A WELL – DISPERSED CATALYTIC COATING PROCESS ONSTAINLESS STEEL MICRO REFORMER

Huang Yuh-jeen Abstract: In this thesis, we explored a binder to coat powders of Cu/ZnO catalyst onto stainless steel plates.. Prepared well-dispersed slurries were coated on the surface

THAI NGUYEN UNIVERSITY

UNIVERSITY OF AGRICULTURE AND FORESTRY

DUONG DINH TUAN

A WELL – DISPERSED CATALYTIC COATING PROCESS

ON STAINLESS STEEL MICRO – REFORMER

BACHELOR THESIS

Study Mode : Full-time

Faculty : International Training and Development Center

Thai Nguyen, September 2015

Thai Nguyen University of Agriculture and Forestry

Degree Program : Bachelor of Environmental Science and Management

Student name : Duong Dinh Tuan

Student ID : DTN 1153070102

Thesis Title : A well – dispersed catalytic coating process on stainless steel

micro – reformer Supervisor (s): Ph.D Phan Dinh Binh

Assocs Prof Huang Yuh-jeen Abstract: In this thesis, we explored a binder to coat powders of Cu/ZnO catalyst onto stainless steel plates The catalyst was prepared in the laboratory through the sequential precipitation method Coated plates were then used for fabrication of micro-channel reactor (MCR) to produce hydrogen Partial oxidation reforming (PO) of butane would

be used as a model reaction for the production.

A series of coating slurries were prepared by mixing the catalyst powders with the dispersant and binder Prepared well-dispersed slurries were coated on the surface or the micro-channels of stainless steel substrates through the coating method.

Adhesion stability of coated layers on the substrates was estimated by fraction of weight loss (FL) during a standard ultrasonic impulsion in a D.I water bath.

Keywords: Cu/ZnO catalyst, stainless steel plates, hydrogen,

micro-channel, weight loss (FL).

Number of Pages: 43

Date of Submission: 30/9/2015

I would like firstly to emphasize the sincere appreciation to teachers inInternational Training and Developments as well as teachers in Thai NguyenUniversity of Agricultural and Forestry, who have taught me knowledge notonly for my subjects but also for my living skills and gave me a chance to do mythesis abroad In addition, I would like to thank all supports and help fromBiomedical Engineering & Environment Science Department, National TsingHua University for the time I did my research in Taiwan

It is my pleasure to work with a great teacher - Associate Professor HuangYuh-Jeen, who always helped me any time She also gave me the bestconditions, supported all materials for my research and discussed about anyproblems I got whenever I did experiments in her Environment Nano Analysisand Energy Laboratory

I would like to give special thank to Dr Phan Dinh Binh, who alwayssupported and cheered me up whole the time I worked oversea He also helps

me a lot on spending much time for checking my thesis report

I consider it is an honor to work with Ms Janet, a master student, whoparticularly helpful in guiding me toward a qualitative methodology andinspiring me in whole period of internship time She was always helpful,friendly and very kind with me Without her guidance, I cannot complete thisthesis

Finally, I would like to express my gratitude to my family and friends,who always beside me all the time Their helps, supports and encouragementcreated the pump leading me to success.

Sincerely,

Duong Dinh Tuan

TABLE OF CONTENT

LIST OF FIGURES 1

LIST OF TABLES 2

LIST OF ABBREVIATIONS 3

CHAPTER 1: INTRODUCTION 4

1.1 Background 4

1.2 Objective 5

1.3 Limitations 5

1.4 Definitions 5

CHAPTER 2: LITERATURE REVIEW 7

2.1 An overview about experiment 7

2.2 Factors influence coating quality 8

2.2.1 Particle dispersion 8

2.2.2 Particle size distribution 9

2.2.3 Coating methodology 10

2.2.4 Drying and calcinations 13

CHAPTER 3: METHODS 15

3.1 Material 15

3.2 Experimental 17

3.2.1 Catalyst preparation 17

3.2.2 Catalyst slurry preparation 19

3.2.4 Adhesion test 21

3.3 Coating methods 21

CHAPTER 4: RESULTS AND DISCUSSION 22

4.1 Comparison between different wafer treatments 22

4.2 Comparison with different coating methods 26

4.3 Effect of binder content on adhesive ability of slurry 27

4.4 Proper catalyst/binder ratio 31

4.5 Effect of ethanol on adhesive ability of slurry 32

CHAPTER 5: CONCLUSION AND DISCUSSION 36

5.1 CONCLUSION 36

5.2 DISCUSSION 38

REFERENCES 39

LIST OF FIGURES

Fig 4.1 Wafer without treatment (only clean by acetone) 23Fig 4.2 Wafer was scraped by abrasive paper and treated with acetone, acid

(5M nitrate acid and hydrogen peroxide) and NaOH

24

Fig 4.3 Wafer treated with a 1:1:1:5mixture of hydrogen peroxide (H2O2),

phosphoric acid (H3PO4), acetic acid (CH3COOH) and NaOH

25

Fig 4.4 Effect of binder content on adhesive ability of slurry (Binder:

8% CeO2 colloid at pH=7, Catalyst: Cu/ZnO 100mg)

Fig 4.7 Effect of catalyst content with 5% PVA on adhesive ability (Binder

PVA, catalyst: Cu/ZnO)

Fig 4.10 The best result after ultrasonic vibration test of 5wt% PVA, 2%

catalyst with the ratio of ethanol to water is 1:1

37

LIST OF TABLES

Table 4.1 Different treatments showed different results of weight loss (2%

PVA, 10% catalyst Cu/ZnO)

22

Table 4.2 Different percentage of weight loss between open-channel injection

directly method and brushing method (boehmite 20%, 0.0224g bentonite, 0.5g

catalyst Cu/ZnO and 50 mL D.I water)

26

Table 4.3 Different ratios of binder PVA showed different results of weight loss 29

Table 4.4 The best result of weight loss is 5wt% PVA and 20mg catalyst in 1g

slurry (the solution is D.I water)

LIST OF ABBREVIATIONS

FC Fuel Cell

FEMFC Proton exchange membrane fuel cells

FL Fraction of weight loss during ultrasonic treatment

W Load amount of catalyst before the treatment

WL Load after the treatment

W2 Weight of wafer after calcine and ultrasonic vibration test

CHAPTER 1: INTRODUCTION 1.1 Background

In recent years, climate change is a “hot” issue on the world It becomes

more and more seriously day by day This is due to the fact that most of energysources people use every day comes from fossil fuel like oil, natural gas andcoal When we burn them, some polluted gas will release such as CO, NO2,

CO2…They are the major sources that cause greenhouse effect to climate

change Therefore, an environmental friendly and renewable energy is importantand necessary Today, there are many kinds of clean energy such as wind power,solar power, geothermal power and biomass could be a positive way to replacefossil fuel and safety for environment Fuel cell (FC) is one of the most efficientand potential devices in renewable energy It is a device that converts thechemical energy from a fuel into electricity through a chemical reaction withlow emission of pollutants

Proton exchange membrane fuel cells (PEMFC) are a type of fuel cell beingdeveloped for transport application as well as for stationary fuel cell applications

and portable fuel cell application One main part in PEMFC is catalytic process Catalyst plays an important role in PEMFC because catalysts increase the

reaction rates of the chemical reactions They lower the energy required for theinitiation of the reactions and thus make the reactions easier to occur Catalysts

do not have influence on the position of the equilibrium and do not enable thereactions that are forbidden by the thermodynamics Typical more than one

chemical reaction occurs in chemical reactors Catalysts may influence only thedesired reaction and thus increase the selectivity of the process This improvesthe utilization of the feedstock materials.

In previous research, they have been used many kinds of catalyst like Cu –

Mn Hopcalite monolithic catalyst, Cu/Mn/ZnO catalyst, nicken or aluminum

catalyst…that shows good results on coating process with low weight loss

(lower than 10 wt% loss) In this research, Cu/ZnO catalysts to coat on stainlesssteel plates

1.2 Objective

The main purpose of this research is to find the best ratio of Cu/ZnO catalystwhich was adapted with polyvinyl alcohol (PVA) as a dispersant and organicbinder to coating on the stainless plates with low percentage of weight loss The

lower rate of weight loss, the higher efficient in PEMFC is.

the reaction rate The smaller the catalyst particle size, the larger the surface areafor a given mass of particles.

CHAPTER 2: LITERATURE REVIEW

The quality of a slurry coating is determined by all its determining properties, such as loading, adherence, thickness and homogeneity.Many factors at each process unit which can influence the coating quality Thesefactors can be measured and controlled by various techniques

performance-2.1An overview about experiment

The most common way to deposit catalysts within the micro-channel is thewash-coating (slurry) technique The advantage of wash-coating method is that

it is a well-optimized catalyst that can be used directly However, catalystimmobilization into micro-channel has been a challenge, influenced by a lowinteraction between substrate surfaces and catalyst Most research studies useinorganic binder to improve adherence For commercial CuO/ZnO/Al2O3catalyst, alumina sol was often used as a binder Some researchers also usedZrO2 sol as binder to immobilize the catalyst onto a stainless steel micro-channel Nevertheless, using inorganic binder has some disadvantages Chen et

al reported that catalytic activity was significantly affected by the acidic solbecause the catalysts were partially dissolved in the acidic slurry Karim et alalso mentioned that the low pH of the catalyst slurry caused catalyst dissolving.Lin et al also reported that catalyst activity decreases as the ratio of CeO2 solincreases in the catalyst slurry Thus, inorganic binder is not recommended

because the catalyst’s active regions may be covered by the inorganic sol

In the present work, water was chosen as solvent because it is eco-friendlyand cost less compared to organic-based slurry However, the catalyst powdercan easily agglomerate in water, which makes it impossible to form a crack-freecatalyst coating Some researchers used hydroxyethyl cellulose to improvecatalyst dispersion Tadd et al used PVA and a ceria–zirconia binder to preparethe catalyst washcoat which then was ball-milled with zirconia grinding mediafor 48 h Hwang et al used polyvinyl alcohol (PVA) as a drying, controlchemical additive for pre-coated alumina adhesive layer Peela et al used PVA

and colloidal alumina for washcoating γ-alumina on stainless steel

microchannels In this study, the catalyst surface was modified by adding PVA

as dispersant to avoid the aggregation of particles and as binder for coatingcatalysts on the stainless steel micro-channel

2.2 Factors influence coating quality

To obtain a good quality in term of coating process, there are severalfactors which control the quality of coating process: particle dispersion, particlesize distribution, slurry characteristics, drying and calcinations etc

2.2.1 Particle dispersion

The Cu/ZnO catalyst particles should be properly dispersed in the slurrybefore milling, and this is enabled by its unique chemistry (i.e structure andproperties)

To model the Cu/ZnO catalyst, a Cu cluster composed of eight Cu atoms wasdeposited on a oxygen–terminated ZnO (0001) surface The electronic structure

was calculated using the DFT/PBE method within the periodic slab approach Afundamental problem when one wants to study heterogeneous catalytic processesfrom a theoretical frame- work is to include the effect of the high temperaturesand pressures typical of industrial heterogeneous catalysis The inclusion ofthermodynamic effects into the description of the catalyst model wasaccomplished via the initio thermodynamics method, which allowed toapproximately calculation surface Gibbs free energy differences from the DFTtotal energies of a set of Cu/ZnO structures As a result, a phase diagram wascomputed showing the thermodynamically most stable structures of a Cu/ZnOcatalyst model as a function of H2 and O2 partial pressures of the gas phase incontact with the catalyst From over 50 different Cu/ZnO structures withvarying hydrogen coverage, oxygen vacancies, Zn atoms, ZnO atoms, and withdifferent cluster shapes and atomic configurations, our results showed that themost stable structures under catalytic conditions presented coverage of 1/2 ML

of hydrogen Additionally, under severe reducing conditions a Zn atom attached tothe Cu cluster was preferred, in agreement with the experimental observation ofCu–Zn alloy formation.

2.2.2 Particle size distribution

Some studies (e.g Germani et al., 2007; Jia et al., 2007) have found thatthe finer particles produced by the milling of Cu/ZnO particles have goodcoating adherence Therefore, the milling process should be targeted at

producing Cu/ZnO particles of the desired particle size distributions Themilling of Cu/ZnO slurries to obtain these distributions have been done usingdifferent mill types: stirred bead mill (e.g Fadhel & Frances, 2001), ballmill (e.g Yang and Sigmund, 2003) and the jet mill (e.g Omura et al.,2005) The particles are mixed and fragmented by the mechanical agitationproduced inside these mills by the moving shafts and the grinding media.Stirred bead mills are increasingly used in many applications (e.g.pharmaceuticals, coatings, paints) for micro-grinding, dispersion anddisintegration of particles in the liquid phase (Fadhel & Frances, 2001) Due totheir increasing demand for producing fine and ultra-fine particles making themadvantageous compared to other mills, stirred bead mills have been thesubject of a growing number of scientific studies (Kwade, 1999).

2.2.3 Coating methodology

At present there are two industrial methods for coating Fecralloy®substrates (similar with stainless steel substrates) The first involves coatingthe Fecralloy®, which is already formed into a monolith, by pouring theslurry through or by suction (i.e sucking the slurry into the monolithchannels) This is followed by the removal of the excess slurry by gravitationaldraining or by applying pressure to clear the channels of all but requisitecoating thickness The composite is dried (normally above 100oC) andcalcined at 350 – 700oC for good adherence of the coating within itself and tothe monolith (Cybulski & Moulijn, 2006).The second method involves coating

the Fecralloy® before the monolith is constructed In this method, the coating

is applied onto the Fecralloy® by dip-coating, painting or spraying Thisprocess may be repeated several times to achieve the required coatingthickness After completion of the coating process, the monoliths can be formed

in the conventional way (Cybulski & Moulijn, 2006) One of the argumentsagainst coating application before the construction of the monolith is thatthe presence of the coating makes welding or brazing difficult (Määthänen &Lylykangas, 1990) For experimental studies in the laboratory, the coating ofFecralloy®, usually in the form of coupons or slabs, is mostly done usingone of the application techniques of the second method, with dip-coatingbeing the most preferred (Jia et al., 2007; Meille et al., 2005; Zhao et al., 2003;Fei et al., 2003; Schimpf et al., 2002; Valentini et al., 2001).Though not ascommonly used as dip-coating, the spraying of coatings on Fecralloy® hasalso been chosen as the laboratory method in some studies (Meille et al.,2005; Sidwell et al., 2003; Wu et al., 2001) Other methods have also been used

in the laboratory, such as electrophoretic deposition (Sun et al., 2007), chemicalvapor deposition (Janickle et al., 2000), physical vapor deposition which can becathodic sputtering, electron-beam evaporation or pulsed laser deposition(Kestenbaum et al.,2002; Roth et al., 1987) Each of these coating methods hasits advantages and disadvantages, and the method to be chosen in a givenapplication depends on the required coating properties (Meille, 2006)

The process of dip-coating involves immersing the Fecralloy® into theslurry for 15 – 20 s and withdrawing at a velocity of 3 – 12 cm/min This isfollowed by drying and calcination (Jia et al., 2007; Fei et al., 2003; Zhao etal., 2003; Schimpf et al., 2002; Wu et al., 2001) A spray of the slurry particles

is applied in the case of spraying This may be repeated until the desired coatingthickness is achieved (Schuessler et al., 2003) The slurry viscosity used for dip-coating differs from that used for spraying as the shear rate is many times largerduring spraying than immersing (Meille, 2006) Wu et al (2001) used both dip-coating and spraying methods to apply slurries on Fecralloy® mesh, though twodifferent slurries were used for comparison A slurry containing γ–aluminasuspended in poly-vinyl alcohol and water was sprayed, while a slurry

containing γ–alumina suspended in boehmite sol was dip-coated They found

that both methods achieved the same coating thickness, but the sprayed coatingwas more adhered than that of the dip-coated one Fei et al (2003) studied thecoating properties of dip-coated Fecralloy® rods using γ–alumina slurriesproduced from two different sol gel routes The effects of a number of thecoating process parameters, such as withdrawal speed (3 – 12 cm/min) andsoaking time (30 s and 5 min), were investigated Their results showed noeffects of the withdrawal speed and the soaking time on the coating thickness.The common disadvantage of dip-coating and spraying is the inconsistency

in the coating quality as evidenced from the wide variation in the coatingloadings (Jia et al., 2007; Wu et al., 2001) There is therefore a need to develop a

more consistent laboratory-scale method of coating Fecralloy® to produce thedesired coating properties In this study, the γ–alumina slurry is coated onto theFecralloy® at a controlled shear rate using an automatic film applicatorequipped with wire-wound bars The film applicator is chosen as the coatinginstrument because it had been designed consistency and reproducibility ofcoatings (Sheen Instruments Manual, 2006) The film applicator is a versatilebench top instrument which has been used not only in coating industries, butalso in paint, ink and paper industries (Braaten et al., 1998).

2.2.4 Drying and calcinations

The drying and calcination conditions used in the coating method caninfluence the coating properties, such as in the formation of cracks Fei

et al (2003) dip-coated Fecralloy® rods in γ–alumina using sol gelpreparation, and dried the coatings at room temperature for 1 h, then in anoven at 100oC for 2 h before finally calcining in a furnace at 500oC for

3 h They found that thicker coatings were achieved by dip-coating severaltimes, and that drying and calcining each layer separately could avoid crackscaused by large stresses It was discovered that to achieve an integrated coatingwithout cracks, the critical coating thickness should not be exceeded The dryingand calcination conditions reported in the literature have been in terms ofdurations (i.e 100oC and 350 – 700oC respectively for 1 – 3 h), without anyconsideration of the profiles adopted (Jia et al., 2007; Cybulski and Moulijn,2006; Fei et al., 2003; Zhao et al., 2003)

Similarly the heating conditions of coatings have been considered Theheating conditions of coatings in a furnace depend on how heat energy issupplied, e.g by blowing a stream of hot air or through an encapsulated filament(Ismagilov et al., 2005) The slow drying of coatings has been encouraged asthis was viewed to be influential in preventing coating cracks The fast drying ofcoatings can lead to high evaporation which quickly thins the coating film, thuscausing a detrimental breakage (Giani et al., 2006; Fei et al., 2003).

in air at ambient temperature is normally achieved with additions of a minimum

of 13% (by weight) chromium, and up to 26% is used for harsh environments.The chromium forms a passivation layer of chromium (III) oxide (Cr2O3) whenexposed to oxygen The layer is too thin to be visible, and the metal remainslustrous and smooth

Stainless steel is generally highly resistant to attack from acids, but thisquality depends on the kind and concentration of the acid, the surroundingtemperature, and the type of steel Type 316 is resistant to sulfuric acid below10% at room temperature All types of stainless steel resist attack fromphosphoric acid

The 300 series of stainless steel grades is unaffected by any of the weakbases such as ammonium hydroxide, even in high concentrations and at hightemperatures The same grades of stainless exposed to stronger bases such assodium hydroxide at high concentrations and high temperatures will likelyexperience some etching and cracking, especially with solutions containingchlorides such as sodium hypochlorite

Figure 3.1: Stainless steel plate and its composition

- Polyvinyl alcohol (PVA) is a water-soluble synthetic polymer It has theidealized formula [CH2CH(OH)]n It is used in papermaking, textiles, and avariety of coatings It is white (colourless) and odorless It is sometimes supplied

as beads or as solutions in water In this experiment, PVA act like a binder thatholds or draws all catalyst particles together to form a cohesive

- Ethanol also commonly called ethyl alcohol, drinking alcohol, orsimply alcohol is the principal type of alcohol found in alcoholic beverages,produced by the fermentation of sugars by yeasts Ethanol is

a volatile, flammable, colorless liquid with a slight chemical odor It is used as

an antiseptic, a solvent, a fuel, and, due to its low freezing point, the active fluid

in post-mercury thermometers Its structural formula, CH3CH2OH, is oftenabbreviated as C2H5OH, C2H6O or EtOH

- Field Emission Scanning Electron Microscope (FESEM): catalyst nanoparticles can be seen more clearly by FESEM FESEM produces clearer, lesselectro-statically distorted images with spatial resolution down to 1 1/2 nm.That's 3 to 6 times better than conventional SEM Smaller-area contamination

spots can be examined at electron accelerating voltages compatible with EnergyDispersive X-ray Spectroscopy Reduced penetration of low kinetic energyelectrons probes closer to the immediate material surface High quality, lowvoltage images are obtained with negligible electrical charging of samples.(Accelerating voltages range from 0.5 to 30 kV.) Need for placing conductingcoatings on insulating materials is virtually eliminated The FESEM (JSM-6330F, JEOL) used in this research is located at National Tsing Hua university,department of Engineering and System Science.

10 min, 1.32 g of glucose was added and the resulting mixture was transferredinto a 120mL of Teflon-lined stainless steel autoclave, followed by heating at

180◦C for 24 h The product was collected and washed with deionized water andethanol for several times by centrifugation The particles were dried in an oven

at 60◦C over night

The previous product ZnO/C nanoparticles 0.1g were mixed with coppernitrate (Cu(NO3)2 3H2O) 0.1g in 36 ml deionized water and ethanol with theratio is 3:1 CTAB 0.125g was added in to the solution and was ultrasonic for 30min 340 μL NH4OH was dropped into the mixture and stirred for more than 30min Then 255uL TEOS was added into the mixture following stirred for 20hr.Finally the core-shell particles were washed with ethanol for several times bycentrifugation again After dried in the oven, the product was calcined at 550◦Cfor 6hr By burning the carbon to make void, the yolk-shell Cu/ZnO was formed.

- Solvothermal synthesis is a method of producing chemical compounds It is

very similar to the hydrothermal route (where the synthesis is conducted in astainless steel autoclave), the only difference being that the precursor solution isusually not aqueous (however, this is not always the case in all literature uses ofthe expression) Using the solvothermal route gains one the benefits of both thesol-gel and hydrothermal routes Thus solvothermal synthesis allows for theprecise control over the size, shape distribution, and crystallinity of metaloxidenanoparticles or nanostructures These characteristics can be altered by changingcertain experimental parameters, including reaction temperature, reaction time,solvent type, surfactant type, and precursor type Solvothermal synthesis hasbeen used in laboratory to make nanostructured titanium dioxide, graphene,carbon and other materials

- Adsorption is the adhesion of atoms, ions, or molecules from a gas, liquid,

or dissolved solid to a surface This process creates a film of the adsorbate on