SIMULATION OF METHYL ETHYL KETONE (MEK)

PRODUCTION IN A FIXED BED REACTOR

FATIN NUR LIYANA BINTI MOHD IDRIS

UNIVERSITI SAINS MALAYSIA

2017
SIMULATION OF METHYL ETHYL KETONE (MEK)

PRODUCTION IN A FIXED BED REACTOR

by

FATIN NUR LIYANA BINTI MOHD IDRIS

Thesis submitted in partial fulfilment of the requirement

for the degree of Bachelor of Chemical Engineering

June 2017
 ACKNOWLEDGEMENT

First and foremost, I would like to convey my sincere gratitude to my supervisor,

Associate Professor Dr. Norashid Bin Aziz for his patience guidance, precious

encouragement, and generous support throughout this work. I want to express my

appreciation to my parents for their enormous love, brother and sisters for their

unconditional support and encouragement.

I would also extend my gratitude towards all my colleagues especially Belinda

Anne Xavier and Amirul Hakim for their kindness cooperation and helping hands in

guiding me carrying out this work. They are willing to sacrifice their time in guiding and

helping me throughout this work besides sharing their valuable knowledge.

Once again, I would like to thank all the people, including those whom I might

have missed out and my friends who have helped me directly or indirectly. Their

contributions are very much appreciated. Thank you very much.

FATIN NUR LIYANA BINTI MOHD IDRIS,

May 2017

ii
 TABLE OF CONTENTS
Page
ACKNOWLDEGEMENT ii

TABLE OF CONTENTS iii

LIST OF TABLES v

LIST OF SYMBOLS vi

LIST OF ABBREVIATION vii

ABSTRAK ix

ABSTRACT x

CHAPTER ONE : INTRODUCTION 1

1.1 Research background 1

1.2 Problem statement 3

1.3 Research objectives 3

1.4 Organization of thesis 4

CHAPTER TWO : LITERATURE REVIEW 6

2.1 Introduction 6

2.2 Dehydrogenation of 2-butanol 6

2.3 Current industrial process 11

2.3.1 Bio-based 12

2.3.2 Petrochemical-based 13

2.4 ASPEN model 15

iii
CHAPTER THREE : METHODOLOGY 17

3.1 Introduction 17

3.1.1 The research methodology steps 18

3.2 Methodology flow chart 19

3.2.1 Data collection 19

3.2.2 Run simulation 22

3.2.3 Examine simulation results 22

3.2.4 Sensitivity analysis 23

3.2.5 Optimization 25

CHAPTER FOUR : RESULTS AND DISCUSSION 27

4.1 Comparison of simulated results and literature data 27

4.2 Model sensitivity analysis of dehydrogenation process 28

4.2.1 Case 1: The effect of pressure towards the conversion of SBA 29

4.2.2 Case 2: The effect of reactor temperature towards the conversion

of SBA 30

4.2.3 Case 3: The effect of SBA feed flow rate towards the conversion

of SBA 32

4.3 Optimization 33

CHAPTER FIVE : CONCLUSION AND RECOMMENDATION 35

5.1 Conclusion 35

5.2 Recommendation 36

REFERENCES 37

APPENDICES 41

iv
 LIST OF TABLES

Page

Table 3.1: MEK plant specifications 20

Table 3.2: Properties of the catalyst bed 20

Table 3.3: Reactor specifications 21

Table 3.4: Reactor length for the first reactor for different conversions 21

Table 3.5: Range operating variables used for sensitivity analysis 24

Table 4.1: Sensitivity results at different reactor pressure 29

Table 4.2: Sensitivity results at different temperature 31

Table 4.3: Sensitivity results at different SBA feed flow rate 32

Table 4.4: Optimum conversion for SBA obtained 34

v
 LIST OF FIGURES

Page

Figure 3.1: Flow for Properties section 17

Figure 3.2: Schematic flow diagrams of methodology 18

Figure 3.3: Experimental data of the reaction (Olsson, 2013) 23

Figure 4.1: The literature and simulated equilibrium curve for SBA

conversion at various operating temperature 28

Figure 4.2: Sensitivity results curve for conversion of SBA over pressure 29

Figure 4.3: Sensitivity result curve for conversion of SBA over temperature 31

Figure 4.4: Sensitivity result curve for conversion of SBA over

SBA feed flow rate 33

Figure A.1: Component used in this simulation 41

Figure A.2: Property method used in this simulation 41

Figure A.3: Flowsheet of Aspen Plus Simulation for dehydrogenation process 42

Figure A.4: Stoichiometric coefficient 42

Figure A.5: Reaction kinetic data 43

Figure A.6: PFR configuration 43

Figure A.7: Catalyst specification 43

Figure A.8: Error percentage between literature and simulated equilibrium

curve 44

vi
 LIST OF SYMBOL

Symbol Description Unit

dp Length m

𝜌p Particle density kg/m3

λs Thermal conductivity W/(m.K)

cp Heat capacity kJ/(kg.K)

𝜀B Porosity of the bed -

X Conversion %

vii
 LIST OF ABBREVIATIONS

Al2O3 Aluminium oxide

BET Brunauer–Emmett–Teller

CDS Chemical Data System

C4H8 Butene

CH2O2 Formic acid

C3H6O3 Lactic acid

C4H6O4 Succinic acid

C4H8O2 Acetoin

C4H10O2 2,3-Butanediol

C6H12O6 Glucose

CO2 Carbon dioxide

COPRE Co-precipitation

Cr2O3 Chromium oxide

Cu Copper

CuO Copper oxide

DRUV-Vis Diffuse Reflectance UV–Vis

EDTA Ethylenediaminetetraacetic acid

H2 Hydrogen gas

H2O Water

H3PO4 Phosphoric acid

IM Impregnation

ICP Inductively coupled plasma

La2O3 Lanthanum oxide

MEK/ C4H8O Methyl Ethyl Ketone

viii
MgO Magnesium oxide

O2 Oxygen

PFR Plug Flow Reactor

RPLUG Plug Flow Reactor

RSM Response Surface Method

SBA/ C4H9OH Sec-Butanol Alcohol

SG Sol-gel

SiO2 Silicon dioxide

TEA Triethanolamine

TPR Temperature-Programmed Reduction

UNIQUAC Universal QuasiChemical

XPS X-Ray Photoelectron Spectroscopy

XRD X-ray diffraction

Zn Zinc

ZnO Zinc oxide

ix
 SIMULASI PENGELUARAN METIL ETIL KETON (MEK) DALAM

REAKTOR KATIL TETAP

ABSTRAK

SBA atau sec-butanol alkohol merupakan produk perantaraan untuk menghasilkan metil

etil keton (MEK) yang merupakan pelarut untuk ejen dan pembuang cat pembersihan.

Proses nyahhidrogenan SBA telah disimulasi, dianalisis dan dioptimumkan dengan

simulator Aspen Plus versi 8.8. SBA model nyahhidrogenan dihasilkan dan disimulasi di

Aspen Plus. Reaktor katil tetap merupakan reaktor berbilang tiub dimodelkan oleh reaktor

aliran palam (PFR). Hasil yang diperolehi bagi pengesahan menunjukkan keputusan

simulasi hampir mengikuti trend data literasi pada SBA penukaran di mana ralat yang

paling tinggi ialah 18% dan yang paling rendah ialah 0.1%. Selain itu, analisis sensitiviti

dilakukan dengan menggunakan model alat analisis dalam Aspen Plus untuk mengkaji

kesan tindak balas tekanan, suhu dan kadar aliran suapan SBA ke penukaran SBA. Nilai

optimum diperolehi dalam analisis sensitiviti adalah 0.5 bar, 400 ° C dan perubahan yang

terlalu kecil untuk kadar aliran suapan SBA. Dalam seksyen pengoptimuman, tujuannya

adalah untuk mengurangkan kadar aliran molar SBA di aliran produk untuk mencapai

tertinggi dioptimumkan penukaran SBA. Pengoptimuman itu dapat dioptimumkan nilai

SBA kadar aliran molar sebanyak 13.4911 kmol / jam dengan tertinggi dioptimumkan

SBA penukaran yang merupakan 99.96% apabila suhu optimum dan tekanan masing-

masing adalah 550 ° C dan 0.5 bar.

x
 SIMULATION OF METHYL ETHYL KETONE (MEK) PRODUCTION IN A

FIXED BED REACTOR

ABSTRACT

SBA or sec-butanol alcohol is an intermediate product in order to produce methyl ethyl

ketone (MEK) which is a solvent for cleaning agents and paint removers. SBA

dehydrogenation process has been simulated, analysed and optimized with the Aspen Plus

software version 8.8. SBA dehydrogenation model is developed and simulated in Aspen

Plus. The fixed bed reactor which is multi-tubular reactor are modelled by plug flow

reactor (PFR). The result obtained for the validation shows the simulation results almost

follow the trend of the literature data on SBA conversion whereby the highest error is

18% and the least is 0.1%. Besides, sensitivity analysis is performed in model analysis

tool in Aspen Plus to study the effect of reaction pressure, temperature and SBA feed

flow rate towards the SBA conversion. The optimum values obtained in sensitivity

analysis are 0.5 bar, 400°C and insignificantly changes for the SBA feed flow rate. In

optimization section, the objective is to minimize the molar flow rate of SBA at the

product stream in order to achieve highest optimized SBA conversion. The optimization

is able to optimized the value of SBA molar flow rate at 13.4911 kmol/hr with highest

optimized of SBA conversion which is 99.96% when the optimum temperature and

pressure are 550°C and 0.5 bar respectively.

xi
 CHAPTER ONE

INTRODUCTION

1.1 Research Background

Butanol can be considered as an important industrial chemical which is useful as

a fuel additive, feedstock chemical in the plastic industry and food grade extractant in the

food and flavor industry. Every year, 10 to 12 billion pounds of butanol are produced by

petrochemical implies and the requirement for this commodity will probably increase

(Bramucci, et al., 2013). Sec-butanol alcohol (SBA) is an intermediate product in order

to produce methyl ethyl ketone (MEK) which is a solvent for cleaning agents and paint

removers. Since SBA is an isomer of butanol it is anticipated to have similar fuel

properties. Besides, volatile esters of SBA are used in perfumes and artificial flavors. It

is a chiral compound, and hence it exists as two enantiomers (Waldron, 2010).

MEK is a volatile, highly flammable liquid, colorless organic liquid with acetone-

like odor (Foxall, 2010). MEK can form azeotropes with a number of organic solvents

since it is partially miscible with water and many conventional organic solvents

(Melorose, et al., 2015). In safety and health, methyl ethyl ketone does not cause a

significant threat to the environment except in cases of major spills or discharges. It is

also a relatively safe organic solvent (Organization, 2014) and there are no prove on

methyl ethyl ketone causing cancer to humans (Foxall, 2010). The production of MEK is

important for automotive, construction, electronic and furniture industries based on local

and global demand.

MEK is produced by dehydrogenation of sec-butanol with the presence of a

catalyst. The catalyst used in this reaction is a copper based catalyst with the production

1
of hydrogen as the by-product. In most of the production plant, the production involves

two reaction steps. Firstly, it will start with n-butene whereby the n-butene will be

hydrated to produce sec-butanol. Then, the hydrated n-butene (SBA) will undergo

dehydrogenation reaction to produce MEK. This production route dominates the MEK

production industry in which 90% of the world production plant has chosen this route

(Schulze and Homann, 1989) and it is the second largest method of utilizing n-butene

produced by refinery of butane and butene in oil and gas industry (Kent, 2007). Due to

the rapid development of the oil and gas industry, these production routes is considered

to be comparatively cheaper.

Usually, complicated problems in industry may not be solved by hand because of

two reasons which is human error and time constraints. There are various simulation

programs used in industry depending on the field, application and desired simulation

products such as for entire process unit or one piece of equipment. Aspen can be a very

powerful tool when used to its full capabilities for a Chemical Engineer in a different of

fields including oil and gas production, refining, chemical processing, environmental

studies and power generation (Bernards and Overney, 2004). The advantages of

modelling using by Aspen plus are it can reduce plant design as well as allows designer

to quickly test various plant configurations. Besides, it also helps in improve current

process as determines optimal process conditions within given constraints and assists in

locating the constraining parts of a process that will be carried out using model analysis

tools such as sensitvity analysis and optimization tool. Furthermore, the use of a process

flowsheet simulator is beneficial in research and development (R&D), design and

production. In R&D, it helps to cut down on laboratory experiments and pilot plant runs

while in design, it enables a speedier development with simpler comparisons of various

2
alternatives. Meanwhile, in production it can be used for risk-free analysis of various

what-if-scenario (Sylvia and Orchanian, 2001).

1.2 Problem Statement

The demand of methyl ethyl ketone (MEK) has been expanding throughout the

years as there is a positive interest viewpoint in the paints and coating industry, where it

is broadly utilized as a solvent. However, the present issue confronted is that the use

feedstock from butane refined in oil and gas industry to deliver MEK. This clearly

pictures the assets are restricted because of its non-renewable properties. Therefore,

depletion of fuels can occur overtime and can lead to their constrained supply.

Additionally, extracting and blazing oil produces hurtful gases that adds to natural

contamination and consequently, to a dangerous atmospheric deviations. This means that

degradation of our ecosystems will be made faster since our feedstock to produce MEK

is based on refined oil and gas industry. Besides, there is lack of studies on optimization

of this production by simulation, thus an optimization of the production of MEK in a fixed

bed reactor using ASPEN plus is introduced in this paper to enhance the yield and

selectivity of the MEK production. Furthermore, in order to maximize the amount of

MEK, some parameters such as operating temperature and pressure are needed to be

controlled because it will affect the reaction process.

1.3 Research Objective

The conversion of SBA into MEK has been known for long and the reaction is

applied industrially. Considering the importance of the present process, the simulation of

the dehydrogenation of SBA in a fixed bed reactor is done by using Aspen plus. Tubular

3
reactor of fixed bed reactor is chosen and it will be modelled as plug flow reactors with

heterogeneous catalytic reactions. The objectives of the present project are following.

i) To simulate the production of MEK in a fixed bed reactor and validate the

literature data.

ii) To study the effect of pressure, temperature and SBA feed flow rate to the SBA

conversion by using sensitivity analysis tool.

iii) To optimize the SBA conversion based on the variables involve in the sensitivity

analysis part.

1.4 Organization of the Thesis

Chapter 1: Introduction

This chapter outlines the general information about research background for example

explanation of the raw material used, product and the reaction process occur in order to

produce the product. It also contains information on sensitivity analysis and optimization.

Besides, the problem statement is elaborated about past research study while research

objectives will be considered about the validation of the simulated and literature data, the

sensitivity of the variables such as pressure, temperature and flow rate affect the SBA

conversion. The last objective is optimization of the SBA conversion based on the

variables in the sensitivity part.

Chapter 2: Literature review

This chapter outlines literature review regarding from previous researcher’s paper that is

related to this project which is a simulation on MEK production in a fixed bed reactor.

The type of MEK production include bio-based and petroleum-based are also be

discussed. The issues on modelling and analysis tools will be discussed too.

4
Chapter 3: Methodology

This chapter presents the methodology of the research. Aspen plus is used to develop the

simulation flowsheet for the RPLUG model in dehydrogenation process. Sensitivity

analysis on pressure, temperature and SBA feed flow rate is analysed in Aspen plus. Then,

the analysis is proceed by optimization of the SBA conversion. All the steps are represent

in a methodology flow chart and it is clearly explained for each of steps involve.

Chapter 4: Results and Discussion

This chapter presents the results and discussion of the simulation results. The simulated

data is compared to the literature data. From sensitivity analysis result, the most

significant effect on the SBA conversion will be used in optimization part. The

optimization is carried out to further increase the SBA conversion as well MEK produce.

Chapter 5: Conclusion and Recommendation

This chapter concludes the research for each of the objectives that have been carried out

in this study. Recommendation is a suggestion or proposal as to the best course of action.

Thus, the recommendation for this paper is made in order to improve or optimize the SBA

conversion.

5
 CHAPTER TWO

LITERATURE REVIEW

2.1 Introduction

This chapter will provide the review from previous researcher’s study that is

related to this project which is a simulation on MEK production in a fixed bed reactor.

Each of the previous research will be discussed on the issues raised from the papers, the

methods used and the results from the experimental works. Besides, this chapter also will

be explained about the type of processes involved and reason for chosen the particular

process. Since, this paper is concerned on software such as simulator, thus the issues on

the modelling and the analysis tools will be discussed too.

2.2 Dehydrogenation of 2-butanol

The researchers of Dehydrogenation of 2-butanol over copper-based catalyst are

studied about the effect of support type (MgO and SiO2) and copper loading on methyl

ethyl ketone (MEK) yield. Besides, they also covered the effect of reaction temperature,

2-butanol feed flow rate and particle size of catalyst. This copper-based catalyst can be

unsupported or supported catalyst and mostly they are supported catalyst where provides

a large surface area compared to the unsupported type. The dehydrogenation reaction will

be favoured if the acidity support has high pH. The 4 basic techniques for preparing

copper-based catalyst are precipitation, urea hydrolysis, electroless plating and

impregnation. Thus, in this study, it used a porous support impregnated copper-based and

kinetic parameter for 2-butanol at atmospheric pressure and 190 to 280°C is determined.

The dehydrogenation of 2-butanol is modelled in a catalytic membrane reactor that is

6
required pure kinetic data. The catalyst is performed in two stages where for the first

stage it will determine the optimum catalyst and for second stage, the optimum catalyst

will determine the kinetic parameter. The result of the reaction of 2-butanol over MgO

and SiO2 shown MEK as the main product and butene as the by-product. MgO supported

catalyst has low BET surface area, so the MEK yield also will low while the silica

supported catalyst gave highest yield of MEK since it has an optimum copper

concentration on the support which is 15wt% and it particles is in the range of 300-850

μm. The amount of butene increased either smaller or larger particles produced. The

selectivity of MEK for a 15wt% Cu on silica catalyst is closed to 100% at 240°C while at

390°C the selectivity is decreased between 83 and 86%. The catalyst is deactivated at

310°C and it stable at 250°C over 24 hours. The MEK adsorption coefficient is the main

concern compared to 2-butanol and hydrogen coefficient. There is a little interphase mass

transfer resistance for this 2-butanol dehydrogenation reaction (Keuler, et al., 2001) .

In another research, they are studied about the chemical activation of olive stone

with phosphoric acid that is prepared for the catalytic conversion of 2-butanol on a

carbon-based acid catalyst. The activated carbon catalyst is found from olive stone

residues. The catalyst activities are measured by the decomposition of 2-butanol that is

operated at 1 atm by using quartz fixed bed reactor placed inside the furnace with

temperature control. From result, the chemical activation of olive stone with H3PO4

produced an activated carbon with a high apparent of surface area and a wide microporous

structure. The mechanism for the activated carbon to be formed is through formation of

phosphate and polyphosphate bridges that connect crosslink biopolymer fragments that

is avoid the contraction of structure from effect of the temperature. Decomposition of 2-

butanol is occurred at 413 to 573 K but in the absence of catalyst there is no reaction takes

place below than temperature of 750 K. The carbon catalyst is active in produce olefin by

7
dehydration of 2-butanol rather than production of MEK by dehydrogenation of 2-butanol

especially at high conversion. This paper proposed of two 2-butanol decomposition

mechanisms for dehydration and dehydrogenation as A and B models. E2 elimination

mechanism is for dehydration reaction of model A while in model B the dehydration

reaction is assumed to take place through an E1 elimination mechanism. The

dehydrogenation reaction occurs from both cases by an E2 mechanism. The rate

expressions derived from both models fit properly with the experimental results that

shown the two mechanisms occur simultaneously (Bedia, et al., 2010).

Next is investigated the catalytic performance tests in vapor-phase

dehydrogenation of 2-butanol to MEK. Two types of copper-based dehydrogenation

nanocatalyst which are Cu/ZnO/Al2O3 and Cu/SiO2 prepared from various precursors by

impregnation method (IM), sol-gel (SG) and co precipitation (COPRE) methods. The

objective of this study to investigate the influences of the preparation method and reaction

conditions on the catalytic properties of copper-based nanocatalysts in MEK production.

The nanocatalyst prepared by co precipitation has the general form of Cu/ZnO/Al2O3

while the Cu/SiO2 is general form for prepared by impregnation and sol-gel methods.

Vapor-phase dehydrogenation of 2-butanol is carried out by CDS (Chemical Data

System) unit. The reactor used is fixed bed reactor that is loaded by 3 g of nanocatalyst

in the middle section of the reactor. To prevent the product from condensate, the transfer

line from reactor outlet to the separated vessel is externally heated and maintain at 120°C.

From the result, impregnation method shown the highest total surface area and pore

volume while the lowest value for the total surface area is co precipitation method and

pore volume for sol-gel method. The COPRE method is obtained the highest selectivity

to MEK and conversion of 2-butanol. Meanwhile, the IM method obtained the lowest

value of MEK selectivity and 2-butanol conversion. The optimization is done by using

8
response surface method (RSM) to optimize the response based on the factors studied.

According to statistical results, the optimum level of parameters that can maximize the

selectivity to MEK is precipitation temperature at 65°C, aging temperature at 68.75°C,

pH of precipitation stage is 7.27 and Cu/Zn molar ratio is 1.38 (Geravand, et al., 2015).

Furthermore, another research is study about the Cu-ZnO-Cr2O3/SiO2 catalysts

that are prepared by impregnation which shown high activity for the dehydrogenation of

2-butanol to 2-butanone. The catalytic activity test is carried out in a 0.6 cm diameter of

a stainless reactor with pressure of 1 atm. The catalyst loaded for each experiment is 0.3

g. From the result, the specific surface area of all the catalysts are smaller that means

some small pores of SiO2 were blocked during impregnation. Furthermore, the specific

surface areas of the catalyst inversely proportional to the increasing of reduction

temperature. This occur might be because of microcrystalline metallic copper sintering.

The conversion of 2-butanol over the catalyst calcined at 673 K for 3 h is very high at the

beginning but it becomes slowly decreasing. The catalyst reduced at 723 K shows a

relatively low conversion because the reduction at high that leads to sintering and to

decrease the amount of active sites on the surface. CuO exhibits high catalytic activity for

dehydrogenation of 2-butanol however it is not stable at the beginning of the reaction.

Thus, with reduction of CuO, the amount of Cu0 increases so the catalytic activity

becomes stable. The selectivity to 2-butanone increases since the catalyst reduced at

higher temperature and the selectivity to 5-methyl-3-heptanone which is the by-product

is decreases. The reduction temperature shown the decreasing of CuO amount and

consequently the selectivity of condensation product 5-methyl-3-heptanone is decreases

because of the reaction reaches a stable state. The condensation of 2-butanone is occurred

due to existence of CuO in the catalyst (Wang, et al., 2002).

9
 The other research using co-precipitation, the copper-based catalysts, supported

on γ-Al2O3, La2O3 and γ-Al2O3- La2O3, respectively, are prepared and tested in the

continuous dehydrogenation of 2-butanol to methyl ethyl ketone. The objective of this

paper is to examine the effect of support acid-base properties and the catalytic activity

and stability on the copper-based catalysts and correlate both of them. The 2-butanol

dehydrogenation is carried out in a tubular, fixed bed reactor with internal diameter of 20

mm and 5 g catalyst is loaded into the reactor. The catalysts are characterized by using

XRD, ICP, XPS, TPR, DRUV-Vis. From the results, the actual Cu contents of the three

catalysts are slightly lower than nominal content which is 30wt%, with Cu-Al2O3- La2O3

having highest amount of Cu. The Cu-Al2O3 had the maximum surface area and pore

volume while Cu-La2O3 is conversely. Conversion of 2-butanol for Cu-Al2O3 is 93.1%

but MEK selectivity is only 83.5% while for Cu-La2O3 the 2-butanol conversion is only

84.1% and 96.4% is the MEK selectivity. For Cu-Al2O3- La2O3 the conversion of 2-

butanol and MEK selectivity is the highest which are 92.2% and 95.8% respectively.

Thus, the catalytic performance of copper-based catalysts are dependent on acid-base

properties of the support in dehydrogenation of 2-butanol. Cu-Al2O3- La2O3 shows the

best activity and stability (Bai, et al., 2013).

Besides, another research is studied about the EDTA and TEA chelating agents

that are used to prepare seven Cu/γ-Al2O3 catalysts of different copper contents, 8.44 -

2.56wt% and 7.09 – 25.92wt% respectively. The activities of these catalysts are tested

using dehydrogenation of 2-butanol as model reaction. The aims of this paper is to study

the effect of the effects of the chelating agents on the surface morphology and dispersion

of copper on the alumina substrate and on the dehydrogenation activity of 2-butanol. The

copper surface area and acidity of these catalysts are measured using volumetric

chemisorption method. The dehydrogenation reactions are carried out in a microreactor

10
and operated under atmospheric pressure. The activities are measured in the range of 473

– 573 K and the reactant flow rate is maintained at 10.93 mL/hr. Based on the result of

this paper, for both chelating agents, the BET surface are gradually decreased which

indicates that the micropores of the support might be blocked by some copper crystallites.

The preparation of Cu/Al2O3 catalyst with TEA took 5 minutes and with EDTA it took

30 minutes to complete the plating process. The rate of dehydrogenation of 2-butanol

increased proportional to the temperature as well as the number of exposed copper atoms.

The complexing capability on the properties of the chelating agent dominates the copper

crystallites particle size, thus, the copper dispersion on the alumina support. The Cu-TEA

chelate enhanced the formation of bulky copper clusters occurring even at low copper

loading. The catalysts prepared by Cu-EDTA chelates had smaller copper crystallites,

higher copper dispersion and higher activity than those prepared by Cu-TEA chelates.

The acidity of the catalyst could be reduced more effectively for the EDTA series (Chang,

1997).

2.3 Current industrial process

Currently, MEK is commercially produced using the two step of n-butene

technique which is hydrating butene to produce SBA and dehydrogenating the SBA.

Besides, there are two methods to produce MEK via dehydrogenation of SBA, first, gas

dehydrogenation and second, liquid dehydrogenation. However, gas dehydrogenation is

widely used in the industry because it is comprise a simpler process flow sheet and its

investment is low.

By this process, it can give high conversion of SBA and high selectivity of MEK

of about 95%. The key to technical development is the improvisation of the catalyst

11
activity, selectivity and operation cycle (Zhenhua, 2006). Furthermore, the commercial

alcohol dehydrogenation catalyst are copper, brass or zinc based. The copper based

catalysts are found to be quite effective in dehydrogenation of SBA because of their lower

operation temperatures and minimum side reactions. Hence, the advantages of this

process are better yield, longer catalyst life, simple production separation and lower

energy consumption.

2.3.1 Bio-based

The production of MEK also can be done through biological path by using

lignocellulosic biomass as the substrate. Lignocellulosic biomass which is rice straw and

bagasse is one of the most abundant waste found on earth that were composed by

cellulose, hemicelluloses and lignin (Wong, et al., 2012). Firstly, alkaline pretreatment is

done to reduce the lignin content then the treated biomass will be subjected to enzymatic

hydrolysis to convert the cellulose and hemicelluloses to fermentable sugar and this

process is costly (Weerasai, et al., 2014) . The hydrolysate that contains fermentable sugar

(i.e glucose) will be sent to the bioreactor for batch fermentation.

Microorganisms such as Klebsiella oxytoca, Klebsiella pneumonia and

Saccharomyces cerevisiae are used to produce 2,3-butanediol from fermentation of

glucose. Since the process will undergo a series of biochemical path through pyruvate

formation, thus there will be other product produced as well such as acetoin, lactic acid,

succinic acid, ethanol, carbon dioxide and hydrogen. The process also will take up to

three days at temperature between 30°C to 37°C with an initial pH of 6 to 6.5 and no

control will be done on the pH throughout the fermentation period (Wong, et al., 2012)

(Wong, et al., 2014) (Cho, et al., 2015). The formation of these products from glucose is

deduced as follow (2.1) to (2.5):-

12
 Glucose (C6 H12 O6 ) → 2,3 Butanediol (C4 H10 O2 ) + 2CO2 + H2 (2.1)

Glucose (C6 H12 O6 ) → Acetoin (C4 H8 O2 ) + 2CO2 + 2H2 (2.2)

Glucose (C6 H12 O6 ) + O2 → Succinic acid (C4 H6 O4 ) + 2 Formic acid (CH2 O2 ) + H2 (2.3)

Glucose (C6 H12 O6 ) → 2 Lactic acid (C3 H6 O3 ) (2.4)

Glucose (C6 H12 O6 ) + O2 → 2 Acetic acid (C2 H4 O2 ) + 2 Formic acid (CH2 O2 ) (2.5)

The fermentation broth will be subjected to a series of separation process to obtain

2,3-butanediol. Then, when the 2,3-butanediol is purified, it will undergo catalytic

dehydration process that take place in liquid phase. The reaction operated at temperature

of 300°C to 400°C with pressure range from 0 to 50 bar. The dehydration reaction is

shown below (2.6):-

C4 H10 O2 → MEK(C4 H8 O) + H2 O (2.6)

Based on the reaction, apart from MEK, 1,3-butanediene also be produced as the

side product (Soo, et al., 2016). In this section, the separation will involve azeotropic

compound of MEK and water.

2.3.2 Petrochemical-based

Production of MEK can be synthesized by the dehydrogenation of sec-butanol (2-

butanol) process. In most of the production plant, the production involves two reaction

steps. Firstly, it will start with n-butene whereby n-butene will be hydrated to produce

sec-butanol. After that, the hydrated n-butene (sec-butanol) will undergo

dehydrogenation reaction to produce MEK. This production route dominates the MEK

production industry in which 90% of the world production plant has chosen this route

13
(Schulze and Homann, 1989). Besides, the proposed production route is the second

largest method of utilizing n-butene produced by refinery of butane and butane in oil and

gas industry (Kent, 2007). Due to the rapid development of the oil and gas industry, these

production routes is considered to be comparatively cheaper. The two reactions involved

in this process is summarized in the following stoichiometry equation:-

n − butene(C4 H8 ) + H2 O → Sec − butanol(C4 H9 OH) (2.8)

C4 H9 OH → MEK(C4 H8 O) + H2 (2.9)

Ion exchange catalyst has the advantage in terms of safety and environmental

friendly characteristics because liquid strong acid is not used and less hazardous waste

water will be generated. Moreover, less by product will be generated by the direct

hydration route too. Up to the date of publication, Deutsche Texaco plant in Meerbeck is

the only plant that utilizes direct hydration (Schulze and Homann, 1989). Followed the

hydration process will be the dehydrogenation of sec-butanol. The hydrogen product

produced in this reaction can be recovered for sale or utility purpose. Even though this

route of production is considered to be easier, its separation may be tedious as the

upstream product and reactant will form azeotrope. The unreacted water may form

azeotrope with the unreacted sec-butanol and MEK. Overall there are two azeotropes

involve.

Controversial been raised on the MEK production using n-butene since n-butene

is considered as a non-renewable source which will depleted sooner and later. However,

instead of waiting the resource to be depleted, scientist been divert their attention to create

the n-butene from renewable source. Recently, research been carried out to produce n-

butene can by the dehydration of butanol that derived from fermentation approach (West,

et al., 2009). Bio-1,3-butadiene hydration could also be another butane production

14
alternative (Umpierre, et al., 2005). Hence, the feedstock sustainability should not be a

large problem in future.

Based on the bio-based and petrochemical-based, this paper is proposed to choose

the petrochemical-based rather than bio-based due to higher cost for the enzymatic

hydrolysis process for the bio-based. Besides, there will be other product produced as

well such as acetoin, lactic acid, succinic acid, ethanol, carbon dioxide and hydrogen

while the petrochemical-based only produce MEK as the main product and hydrogen.

The catalyst used in petrochemical based has the advantage in terms of safety and

environmental friendly characteristics and the production route is cheaper.

2.4 ASPEN model

The importance of having simulation in predicting reaction process model by

using basic engineering relationships, such as mass and energy balances, and phase and

chemical equilibrium. Given reliable thermodynamic data, realistic operating conditions,

and rigorous equipment models, the actual plant behaviour can be simulated (Aspen

Technology, 2001). The Aspen plug flow reactor is used to simulate the dehydrogenation

of SBA in a fixed bed reactor because the reaction kinetic of this reaction is known. In

this step, SBA is converted to MEK and hydrogen. Other reactions are possible, but their

reaction mechanisms are still not clear. RPlug handles kinetic reactions, including

reactions involving solids, since the kinetic is given in the literature, so the RPlug is

chosen as the reactor model. The reaction kinetics must be determined before using RPlug

to model a reactor.

Sensitivity analysis is a tool for determining how a process changes with varying

key operating and design variables. It can be used to vary one or more flowsheet variables

and study the effect of that variation on other flowsheet variables. It is a valuable tool for
15
carrying out “what if”’ studies. The flowsheet variables are inputs to the flowsheet.

Besides, sensitivity analysis also can be used to verify whether the solution to a design

specification lies within the range of the manipulated variable. The optimization process

is carried out to achieve the optimum output variables.

16
 CHAPTER THREE

METHODOLOGY

3.1 Introduction

In order to achieve the objectives as stated in Chapter 1, a process model is

developed for SBA dehydrogenation in a fixed bed reactor using Aspen plus version 8.8

software. First, a suitable reactor block in Aspen plus is chosen to simulate the data. Next,

suitable information and assumption were taken into account for the reactor block. By

using the model developed, the three parameters which are pressure, temperature and

SBA flow rate towards reactor were studied. The data were simulated by using sensitivity

analysis in Aspen plus. Regarding to validate the simulation results, experimental data

from literature is used. The experimental data from case study is used to validate the

model in order to determine the model fits with the experimental data. If the validation

succeeded, the optimization of the model can be performed by using optimization tool in

Aspen plus.

There are few steps involved before creating a simulation model. First, open the

software, the menu will display the Properties. Select the Specifications and fill the

components used for this simulation. Then, click Methods in order to choose the Base

method before select the Simulation part. The Figure 3.1 shows the flow for the Properties

section in Aspen Plus.

Properties Specification for the Method Run the Properties

Figure 3.1: Flow for Properties section

17
3.1.1 The research methodology steps

A general flow of the methodology is shown in Figure 3.2.

Start

Collection of data from literature

Simulation on MEK production

Comparison the result with

NO
Validation

YES

Sensitivity analysis

Study the effect of different

Optimization

Satisfied
?n

End

Figure 3.2: Schematic flow diagrams of methodology

18
3.2 Methodology flow chart

The detail information based on the flow diagrams in Figure 3.2 will be

explained as below.

3.2.1 Data collection

Aspen plus simulator is used to model and predict the performance of a process

which involves the decomposition of the process into its constituent elements for

individual study of performance. It is widely used to study and investigate the effect of

various operating parameters on various reactions. The fixed bed reactor which is tubular

reactor were modelled by PFR in Aspen plus according to the data that found from case

study provided (Olsson, 2013). UNIQUAC (Universal QuasiChemical) is used as the

property method for the Aspen model. A PFR model in the Aspen plus is known as

RPLUG. In the case study setup the reactant which is SBA is fed to the RPLUG at 12.08

tons/hr. The mass fraction of feed solution is 1. Temperature and pressure of feed stream

are set as 500 °C and 300 kPa respectively. Length of the reactor is 0.46 m from calculate

by interpolation and the diameter of the tube is 32 mm.

There are some assumptions used:

1. There is no accumulation of impurities in the process.

2. The simulation does not take into account the heat loss that is present throughout the

process.

3. In the SBA dehydrogenation process, the main component for the output is MEK and

the by-product is hydrogen gas.

4. No other reaction is occurred others than main reaction.

19
5. Mole of hydrogen gas in output is ignored in calculation for conversion, selectivity

and yield in order to get accurate results of calculation.

6. Since there is no fixed bed reactor in Aspen model, RPLUG is chosen as the model

to simulate the reaction.

Table 3.1 to Table 3.4 are the data obtained from the literature. The Table 3.1 shows the

MEK plant specifications. Table 3.2 is the properties of the catalyst bed meanwhile the

Table 3.3 is reactor specifications. The Table 3.4 is reactor length for the first reactor for

different conversions.

Table 3.1: MEK plant specifications

MEK production 90000 tons/year

Raw material SBA

Conversion in the reactor 96%

Purity of produced MEK 99 wt% MEK

Loss of MEK from the plant 1 wt% based on feed

Operation time 8000 hr/year

Flue gas stack temperature 150 °C

Table 3.2: Properties of the catalyst bed

Length dp 0.0032 m

Particle density 𝜌p 8500 kg/m3

Thermal conductivity λs 125.4 W/(m.K)

Heat capacity cp 0.38 kJ/(kg.K)

Porosity of the bed 𝜀B 0.39

20
 Table 3.3: Reactor specifications

Given values from the course project

Fluid viscosity 1.89×10-5 Pa.s

Wall thickness 4 mm

Thermal conductivity 15 W/(m.K)

Overall heat transfer coefficient 50 W/(m2.K)

Minimum temperature difference for 50 °C

heating the reactor

Calculated values

Tube diameter 32 mm

Number of tubes 1156

Wall heat transfer coefficient 100 W/(m2.K)

Table 3.4: Reactor length for the first reactor for different conversions

Conversion (%) Length (m)

85 0.35

90 0.38

95 0.42

21
3.2.2 Run simulation

To run a simulation:

1. Click the Next button to confirm that you have finished entering all required input.

The Required Input Complete dialog box appears.

2. Click OK. A Control Panel appears and shows the run progress.

3. As the run proceeds, status messages appear in the Control Panel. When the

calculations are completed, the message Result Available appears in the status bar

at the right corner of the Aspen Plus main window.

3.2.3 Examine simulation results

When the message Results Available appears in the status bar simulation results

can be examined.

To examine the results:

1. From the Menu Tree, double-click the Results Summary folder.

2. In the Results Summary folder, click Run Status.

3. The Summary sheet appears.

4. The Aspen Plus version, run starting time and run status are summarized on this

sheet.

5. From the Results Summary folder, click Streams.

6. The Material sheet appears with the stream results.

The summary result of the simulation by Aspen plus will be compared with the

result from literature data (Olsson, 2013) especially the effect of temperature towards the

SBA conversion. Figure 3.3 shows the literature data for the reaction.

22
 Figure 3.3: Equilibrium curve for MEK (Olsson, 2013)

The determination of the conversion in the SBA dehydrogenation is related to the SBA

conversion. The definition of conversion (X) is the relation between the related amount

of SBA and the initial amount of SBA. In terms of moles (3.1) is:

(𝑀𝑜𝑙𝑒𝑆𝐵𝐴 )𝑖𝑛𝑙𝑒𝑡 − (𝑀𝑜𝑙𝑒𝑆𝐵𝐴 )𝑜𝑢𝑡𝑙𝑒𝑡

3.2.4 Sensitivity analysis

A sensitivity analysis is carried out for the SBA dehydrogenation reactor. The

effects of operating condition on the product will be analysed below. By using sensitivity

analysis tool at Aspen plus, the reactor pressure is varied from 0.5 to 3.5 bar in case 1,

reactor temperature is varied from 50°C to 600°C in case 2. SBA feed flow rate is varied

from 1000 kg/hr to 13000 kg/hr in case 3. Table 3.5 shows the range operating variables

used for sensitivity analysis.

23
 Table 3.5: Range operating variables used for sensitivity analysis

Parameter Lower range Upper range Increment

Temperature (°C) 50 600 50

Pressure (bar) 0.5 3.5 0.5

SBA feed (kg/hr) 1000 13000 1000

The steps for sensitivity analysis:

1. Select Model Analysis Tools and then select Sensitivity from the submenu.

2. Click New and a dialog box of Create New ID will appear. It will display an

automatically generated S-1. Rename the Enter ID in that box or just use as the

default.

3. Click OK to accept the Create New ID dialog box.

4. The Sensitivity Input Vary sheet for S-1 appears and click New.

5. Variable labelled with 1 will display for the Manipulated variable.

6. Edit the selected variable for the Type such as Stream-Var, then click the S1 and

TEMP for variable that are needed to manipulate.

7. Specify the limits for the manipulated variable limits and put the number points,

increment or list of values.

8. Repeat the step 4 to 7 for another parameters which are pressure and SBA feed

flow rate.

9. Click Next for Input Define sheet and edit selected variable by click New and

name it by BUTIN for feed stream of SBA. Add in the variable for BUTOUT

which is product stream of SBA.

10. Choose any of the category and the reference.

24