Oxygen transport in {Ba(Fe,Co,Zr)O_1tn3_1tn-_1tn_d63 [Ba(Fe,Co,Zr)O 3-delta] membranes [Elektronische Ressource] / vorgelegt von Young Chang Byun

Ybyun

Informations

• Publié par: rheinisch-westfalischen_technischen_hochschule_-rwth-_aachen
• Publié le: 01 janvier 2009
• Nombre de lectures: 13
• Langue: English
• Poids de l'ouvrage: 3 Mo

Extrait

Oxygen Transport in Ba(Fe,Co,Zr)O membranes 3- δ


Von der Fakultät für Mathematik, Informatik und Naturwissenschaften
der Rheinisch-Westfälischen Technischen Hochschule Aachen
zur Erlangung des akademischen Grades eines
Doktors der Naturwissenschaften
genehmigte Dissertation


vorgelegt von
Master of Science Young Chang Byun
aus Daejeon, Republic of Korea


Berichter: Priv.-Doz. Dr Michael Schroeder
Uni.-Prof. Dr Manfred Martin

Tag der mündlichen Prüfung: 19. Januar 2009




Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar.




Acknowledgements

Here I would like to express my cordial thanks to:

● Priv. -Doz. Dr. Michael Schroeder for the opportunity to conduct my Ph.D in Institute of
Physical Chemistry, RWTH Aachen. I also benefit from his never failing and
motivating support, scientific discussion, and corrections of the dissertation.
● Prof. Dr. Manfred Martin for scientific discussion and kind advices.
● David Müller for the assistance with thermal gravimetry.
● Daniel Röhrens for X-ray measurements.
● Badreddine Belghoul for the help with Rietveld Refinement.
● Dr. Thomas Kirschgen for the kind explanation about german things and institution life.
● Dr. Thomas Schiestel in IGB for sending Ba(Fe,Co,Zr)O powders. 3- δ
● Mrs. Resi Zaunbrecher for the help with SEM operation.
● Mrs. Alwine Naujoks for the fast purchase of chemicals and the explanation about
various equipments such as diamond saw and polishing machine.
● All my colleagues in Institute of Physical Chemistry, RWTH Aachen.
● All the people in workshop who gave me a special help.










Table of Contents
1 Introduction ….. …….. …….. …….. …….. …….. …….. …….. …….. …… 1
1-1 General introduction ….. …….. …….. …….. …….. …….. …….. …… 2
1-2 State of the art …. …….. …….. …….. …….. …….. …….. …….. …… 6
1-3 The Scope of the work … …….. …….. …….. …….. …….. …….. …… 7
2 Experimental Details … …….. …….. …….. …….. …….. …….. …….. …… 9
2-1 Preparation of membranes …… …….. …….. …….. …….. ……..
2-1-1 Fabrication of disc type membranes …….. …….. …….. …….. …… 9
2-1-2 Grinding and polishing of disc type membranes …….. ……..
2-1-3 Fabrication of fibre type membranes ………….. …….. …….. …… 10
2-2 Oxygen permeation ……………. …….. …….. …….. …….. ……..
2-2-1 Oxygen permeation for disc type membranes ………… ……… …… 10
2-2-2 Oxygen permeation through fibre type membranes ….. ……...
2-3 Thermal gravimetric experiment …….. …….. …….. …….. ……... …… 15
2-3-1 Relaxation by TG … …….. …….. …….. …….. …….. ……...
2-3-2 Nonstoichiometry of Ba(Fe,Co,Zr)O … …….. …….. …….. …… 15 3- δ
2-4 XRD …….. …….. …….. …….. …….. …….. …….. …….. …….. …… 19
2-5 SEM …….. …….. …….. …….. …….. …….. …….. …….. ……..
3 Model development of oxygen transport ….. …….. …….. …….. …….. …… 20
3-1 Model development of surface exchange reaction ….. …….. ……..
3-1-1 The First model (Model 1)
- prevailing incorporation of atomic oxygen ….. …….. …….. …… 22
3-1-2 The second model (Model 2)
- prevailing incorporation of molecular oxygen …….. …….. …… 25
3-2 Bulk transport …. …….. …….. …….. …….. …….. …….. …….. …… 26
3-3 Summary of both models …….. …….. …….. …….. …….. …….. …… 28
3-4 Determination of a and a ….. …….. …….. …….. …….. …….. …… 29 s1 s2
0 3-5 Determination of parameters (k and σ )ion
from permeation data of disc type membranes …….. …….. …….. …… 31
4 Results and Discussion ……... ……... …….. …….. …….. …….. …….. …… 32
4-1 Structure of Ba(Fe,Co,Zr)O and its lattice constant …….. ……... 3- δ
4-2 Nonstoichiometry of Ba(Fe,Co,Zr)O …….. …….. …….. …….. …… 33 3- δ
i

4-3 The dependency of oxygen vacancy concentration on oxygen activity …… 35
4-4 The thickness dependency of the oxygen permeation flux
0 and determination of parameters (k, σ , n) …….. …….. …….. …… 37 ion
0 4-4-1 Fit calculations with two parameters (k and σ ) …... …….. …… 38 ion
4-4-2 Influence of the existence of secondary phase
0 on two parameters (k and σ ) ... …….. …….. …….. …….. …… 47 ion
0 4-4-3. Fit calculations with three parameters (k, σ , n) …… …….. …… 50 ion
4-5 Modification of oxygen transport models …… …….. …….. …….. …… 53
4-5-1 Variable oxygen vacancy mobility …….. …….. …….. …….. …… 54
4-5-2 Variable surface exchange coefficient …. …….. …….. …….. …… 59
4-5-2-1 Surface exchange coefficient dependent
on gas phase oxygen activity …….. …….. …….. …….. …… 60
4-5-2-2 Surface exchange coefficient dependent
on surface oxygen activity … …….. …….. …….. …….. …… 65
4-5-3 Comparison of the modelling approaches outlined above …… …… 71
4-5-3-1 Surface exchange coefficient
independent of oxygen activity (k ≠f(a(O )) …….. ……... …… 72 2
4-5-3-2. Non-constant surface exchange coefficient (k=f(a(O )) …. …… 73 2
4-6 Oxygen activity profile across Ba(Fe,Co,Zr)O membranes ……... …… 74 3- δ
4-7 Oxygen permeation in fibre type membranes ……... …….. ……... …… 77
4-7-1 An analysis using a simple Wagner equation …. …….. …….. …… 79
4-7-2 Evaluation of oxygen permeation fluxes
0 with two parameter model ( σ , k) ……. …….. …….. …….. …… 84 ion
4-7-2-1 Co-current configuration ….. …….. …….. …….. ……..
4-7-2-2 Counter-current configuration …… …….. …….. …….. …… 86
4-7-2-3 Comparison of the oxygen permeation rates
in co-current and counter-current configuration .. ……... …… 87
4-8 Relaxation by thermal gravimetry …... …….. …….. …….. …….. …… 96
4-8-1 An analysis of relaxation curve … …….. …….. …….. …….. …… 102
4-8-2 Parametric study … …….. …….. …….. …….. …….. ……... …… 109
5 Summary and open questions …….. …….. …….. …….. …….. ……... …… 115
References ………………………. …… 120
Abstract …………………………. …… 126

ii
Chapter 1. Introduction

1. Introduction

1-1. General introduction
Oxygen, a colourless, odourless and tasteless gas, is widely used in industry. It is typically
supplied with a purity of 99.5 % or better. Major industrial consumers are metallurgical
industry (49 %), chemical industry (31 %), glass industry (6 %), cutting and welding
industry (6 %), among others [93]. Currently, oxygen is produced commercially by
cryogenic process (fractional distillation of liquefied air), PSA (pressure swing adsorption)
and VSA (vacuum swing adsorption). In cryogenic process, pure oxygen (above 99.5 %)
is obtained. However, production of liquefied air requires large amounts of energy, which
means that its primary cost of production of oxygen comes from the energy cost for
liquefying the air. On the other hand, in PSA or VSA, only oxygen with rather low purity
(O purity: 90-93%) can be obtained in a more economical way. 2
In recent years, membrane technology has been suggested as a promising alternative for
many industrial processes which require a continuous supply of pure oxygen [1-3, 13-14].
Compared to cryogenic process, PSA and VSA, the membrane technology may provide a
way to produce high purity oxygen from air at lower cost. The principle of operation of
the membrane technology for the oxygen separation from air is illustrated in Fig. 1-1
(MIEC membrane).
2-O Sweep gas O depleted 2
• + hAir
O2
••VO
Air - Sweep gas e

Fig. 1-1: Operation principle of MIEC membrane for oxygen separation
1
Chapter 1. Introduction
2- ••In Fig. 1-1, O denotes an oxygen ion and V denotes an oxygen vacancy with an O
•• • -effective double positive charge (indicated by the superscript ). h and e denote and
electron hole and an electron according to the Kroeger-Vink notation [30], respectively.
When oxygen activity gradient is imposed across the MIEC membrane, oxygen is driven
from the high oxygen activity side to the low oxygen activity side. Provided that the
membrane is fabricated dense, i.e. free of cracks and connected through porosity, it enables
the oxygen separation from the air with infinite selectivity because oxygen transport across
MIEC membrane takes place in the form of oxygen ions instead of oxygen molecules.
Owing to the ability to conduct both oxygen ions and electronic charge carriers, it operates
without the need of electrodes and external circuitry, in contrast to solid oxide fuel cells
and electrochemical oxygen pumps. The presence of significant electronic conductivity in
MIEC membranes