Great news for Emmishield! First test from the European Commission Improof Project shows that great benefits as energy savings, protection increase, cleaner air emissions. Conclusions: more uniform heat transfer, increased run lenghts, improved product selectivities, longer lifetime of the furnace, big energy savings, production increase. Emmishield works!!!!

1. 124B
479642
IMPROOF: INTEGRATED MODEL GUIDED PROCESS
OPTIMIZATION OF STEAM CRACKING FURNACES
K.M. Van Geem, F. Battin-Leclerc, G. Bellos, G. Heynderickx, W. Buysschaert, B.
Cuenot, M.R. Djokic, T. Faravelli, G. Theis, D. Jakobi, P. Lenain, A.E. Muñoz, J.
Olver, J.N. Dedeyne, S. Vangaever, P. Honnerová, Z. Veselý, P. Oud
DEPARTMENT OF MATERIALS, TEXTILES AND CHEMICAL ENGINEERING (MATCH)
LABORATORY FOR CHEMICAL TECHNOLOGY
2017 Spring National Meeting, San Antonio, Texas, March, 28, 2017

2. A SPIRE project

3. IMPROOF IS ALL ABOUT:
3
Renewable fuel characterization
- experimental activity
- kinetic mechanisms
Innovative Furnace System developments and integration
- oxy-fuel combustion - emissivity coating
- coke formation - 3D reactor testing
Advanced 3D modeling
- CFD RANS/LES
- Reactor optimization
- pilot plant simulation
Upscaling and Demonstration

4. OUTLINE
̶ Introduction
̶ Objectives
̶ High Emissivity Coatings
̶ 3D Reactor Technology
̶ Conclusions
4

5. OUTLINE
̶ Introduction
̶ Objectives
̶ High Emissivity Coatings
̶ 3D Reactor Technology
̶ Conclusions
5

6. CONCEPT AND OBJECTIVES
6
1. Novel combustion technology using alternative
fuels and oxy-fuel combustion
2. Demonstrate the individual impact of novel
emissive, reactor and refractory materials on
pilot scale (TRL5)
3. Demonstrate the power of advanced process
simulation (high performance computing and
CFD) for furnace design and optimization
4. Demonstrate the technical economic and
environmental sustainability of the IMPROOF
furnace at TRL6
5. Coke formation reduction and real time
optimization

7. OUTLINE
̶ Introduction
̶ Objectives
̶ High Emissivity Coatings
̶ 3D Reactor Technology
̶ Conclusions
7

8. EMISSIVITY: RADIATIVE HEAT TRANSFER
Three fundamental modes of heat transfer:
• Conduction
• Convection
• Radiation
Planck’s law describes the spectral emission intensity of a
blackbody:
8
Over 90 % of the heat transfer from the steam cracking furnace to the reactor coils arises from radiation
Heynderickx, G.J. and M. Nozawa, Banded gas and non-gray surface radiation models for
high‐emissivity coatings. AIChE journal, 2005.
( , ) =
2
( / − 1) = /
=
With:
the speed of light in vacuum
Planck’s constant
Boltzmann constant

9. EMISSIVITY: RADIATIVE HEAT TRANSFER
9
Integration of Planck’s law yields the Stefan-Boltzmann equation:
With the Stefan-Boltzmann constant
( ) the total hemispherical emissivity
In reality no object behaves as a perfect blackbody
Annaratone, D. (2010). Engineering Heat Transfer, Springer Berlin Heidelberg.
= , d = ( ) !
The deviation of the
spectral radiance from
a perfect blackbody is
given by the emissivity:

10. The following figure shows the smoothened surface "# and the actual rough surface "$
The following concepts are defined:
apparent emissivity %&& (experimentally observed)
intrinsic emissivity '()$ (material characteristic)
Using geometric optics the relation between apparent emissivity
and intrinsic emissivity is given by:
%&& = 1 +
1
'()$
− 1 +
,
The roughness factor + is introduced, with + being the ratio of the
smoothened surface area, "#, over the actual rough surface area, "$.
RELATION BETWEEN APPARENT AND INTRINSIC EMISSIVITY
10
Wen, C.-D.; Mudawar, I., Modeling the effects of surface roughness on the emissivity of aluminum alloys. International Journal of Heat and Mass Transfer 2006
"#, %&&
"$, '()$
-./01
= 2 and R =0.5
%&&=0.66

11. RELATION BETWEEN APPARENT AND INTRINSIC EMISSIVITY
11
S.G. Agababov, Effect of roughness factor on the radiation properties of a solid body with random roughness, Teplofizika Vysokikh Temperature 1975
How to determine the surface roughness factor +:
1) Optical 3D profilometry
2) Develop empirical correlation relating + with more conventional surface roughness parameters
+ = 1 + 1.25 4 5 +%
,
http://www.filmetrics.com/opticalprofilers
With +% the arithmetical mean roughness and 5 the number of intersections of the
surface profile with the mean per unit length of the mean line. In this case the
profilogram is also required!

12. EMISSIVITY: EXPERIMENTS AT NTC
12

13. EMISSIVITY: SAMPLES
13
Metal samples
Ceramic samples
• Steel, ceramic and coated disc substrates with diameter of 25 mm thickness of 5 mm
• Two novel high-emissivity coatings by Emisshield designed for radiant walls/floor of furnace
and process coils/tubes
• Reference 304 SS and Alamo® refractory bricks by HarbinsonWalker International

14. EMISSIVITY: CERAMIC SAMPLES
14
• Temperature independent in 700 – 850 °C
Emisshield Coated Alamo® Coated vs. Uncoated Alamo®
• Big difference in the sub 5 µm region
• Coating the refractory wall can significantly
improve the thermal efficiency
• Decrease of spectral emissivity is observed for wavelengths shorter than 5 µm.
• Method needs further improvement to evaluate this.
90% of blackbody
radiation is emitted
with a wavelength
smaller than 7.5 µm
at 980 °C

15. EMISSIVITY: METAL SAMPLES
15
Coated vs. Uncoated 304 SS Metal Coating vs. Ceramic Coating
• Decrease of spectral emissivity is observed for wavelengths shorter than 5 µm.
• Clear difference exists between the samples
throughout the entire range of wavelengths
• Once the 304 SS material is oxidized, spectral
emissivity will increase and close the gap
• Emissivity of 304 SS is slightly higher than in
literature – due to roughness
• Two coatings designed for metal and ceramics
• Although different in composition they have
very similar emissivity

16. NEXT STEPS
̶ Further investigation of the sub 5 µm region
̶ Determination of the effective directional emissivity
̶ Comparison of 5 different coatings for wall and coils
̶ Selection of the best coatings and testing their
performance in the pilot plant furnace/cracker at
UGent at TRL5 level
• benchmark to uncoated ethylene furnace by calculating the
thermal efficiency of the improved furnace
16

17. OUTLINE
̶ Introduction
̶ Objectives
̶ High Emissivity Coatings
̶ 3D Reactor Technology
̶ Conclusions
17

18. Optimization by
- Feed additives
- Metallurgy & surface technology
- 3D reactor technology
Deposition of a carbon layer on the reactor surface
Thermal efficiency
Product selectivity
Decoking procedures
Estimated annual cost to industry: $ 2 billion
L. Benum, "Achieving Longer Furnace Runs at NOVA Chemicals," in AIChE Spring National
Meeting, 14th Annual Ethylene Producers’ Conference, New Orleans, Louisiana, 2002.
COKE FORMATION
18

19. 3D REACTOR TECHNOLOGY
Improve the reactor by decreasing Tgas/coke
Increase tube area (A)
Increase heat transfer coefficient (U)
19
6 = 7 · " · 9%#/ :;< − =>;
M. Zhu, "Large eddy simulation of thermal cracking in petroleum industry," 2015.

20. (3D) SIMULATIONS ARE NEEDED
20
NEED VALIDATION!!!

21. PIV Liquid Crystal Thermography
Measuring velocity
and 3D flow profile
Measuring tube wall temperature
and heat transfer
Both experiments were performed at the
von Karman Institute for Fluid Dynamics
EXPERIMENTS
21

22. PIV SETUP
22

23. PIV SETUP
• Length = 980 mm
• D = 150 mm
• e = 5.4 mm
• P = 63 mm
• Re ~ 24000
23

24. LIQUID CRYSTAL THERMOGRAPHY SETUP
24

25. Nikon Camera > RGB > Hue > Temperature
35 °C – 55 °C ± 1 °C
IR Camera > Intensity > Temperature
-20 °C – 1500 °C ± 0.5 °C
25
LIQUID CRYSTAL THERMOGRAPHY SETUP

26. Computational domain
5 000 000 cells
26
LES VS RANS – RIBBED DUCT (RE = 40000)

27. DO YOU REMEMBER SLEDGE HAMMER!?
27
trust me I know what I'm doing

28. Reattachment behavior
differs along the rib
Average reattachement point per pitch
1.40
0.00
U/Ub
X/e (-) 28
LES SIMULATION

29. ̶ Feedstock 118.5 kg/h propane
̶ Propane conversion 80.15 % (± 0.05%)
̶ Steam dilution 0.326 kg/kg
̶ CIT 903.7 °C
̶ COP 170 kPa
Different geometries simulated
̶ Same reactor volume
̶ Same axial length
̶ Same minimal wall thickness
Bare c-RibFin
MILLISECOND PROPANE CRACKER SIMULATION
29

30. RUN LENGTH SIMULATION
• Several mesh updates, each corresponding to 24 hours (c-rib,
bare) or 48 hours (fin) of coke layer growth
• Heat flux updated to keep propane conversion constant
SOR 96h48h
Increasing run length
30

31. PRODUCT SELECTIVITIES
Minor effect on total olefin
selectivity
Radial mixing effects cannot
be predicted based on 1D
simulations only
Reactor pressure drop
30% higher (fin)
300% higher (c-Rib)
1D: “Lower olefin selectivity”
66.1% 65.8% 66.5%
0%
10%
20%
30%
40%
50%
60%
70%
bare fin c-rib
1.3-C4H6
C3H6
C2H4
31

32. NON-UNIFORM COKE LAYER GROWTH
SOR (0 hrs)
48 hrs
Fin c-Rib
SOR
10 days
32

33. INCREASED HEAT INPUT
Bare
c-Rib
Fin
Heat
input
relative
to SOR
[-]
Heat input to the reactor is updated after each mesh update, to keep the
propane conversion constant: more cokes = more heating.
33

34. PRESSURE DROP
Pressure drop increases
Cross-sectional flow area
decreases due to coke
Bare
c-Rib
Fin
Less fast increase for c-rib compared to bare and finned geometry
34

35. TUBE METAL TEMPERATURE
Max. TMT increasesThermal resistance coke layer
Bare
c-Rib
Fin
TMT increases at the same rate for all geometries, but
absolute max. TMT lower for 3D geometries
35

36. OUTLINE
̶ Introduction
̶ Objectives
̶ High Emissivity Coatings
̶ 3D Reactor Technology
̶ Conclusions
36

37. CONCLUSIONS
̶ All spectral emissivity experiments appear to be temperature independent in the
observed temperature range (700 – 980 °C)
̶ The tested Emisshield coatings significantly increase the spectral normal emissivity
compared to the uncoated material
̶ Most of the profit will come from coating the refractory wall
̶ Combining the advanced coil materials and novel 3D reactor designs can lead to
more uniform heat transfer
increased run lengths
improved product selectivities
longer lifetime of the furnace
37

38. ACKNOWLEDGMENT
̶ The work leading to this invention has received funding from the
European Union Horizon H2020 Programme (H2020-SPIRE-04-
2016) under grant agreement n°723706.
Thank you for your attention!
38

39. GLOSSARY
̶ TRL: technology readiness level
̶ LES: large-eddy simulation
̶ RANS: Reynolds-averaged Navier-Stokes
39

40. Kevin M. Van Geem
Associate Professor
LABORATORY FOR CHEMICAL TECHNOLOGY
E Kevin.VanGeem@ugent.be
T +32 9 264 55 97
M +32 478 57 38 74
www.ugent.be
Ghent University
@ugent
Ghent University