSCH
Andrey Tarasov, FHI MPG 41
Methods and Instruments
Type
Composition
Temperature range, K Output voltage (0°C), mV Tmin Tmax
T Cu /constantan 3 670 20
J Fe/constantan 70 870 34
E chromel /constantan - 970 45
K chromel /alumel 220 1270 41
S Pt/PtRh (10) 270 1570 13
P Platinel/AuPd 500 1400 12
C W/WRe(26) - 2670 39
N Nicrosil/Nisil - 1200 39
Constantan - 58% Cu, 42% Ni
Chromel – 89% Ni, 10% Cr, 1%Fe
Alumel – 94% Ni, 2% Al, 1,5% Si, 2,5 % Mn
Nicrosil/Nisil – Ni, Cr, Silicon/Ni, Silicon
CuNi disk sensor
on Ag plate.
Cu/Constantan
disk sensor on
Si(X) waffer.
μ sensor τ sensor
Pt-PtRh(10)
diskshaped
thermocouple
Thermocouples
Platinel – 55% Pd, 31% Pt, 14% Au
Andrey Tarasov, FHI MPG 42
Methods and Instruments
2. ‘Heat-flux’, or ‘heat-conductive’, or Calvet-type DSC
– heat flux is measured outside of the sample assuming that it is
proportional to the T on different distances from it
Calvet calorimetry (incl. DSC):
- none of the TC’s measures the
temperature of the sample (!!!);
- the DSC signal is formed as T
between two cylindrical surfaces
(inner and outer) on different
distances from the sample
Andrey Tarasov, FHI MPG 43
Basic Principles and Terminology
2. ‘Heat-flux’, or ‘heat-conductive’, or Calvet-type DSC
– heat flux is measured outside of the sample assuming that it is
proportional to the T on different distances from it
Basic Principles and Terminology
Andrey Tarasov, FHI MPG 44
3D vs. 2D sensing: more efficient (complete) capturing of heat fluxes
from/to the sample – lower sensitivity to the sample properties !!!
the sample does not play the role of
‘thermal resistance’;
– no tight contact between sample (and reference) holder(s) and the sensor is required;
– 3D heat sensing;
– good gas-solid mass-transfer
conditions;
– high compatibility (e.g. with TGA)
– relatively low temporal resolution
Andrey Tarasov, FHI MPG 45
heat removal and storage
𝑞 = 𝐾 ∙ ∆𝑇 + 𝐶 ∙𝜕𝑇
𝜕𝑡≈ 𝐾 Δ𝑇 + 𝜏
𝑑Δ𝑇
𝑑𝑡 - Tian equation
q – heat flux at every moment of time K – specific thermal conductivity C – specific thermal capacity τ- K/C characteristic time constant
Concept
Andrey Tarasov, FHI MPG 46
q = kT + C(dT/dt) kT + C(dT/dt)= k[T + *(dT/dt)]
* = C/k – time constant of fluxmeter
Time constant is depended on design, construction and materials of the calorimetric cell.
Andrey Tarasov, FHI MPG 47
Concept
Andrey Tarasov, FHI MPG 48
k * ( C/k) k * ( C/k)
q @ * q @ *
Concept
q = k[T + *(dT/dt)]
Three ranges of characteristic times of the process should be considered:
Andrey Tarasov, FHI MPG 49
Concept
τӨ<0.1 τ
τӨ>τ
0.1 τ<τӨ < τ
Thermal inertia of the registration system completely shades the kinetics
When the kinetics may rather accurately be determined using Tian equation
When the available methods give kinetics that is somewhat distorted by the thermal inertia of a sample
50 Andrey Tarasov, FHI MPG
In-situ DSC
51 Andrey Tarasov, FHI MPG
15cm 2cm 1cm
0.4cm
In-situ DSC setup
In-situ DSC setup
GC + MS
52 Andrey Tarasov, FHI MPG
𝑞 =𝑑𝑄
𝑑𝑡/
𝑑𝑁
𝑑𝑡
Sinev et al.
2cm
1cm
Inlet
Outlet
53 Andrey Tarasov, FHI MPG
In-situ DSC setup
2cm
1cm
In-situ DSC setup
54 Andrey Tarasov, FHI MPG
Without quartz tube
With quartz tube
Transformation of gas phase molecules • Information on reaction kinetics (Ea) • Simulation of different conditions in-situ • Adsorption - desorption enthalpies
Simplified schematic representation
Transformation in solid state • Redox dynamics • Effects of oxygen diffusion into subsurface • Redestribution of bulk and surface oxygen • Oxygen binding energy (oxides) • Thermochemistry of defects formation
active state
∆HadsA
Eeq(A)
Gas state
Product state
∆HadsB
Adsorbed state
∆Hr
∆H#
Activation barrier
Eapp
Eeq(α)
´
Deactivated
reactivated
adaptive state
Phase α
Deficient Phase α
∆Hrelx
Phase β
∆Hr
∆H#
Adapted using: van Santen, R. A., Modern Heterogeneous Catalysis, Wiley, 2017 and Schlögl, R., Introduction to Heterogeneous Catalysis, Lecture, FHI Berlin, 2017
55 Andrey Tarasov, FHI MPG
What can we measure?
20 40 120 140
-48
-46
-44
-42
-40
-38
-36
-34
-32
-30
-28
-26
-24
-22
Q / J
/s/g
time / min
heat flowexo
I / A
He (*1/15)
C2H
4
Case study: Validation, C2H4 adsorption on Ag catalyst
∆Hads=-94kJ/mol(C2H4)
N(C2H4)=26µmol/g(cat)
5%Ag/SiO2, Max Lamoth
reversible
In-situ DSC 30°C, total flow 30mlm, Ptotal 1,15bar
12 hours SynAir 300°C
∆Hads=-87-33kJ/mol(C2H4)
N(C2H4)=33µmol/g(cat)
Microcalo 40°C, Prange 0-3mbar 3 hours SynAir 300°C
reversible
56 Andrey Tarasov, FHI MPG
Case study: Validation, C2H4 adsorption on Ag catalyst
∆Hads=-94kJ/mol(C2H4)
N(C2H4)=26µmol/g(cat)
5%Ag/SiO2, Max Lamoth
reversibile
In-situ DSC 30°C, total flow 30mlm, Ptotal 1.15bar, PC2H40.6mbar
12 hours SynAir 300°C
∆Hads=-87-33kJ/mol(C2H4)
N(C2H4)=33µmol/g(cat)
Microcalo 40°C, Prange 0-3mbar 3 hours SynAir 300°C
reversibile
57 Andrey Tarasov, FHI MPG
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035
Q /
kJ/
mo
l(C
2H
4)
N (C2H4) / mmol/g
Low energy centers may not be concerned in the flow conditions
25 30 35 40 45
-40
-35
-30
-25
Q / J
/s/g
time / min
heat flowexo
I / A
He (*1/15)
C2H
4
Case study: Validation, C2H4 adsorption on Ag catalyst
5%Ag/SiO2, Max Lamoth
In-situ DSC 30°C, total flow 30mlm, Ptotal 1.15bar, PC2H40.6mbar
12 hours SynAir 300°C
Microcalo 40°C, Prange 0-3mbar 3 hours SynAir 300°C
58 Andrey Tarasov, FHI MPG
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035
Q /
kJ/
mo
l(C
2H
4)
N (C2H4) / mmol/g
Low energy centers may not be concerned in the flow conditions
Desorption and readsorption are occurring simultaneously with diffusion
In UHV systems diffusion takes part in a very limited extent, while readsorption can be avoided using sufficiently high pumping speed
25 30 35 40 45
-40
-35
-30
-25
Q / J
/s/g
time / min
heat flowexo
I / A
He (*1/15)
C2H
4
Case study: alkane ODH over VSb/Al2O3
CnH2n+2 + ½ O2 CnH2n + H2O
CnH2n+2 + zO2 nCOX + (n+1) H2O
Catalysts – V-containing bulk and supported oxides
’Classical’ Red-Ox Kinetics over V-containing catalysts –
Mars-van-Krevelen* or oxygen rebound-replenish (ORR) mechanism
[O]S + A [ ]S + AO
[ ]S + O2 … [O]S
Sinev et al.
59 Andrey Tarasov, FHI MPG
CnH2n+2 + ½ O2 CnH2n + H2O
V(4+) state is optimal for
high selectivity to ethylene
V(5) V(4+) – jump of S @ small variations of E[O];
V(4+) (V3+) – slow decrease of S in a wide range
of E[O] variation
‘Chemical’ factor (i.e. state of surface sites and their interaction with HC’s)
is more important than ‘energy’ one
Case study: alkane ODH over VSb/Al2O3
60 Andrey Tarasov, FHI MPG
Sinev et al.
Case study: alkane ODH over VSb/Al2O3
Is E[O] (Hf)M2OX-1 – (Hf)M2OX ???
Sb2O5 Sb2O4 + ½ O2 -171 kJ/mole O2
Sb2O4 Sb2O3 + ½ O2 -383 kJ/mole O2
V2O5 V2O4 + ½ O2 -224 kJ/mole O2
V2O4 V2O3 + ½ O2 -442 kJ/mole O2
V2O3 2VO + ½ O2 -710 kJ/mole O2
(Hf)M2OX-1 – (Hf)M2OX E[O]
61 Andrey Tarasov, FHI MPG
Sinev et al.
M2OX M2OX-1
Phase A O-deficient Phase A
– D[O] (endo)
M2OX-1
Phase B
Hrel (exo)
(Hf)M2OX-1 – (Hf)M2OX
(endo)
D[O] – O-binding energy (in ‘phase homogeneity’ range);
(Hf)M2OX, (Hf)M2OX-1 – standard (tabulated) values;
Hrel – enthalpy of relaxation (elimination of vacancies, lattice re-organization,
etc.)
|(Hf)M2OX-1 – (Hf)M2OX| < |E[O] | < |D[O]|
Sinev et al. Experimentally measured value
Experimental approaches
62 Andrey Tarasov, FHI MPG