Flammability Limits StudyofVapour Mixtures

above Crude Oil at Low Temperatures

by

DebbieAloysius Sitiol

Dissertation submittedin partial fulfilmentof

the requirements for the

Bachelorof Engineering (Hons)

(Chemical Engineering)

MAY 2012

Universiti Teknologi PETRONAS

Bandar Sen Iskandar

31750 Tronoh

Perak Darul Ridzuan

CERTIFICATION OF APPROVAL

Flammability Limits Study ofVapour Mixtures

above Crude Oil at Low Temperatures

by

Debbie Aloysius Sitiol

A project dissertation submitted to the

Chemical Engineering Programme

Universiti Teknologi PETRONAS

In partial fulfilment of the requirements for the

BACHELOR OF ENGINEERING (Hons)

(CHEMICAL ENGINEERING)

(Dr. Mohanad EI-Harbawi)

W.W0H^ADEL-«ARB*1

Unbend

UNIVERSITI TEKNOLOGI PETRONAS

TRONOH, PERAK

MAY 2012

CERTIFICATION OF ORIGINALITY

This is to certify that I am responsible for the work submitted in this project, that the

original work is my own except as specified in the references and

acknowledgements, and that the original work contained herein have not been

undertaken or done by unspecified sources or person.

!&>

DEBBIE ALOYSIUS SITIOL

Abstract

Study of fire and explosion is very important mainly in oil and gas industries due to

the severity of fire and explosion incidents. Fire and explosion could cause property

damage and loss of lives. In this work, investigation had been carried out on the

flammability of crude oil at low temperatures 35°C, 40°C and 50°C. Hydrocarbon

components derived from refinery storage was assessed. The oil-liquid phase was

analyzed using Headspace-Gas Chromatography (HS-GC) and Gas Chromatography

Mass Spectrometry (GC-MS) to examine the composition of the sample.

Hydrocarbon compounds ranging from C6 to C9 were detected. Lower Flammability

Limits (LFLs) and Upper Flammability Limits (UFLs) for individual components

were calculated at each temperature using stoichiometric concentration method

proposed by Zabetakis et. al. Flammability limits of themixture, LFLmiX and UFLn^

were calculated using the Le Chatelier equation. Limiting Oxygen Concentration

(LOCs) for each temperature are calculated using Hansen and Crowl method, while

the estimation ofLOC for the mixtures (LOC„ux) is calculated using Zlowchower and

Green method. Flammability diagramwas constructed which is used to determinethe

flammability of the mixture at respective temperature. It is found that as the

temperature increases, the flammability range of vapours above crude oil increases

too. The findings of this studymay assist in minimizing fire hazards associated with

presence ofhydrocarbon vapours.

Acknowledgements

This dissertation for "Flammability limits Study of Vapour Mixtures above Crude

Oil at Low Temperatures" would not have been possible without the support,

guidance andmotivation from various individuals.

My utmost gratitude to my project supervisor, Dr. Mohanad El-Harbawi for his

continuous support and guidance throughout the project duration. He has beenvery

patient and committed in helping me to complete this project. His immeasurable

guidance has provided me a good opportunity to contribute to the success of this

research.

I would like to thank the Final Year Project Coordinators, Dr. Norhayati Bt. Mellon

and Puan Asna Bt. M Zain for disseminating required information effectively and

also for giving moral support to the students. Same goes to all lecturers and staffs of

Chemical Engineering Department, Universiti Teknologi PETRONAS.

Deepest thanks to Dr. IvyChai Ching Hsia and herteam from PETRONAS Research

Sdn. Bhd. for their support and cooperation in conducting some of the experimental

work needed in this research project.

Not forgetting, my family members and friends who have continuously given moral

support and encouragement to me to complete this project. Thank you for walking

beside me through the joy and rough patches of my life.

Thank You all.

u

TABLE OF CONTENTS

LIST OF FIGURES iv

LIST OF TABLES iv

ABBREVIATIONS AND NOMENCLATURES v

CHAPTER 1: INTRODUCTION 1

1.1. Background 1

1.2. Problem Statement 2

1.3. Objective 3

1.4. Scope of Study 3

CHAPTER 2: LITERATURE REVIEW 4

2.1. Introduction 4

2.3. Flammable Materials 7

2.4. Flammability Limits 8

2.5. Limiting Oxygen Concentrations 10

2.6. Flammability Diagram 10

2.7. Inherent Safety 12

CHAPTER 3: METHODOLOGY 13

3.1. Material and Method 13

3.2. Experimental and Theoretical Methods 13

3.2.1. Mole Fraction 15

3.2.2. Flammability Limits 15

3.2.3. Limiting Oxygen Concentration 18

3.2.4. Flammability Diagram 19

3.3. Gantt Chart 20

CHAPTER 4: RESULTS AND DISCUSSION 21

4.1. Components Identification 21

4.2. Mole Fraction in Vapour Phase 29

4.3. LFL5UFLandLOC 30

4.4. Flammability Diagram 33

4.5. Inherent Safety 35

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS 37

5.1. Recommendations 37

References 38

Appendices 42

m

LIST OF FIGURES

Figure 1: Fire Triangle 7

Figure 2: Flammability Diagram of Methane at 25°C and 1 atm (Brooks & Crowl,

2007) 11

Figure 3: Methodology flowchart 14

Figure 4: Chromatogram at 35°C 23

Figure 5: Chromatogram at 40°C 24

Figure 6: Chromatogram at 50°C 25

Figure7: Mole fraction in vapourphase at (a) 35°C (b) 40°C and (c) 50°C 30

Figure 8: LFLmix vs. Temperature 32

Figure 9: UFLmjX vs. Temperature 32

Figure 10: LOCmix vs. Temperature 32

Figure 11:Flammability Diagramat 35CC 33

Figure 12: Flammability Diagram at40°C 34

Figure 13: Flammability Diagram at 50°C 34

LIST OF TABLES

Table 1: Type oftank contents (J. I. Chang & Lin, 2006) 5

Table 2: Type ofaccidents (J. I. Chang & Lin, 2006) 5

Table 3: Major Tank Incidents between 1963 and 2002 (J. I. Chang & Lin, 2006) 6

Table 4: Components at 35°C 26

Table 5: Components at 40°C 27

Table 6: Components at 50°C 28

Table 7: Flammability Limits and LOC at 35°C 31

Table 8: Flammability Limits and LOC at 40°C 31

Table 9: Flammability Limits and LOC at 50°C 31

Table 10: LFL, UFL and LOC at each temperature 32

iv

ABBREVIATIONS AND NOMENCLATURES

AIChE The American Institute ofChemical Engineers

AIT Autoignition Temperature

BTX Benzene, Toluene, Xylene

FP Flash Point

GC-MS Gas Chromatography-Mass Spectroscopy

HS-GC Headspace-Gas Chromatography

LFL Lower Flammability Limit

LFLmix Lower Flammability Limit ofmixture

LOC Limiting Oxygen Concentration

LOCmix LimitingOxygenConcentration ofmixture

LNG Liquefied Natural Gas

MIE Minimum Ignition Energy

UFL Upper Flammability Limit

UFLaux Upper Flammability Limit ofmixture

FP Flash Point

LPG Liquefied Petroleum Gas

CHAPTER 1

INTRODUCTION

1.1. Background

Crude oil is a mixture of hydrocarbons in different lengths and structure. It ranges

from the lightest compound to complex paraffin and aromatic molecules. Complex

molecules play major role in defining the properties and processing method of

different types ofcrude oil (Speight & Ebrary, 1999). Oil refining converts crude oil

into various products depending on its composition. Most common end-product of

refinery are petroleum gas, gasoline, naphtha, kerosene, gas oil and lubricating oils.

In the industries, incidents that can cause toxic effects, fire or explosion couldresults

in property damage, lost of production, environmental impact and serious injuries

(AIChE, 2010b). With respect to hydrocarbon components, highly flammable

compounds such as benzene, toluene, ethylbenzene and xylenes may be present and

they pose significant threat of fire hazard. Light ends components until medium

naphtha fractions are most flammable components. Some of these compounds can

evaporate and turn into vapour form at ambient temperature and atmospheric

pressure, thus formingflammable mixtureswith air.

In the presence of sufficient amount of fuel, oxidizer and ignition source, vapour

mixture will burn if its concentration is in between Lower Flammable Limit (LFL)

and Upper Flammable Limit (UFL). The presence of flammable mixtures exposes

storage system to the possibility of fire and explosion events.

1.2. Problem Statement

Crude oil in refineries contains hundreds of substances namely hydrocarbon. At

ambient temperature and atmospheric pressure, some of these components could

vaporize to the atmosphere forming vapour mixtures. Different components in crude

oil vaporize at different temperature and pressure. Vapour mixture might be

flammable depending on the composition of the vapourmixture.

Industries especially the ones thatare involved in theoil andgas operation havebeen

operating at a risk by maintaining operations of flammable vapour mixtures within

flammability ranges especially in storage areas. This has beena threat to the industry

where many fires and explosion incidents have occurred in the past 50 years

involving such flammable vapour mixtures.

Common practice in determining flammability of dustandvapour is by using closed

explosion chamber with various volumes. 20-L sphere explosion chamber is known

internationally as laboratory test apparatus in determining the combustion

characteristicsofcombustible dusts and gases (Wang et al, 2010).

Although sphere explosion chamber are internationally recognized, it is highly costly

to purchase the apparatus. Experts are needed to operate the apparatus as well.

Besides that, it is also expensive to run the sphere explosion chamber to determine

the flammability ofvapour mixture.

This work is aimed on the study of the effect of temperature on the flammability of

hydrocarbon mixtureby incorporating fundamentals of thermodynamics with process

safety concept.

1.3. Objective

The objectives of this project are:

1. To estimate the UFLn^, LFLmix and LOCmn of hydrocarbon vapour mixtures

above refinery crude oil by using fundamental of thermodynamics and

process safety concept

2. To determine the flammability range of vapour mixture above crude oil at

differentlow temperatures based on flammability diagrammethod

3. To investigate the possible presence of flammable mixture in oil refinery

storage spaces and recommend safetymethods to prevent fire and explosion

incidents

1.4. Scope of Study

Crude oil in refineries could volatize and form flammable vapour mixture. When

flammable vapour mixture is exposed to ignition source and oxidizer, fire or

explosion will occur. Experimental work will be conducted to determine the

components of crude oil by using Headspace-Gas Chromatography (HS-GC) and

Gas Chromatography - Mass Spectrometry (GC-MS). Components of the crude oil

in vapourphase will be determined by applyingfundamental ofthermodynamics.

LFLmix and WL^ at low temperatures (35°C, 40°C and 50°C) will be calculated

using Le Chatelier equation (Le Chatelier, 1891) to predict the range of flammability

of the vapour mixture. LOCn«x is estimated using Zlochower & Green (2009)

method. Lastly, flammability diagram will be constructed to determine the

flammability of the sample.

The results from this work will be used to recommend safety methods to prevent the

flammable mixture from occurring in storage system. This work can contribute to

minimizing the loss of production, property damage and ensuring safety of

personnel.

2.1. Introduction

CHAPTER 2

LITERATURE REVIEW

Fires, explosions and toxic releases are common accidents in chemical plant.

Accident resulting from fires andexplosions canbe prevented if engineers know well

about the fire and explosion properties of materials (Crowl & Louvar, 2011).

Zabetakis (1965), also mentioned that knowledge on flammability characteristics

could prevent unwanted fires and gas explosion.

Fire and explosion incidents in oil refineries are not uncommon. Study shows that

74% of accidents occurred in petroleum refineries, oil terminals or storage and 85%

of the accidents are fire and explosion (J. I. Chang & Lin, 2006).

Refineries are classified as majorhazard installation as it possess a largeinventory of

hazardous material which exceed the threshold quantities (Shalufet al, 2003). There

are various complex processes and hazardous components' storage in a refinery. As

mentioned by Chang and Lin (2006), storage tanks in refineries and chemical plants

contain flammable and hazardous chemicals which are stored in large volume. At

any event where there is leakage or release ofthese hazardous materials, there will be

potential risk of fire and explosion.

Understanding the properties of flammable and hazardous material is critical.

Unwanted fire and explosion incident canbe prevented by knowing the flammability

limits ofthe hazardous material.

2.2. Previous Accidents

Chang and Lin (2006) had reviewed 242 accidents that occurred in industrial

facilities over the last 40 years. Based on their findings, storage tanks accidents

occurrence is more frequent in petroleum refineries with weightage of 47.9%

followed by terminal and storage area (26.4%).

Types of tank contents are provided in Table 1 where crude oil and oil products

storage tank is themajor contributor to storage tank accidents.

Tabic 1: Type of tank contents (J. I. Chang & L n,2006)

Year1960-

1969

1970-

1979

1980-

1989

1990-

1999

2000-

2003Total

Crude Oil 6 8 17 23 12 66

Oil Products (Fuel oil,diesel, kerosene, lubricants)

3 7 14 19 16 59

Gasoline/Naptha 0 13 17 21 6 55

Petro-Chemicals 3 3 4 11 6 27

LPG 3 3 1 5 1 15

Waste oil water 2 2 0 4 1 9

Ammonia 0 0 0 0 3 3

Hydrochloric Acid 1 2 3

Caustic Soda 3 3

Molten Sulfur 1 1 2

Total 17 36 53 85 51 242

Also in Chang and Yin (2006) study, fire and explosion together carried 85% of the

totalcases. Statistics on types of accidents is illustrated in Table2.

Table:.: Tvpco1'accidcn(s(J. . Chun"& .in, 2006)

Year1960-

1969

1970-

1979

1980-

1989

1990-

1999

2000-

2003Total

Fire 8 26 31 59 21 145

Explosion 8 5 16 22 10 61

Spill 0 5 3 2 8 18

Toxic Gas Release 0 0 2 1 10 13

Misc. 1 1 1 2 5

Total 17 36 53 85 51 242

Meanwhile Table 3 is an extractionfrom Ching and Yin (2006)study. Therewere 10

major storage tanks incidents that were recorded from 1963 to2002.

Tab c 3: Major Tank incidents between l%3 and 2002 (J. I. Chang & Lin, 2006)

No. Date Location Loss Description

1 2/24/86Thessaloniki,Greece

330

Sparks from a flame cutting torch ignitedfuel from a tank spill in a dike of a fuel tank.The fire spread to other areas resulting indestruction of 10 out of 12 crude oil tanks.

2 4/3/77 UMM said Qatar 179

A 260 000-barrel tank containing 236 000barrels of refrigerated propane at -45°Ffailure massively. An adjoining refrigeratedbutane tank and most of the process areawere also destroyed by fire.

3 1/20/68Pernis

Netherlands141

Frothing occurred when hot oil and wateremulsion in a slop tank reacted with volatileslop, causing a violent vapour release andboil-over. The fire destroyed 3 hydrocarbon,a sulphur plant and 80 storage tanks

4 9/1/79Deer Park,Texas, USA

138

Nearly simultaneous explosion abroad a 70000 DWT tanker off-loading and in an 80000 barrel ethanol at a refinery occurredduring an electric storm.

5 5/30/78Texas City,Texas, USA

120

An unidentified failure led to the release of

light hydrocarbons which spread to anignition source. 11 tanks in this alkylationunit were destroyed

6 8/20/81 Kuwait 73

Fire destroyed 8 tanks and damaged severalothers. The cause of the fire has not beendisclosed

7 9/14/97Vishakhapatnam,India

64

LPG ignited during tank loading from a ship.A thick blanket of smoke spreading panicamong the residents. 15 storage tanks burnedfor two days

8 12/21/85 Naples, Italy 60

24 out of 32 tanks at a marine petroleumproducts terminal destroyed by fire thatbegan with a tank overfill. Explosion causedcomplete destruction of the terminalbuildings and nearby industrial andresidential structures.

9 1/7/83Newark, NewJersey, USA

52

An overfilling of a floating roof tank spilled1300 barrels of gasoline into the tank dike.The vapour cloud carried by wind to anearby incinerator and was ignited. 2adjacent tanks and terminal was destroyed.

10 5/26/83Prodhoe Bay,Alaska, USA

47

A low-pressure NGL feed drumruptured in acrude oil station, resulting in fire damage toone third of the module and exterior of

surrounding structure within 100 ft.

2.3. Flammable Materials

Flammable gases and liquids can be characterized based on its properties such as

Lower flammability limit (LFL), upper flammability limit (UFL), limiting oxygen

concentration (LOC), flashpoint (FP), minimum ignition energy (MIE) and

autoignition temperature (AIT) (Crowl, 2012). Fuel combustion always occurs in

vapour phase. Flammable liquids will be volatilized to vapour phase andflammable

solids are decomposed to vapour and then onlythe vapourphasewill be ignited.

The essential element for fire to occur is fuel, oxidant (which is mostly oxygen) and

also an ignition source/heat. In theevent whereby one oftheelements is missing, fire

will not occur. Fire triangle (Figure 1) is a graphical representation of the three

elements that need to present for a fire to start.

Figure I: Fire Triangle

Although all three elements are present, there are certain concentration limits for a

fire to be ignited. Fuel and oxidant needs to be available in certain concentration

while ignition source must be strong enough to start the fire. The most common

oxidant in fire occurrence is oxygen.

Crude oil is made of mostly of carbon (80-87%) and hydrogen (10-15%). These

compounds are called as hydrocarbons. In crude oil, hydrocarbon exists in various

length and structures. Carbon atoms serve as the backbone with hydrogen atoms

surrounding it. Hydrocarbon chains can be broken and linked in various ways by

different processes. Longhydrocarbon chains can be restructured into shorter chains

and vice versa. Besides hydrocarbons, small amount of other elements in crude oil

are sulphur (0-10%), nitrogen (0-1%) andoxygen (0-5%).

Crude oils characteristics and types are depending on its geographical location.

Crude oils from the South America are thick and tarry while crude from North Africa

are lighter with lower density (BP Educational Service, 2006).

In general, crude oil is highly flammable. It can be ignited easily by heat, sparks or

flames. In vapour form, it may form combustible mixtures with contact of air. Crude

at temperature lower than its boiling point will give up vapour to the surface until the

vapour is in equilibrium with the crude oil. Highly flammable components in crude

oil are BTX (Benzene, Toluene and Xylene), ethylbenzene, cyclohexane etc

(CountryMark, 2009).

2.4. Flammability Limits

Flammable vapour mixture mixed with air in the range of flammable concentration is

likely to be caught on fire with the presence of ignition source (Carson & Mumford,

2002). Knowledge on the flammability limit of vapour mixture could prevent the

formation of vapour mixture in the flammable range. Characterisation of chemical

substances are important to determine its flammability to prevent unwanted fire and

explosion accident due to poor handling ofthe chemicals.

Flammable vapour and volatile liquids are highly hazardous when in contact with air

within the flammable range. Each component in the mixture has its own flammability

limits. Flammability limits measure the ability of a flame to propagate away from the

ignition source.

Lower Flammable Limits (LFL,-) and Upper Flammable Limits (UFL,) are the lower

and upper flammability limit respectively for component i in fuel and air. Below the

LFL, there is not enough fuel to cause ignition (too lean) while above the UFL, there

is not sufficient oxygen to promote ignition (too rich). Vapour mixture will be

categorized as flammable if its concentration is in the flammable range which is in

between LFL and UFL.

Several experimental works have been conducted to predict the flammability limits.One well known apparatus to obtain flammability data is a 20-litres spherecombustion apparatus; Brooks and Crowl (2007) studied experimentally on vapourflammability above aqueous solutions of ethanol and acetonitrile. Liekhus et al.(2000), conducted experiments to predict the flammability ofgas mixtures containinghydrogen and flammable or non-flammable volatile organic compounds (VOCs) inair. Cashdollar et al. (1992), performed experiment using 120-L chamber under

quiescent and turbulent conditions to study on the flammable gas generation fromnuclear waste tanks. Petersen et al estimated LFL ofcrude oilvapour from relieftank

vents by constructing 1:50 scaled models ofrelief tanks (Petersen et al, 1997).

Mixture concentration outside the range of the flammable limits will not ignite.

Basically, wider flammable range poses higher risk of fire. Flammable range isclosely related to temperature. According to (Zabetakis et al, 1958), flammable

ranges increaseswith temperature.

Estimation of flammability limits is required for some situation where experimental

data is not available. LFL/ and UFL, of many hydrocarbon vapours are found to be a

function of stoichiometric concentration (Jones, 1938). Flammability characteristics

of about 500 different substances of pure hydrocarbon fluids were predicted using

structural group contribution method (Albahri, 2003). Meanwhile, Crowl &

Mashuga, (1999) predicted flammability zone using calculated adiabatic flame

temperatures (CAFT).

Empirically derived equations are used to estimate LFL,- and UFL, at different

temperatures. Various components in mixtures' flammability limits (LFLm/* and

UFLmfc) are computed using Le Chatelier equation (Le Chatelier, 1891) which are

still widely used today. Hanley (1998) correlated LFIw with a quantity

representative of the heatevolved during combustion (Vidal et al., 2004).

In recent studies, El-Harbawi et al. (2012) predicted the flammability of vapours

above refinery wastewater laden with hydrocarbon mixtures. The flammability

estimation was conducted by correlating thermodynamics with process safety

concepts (El-Harbawi et al, 2012).

2.5. Limiting Oxygen Concentrations

Limiting oxygen concentration (LOC) is the minimum oxygen concentration

required to propagate flame. Below LOC, reaction will not be able to generate

sufficient energy to heat the mixture. Thus, mixture could not self-propagate.

Reducing the concentration of oxygen could prevent fire and explosion from

occurring. Razus et al, (2004) estimated LOC using values of lower explosion limit

of fuel-air mixture and also calculated adiabatic flame temperature (CAFT).

Correlation was establishedbetween computed CAFT for fuel-air-nitrogen mixtures

at LOC and CAFT at LEL, for a large numberof flammable gases and vapors.

In the absence of literature data, LOC can be estimated using the stoichiomety from

the combustion reaction and LFL. Besides that, LOC can be estimated using method

proposedby Hansen& Crowl (2010);

LOC^.(LFLi-C^UFL)& (2-1)1_(-loc VbL

Meanwhile, LOCmix is estimated using Zlochower & Green (2009) method. This

method was from the studyon spark ignited explosions in large, spherical laboratory

vessels using 7% pressure-rise criterion for explosion propagation. The calculated

results using this method predicts theexperimental values accurately for hydrocarbon

mixtures (Zlochower & Green, 2009).

2.6. Flammability Diagram

Flammability diagram is representing the flammability of vapour. It is the best tool to

determine the flammability region and also to determine if a flammable mixture is

present as it has axes for fuel, oxygen and nitrogen (Mashuga & Crowl, 1999). An

example of flammability diagram can be seen in Figure 2. Concentration of fuel,

oxygen and inert material are plotted on each axes of the triangle. Referring to the

flammability diagram, air line represents all the possible concentrations of fuel and

air. Stoichiometric line signifies the combinations of fuel and oxygen. If

10

stoichiomeric line intersects the flammability zone bounded by LFL and UFL line,

vapour mixture is flammable and can cause fire.

Various individuals have used flammability diagram to represent the flammability of

components and mixtures. Flammability characteristics of 3-picoline/water mixtures

were studied using a 20-L vessel. Findings from the study was represented in

flammability diagram to show the possible flammable mixture ratio (Y. M. Chang et

al, 2006). Mao et al. (2011) also illustrated possible mixture composition using

standard flammability diagram in their study on backdraught in tunnel fires. The

flammability envelope plotted for mass fraction of n-heptane (fuel), oxygen and

nitrogen (Mao et al, 2011).

Flammability diagramis not only used to determine the flammability range of single

component fuel. Chang et al. (2006) studies the flammability on benzene and

methanol with different vapour mixing ratios under different initial conditions. The

flammability of benzene, methanol and their mixture was investigated in a 20-L

spherical explosion vessel (Y. M. Chang et al, 2006). A triangular diagram was

plotted with different ratios ofbenzene and methanol.

Figure 2: Ffctiimiability Diagram of Methane at 25°C arid 1 aim (Brooks & Crow!, 2007)

11

2.7. Inherent Safety

"What you don't have, can't leak" (T. A. Kletz, 1978) is a catchy line by Kletz

(1978) whom started the concept of reducing rather than controlling hazards.

Inherent Safety is defined as a condition in which the hazards associated with the

materials and operations used in the process have been reduced or eliminated

permanently (Bollinger & Crowl, 1997). There is no single index or numerical

values that could define Inherently Safer (Dowell Iii, 2006). Dowell Iii (2006), also

mentioned that there is always a trade-off between different types of hazards.

Eliminating onehazard might bring another newhazard to the situation.

Applying Inherent Safer design concepts can reduce the number of accidents in

industry. Applying this concept at the designing phase could benefit in having low

cleaning up cost in the future. Inherent safer design will enhance overall risk

management by reducing the frequency of potential accident such as fire and

explosion.

There are 11 approaches to inherent safer design: Substitution,

mmimization/intensification, moderation/attenuation, simplification, limitation of

hazardous effects, avoiding knock-on effect, making incorrect assembly impossible,

make status clear, tolerance, ease of control and administrative controls/procedures

(T. Kletz & Kletz, 1998).

AIChE Center for Chemical Process Safety (CCPS, 2009) reduced the 11 concepts of

inherently safer processes to four principles:

• Minimize the amount ofhazardous material present at any given time.

• Substitute hazardous materials by least hazardous materials.

• Moderate the operating conditions of pressure, temperature and

concentrations.

• Simplify the plant since simple process plants are easier to operate and

maintained with fewer chances of things going wrong.

12

CHAPTER 3

METHODOLOGY

3.1. Tools and Materials

1. Sample from one of the refinery in Malaysia

2. Gas Chromatography - Mass Spectrometer (GC-MS)

3. Gas Chromatography- Headspace (GC-HS)

3.2. Experimental and Theoretical Methods

Sample of crude oil is obtained from a refinery and was brought back to Universiti

Teknologi PETRONAS fully insulated in an air tight bottle. Simple distillation was

conducted to remove the water content in the sample. A portion of the sample was

analysed using HS-GC at 35°C, 40°C and 50°C. The crude sample analysis was

performed by PerkinElmer Clarus 500 Mass Spectrometer. MS spectra was

compared using Turbomass Gold Software. For all temperatures, samples were

injected into Perkin Elmer Elite 5MS N9316282 capillary column with 30m length,

0.22 mm i.d and 0.25 um film thickness. Inlet pressure of column is 15psi. Volume

of injected sample is 1.0 uL with electron energy 70 eV. Split vent is set to 20

ml/min with ratio of 19:1.

GC identifies various compounds at different retention times. From GC-MS, peak

areais proportional to the amount of compound presence in thesample.

Resulting data from the GC-MS will be screened and analysed. Compound with

>95% similarities with the library will be taken into consideration. Besides the

percentage of similarity, boiling points of each component selected will be studied.

Components with lower boiling points should be eluted and get detected earlier

compared to higher boiling points components. Thus, boiling points of each

13

component selected should be increasing together with the retention time. Boiling

points data are extracted from Yaws' Thermophysical Properties of Chemicals and

Hydrocarbons (Carl L; Yaws et al, 2009).

Sample collection

Identify the sample contents using HS-GC & GC-MS

Is the mixture flammable?

Estimate y;

Estimate LFLU UFLU andLOQat 30,40 & 50°C

Estimate LFLmix, UFLmi}L, and LOCmix at

30,40 & 50°C

J.Draw the flammability diagram

?igure3: Methodology flowchai

14

Yes

3.2.1. Mole Fraction

Mole fraction in vapour phase can be obtained from each compound's mass fraction.

Massfraction of eachcomponent in the vapour phaseof crudeoil is determined from

peak area data and will be calculatedusing Eq. (3-1):

Jxrr (3-1)

where xt - Massfraction of component/

A{ = Peak area of component i

AT = Peak area of all components

Depending on the temperature, some components in the liquidphasewill vaporise to

the vapour phase. Since fire and explosion occurs in the vapours phase, it is required

to know the concentration of hydrocarbon components in the vapourphase. The mass

fraction of each component in vapour phase can be converted to mole fraction in

vapour phase, yt using Eq. (3-2):

ytx; IMt

Hxi/M* (3-2)

Where Mt= the molecular weight of component i.

3.2.2.Flammability Limits

LFL, and UFL, ofeach component i in vapour mixture was calculated using equation

proposed by Jones (1938) in Eq. (3-3) and Eq. (3-4). Jones (1938) found that for

many hydrocarbon vapours, LFL and UFL both are functions of stoichiometric

concentration of fuel (Cst);

LFL =0.55C„ (3_3)

UFL =3.5C„ (3^}

15

The stoichiometric concentration for most organic compounds was determined using

the generalcombustionreaction(Eq. 3-5)

CmHxOy +z02 -> mC02 +\~\H20(3-5)

Where z is equivalent moles of (Vmoles fuel and can be found from Eq. (3-6);

jc yz~m + —

4 2

(3-6)The stoichiometric concentration (Cst)can befound asa function ofz (Eq. (3-7))

r r moles fuelCs( = [ ] x 100

moles fuel + moles air

100 (3-7)z

1 +0.21

LFL; and UFL, at 25°C are calculated by substituting Eq. (3-6) into Eq. (3-7) and

applying it into Eq. (3-3) and Eq. (3-4). The resulting equations are Eq. (3-8) and Eq.

(3-9) for LFL* and UFL,- respectively.

0.55(100)LFL25 4.76m +JJ9x-2.38y +l (3_8)

UFL„ =25 4.76m +1.19x-2.38y +l « «

3.50(100)

Calculation of LFL, and UFL, at 35, 40 and 50°C were calculated using empirically

derived equations (Zabetakis etal, 1958) (Eq. (3-10) andEq. (3-11)).

0 75LFLr =LFL«-—(r-25)

^ (3-10)

UFLr=UFL25 +̂ (T-25)^ (3-11)

16

Where AHC is the net heat of combustion in kcal/mole and T is the temperature in °C.

AHC for each components are obtained from Yaws' Handbook of Thermodynamic

and Physical Properties ofChemical Compounds (Yaws, 2003).

Le Chatelier (1891) proposed empirical models to calculate the LFL and UFL of

multiple fuel mixtures. The modelswere givenby;

LFL . (vol%)= ;mix n ,

I bAFLT)i = l (3-12)

UFLmix(vol%) = *1 btMLr)

i = l (3-13)

where, LFLT- the lower flammable limit for component i (in volume %) of

component i in fuel and air at T,

UFLT= the upper flammable limit for component i (involume %) of

component / in fuel and air at T

yt = the mole fraction ofcomponent i on a combustible basis and

n = the number of combustible species.

Saturated vapour pressure, P"at of each component was calculated using Antoine

equations (Eq. (3-14)). Antoine Coefficient were extracted from Yaws' Handbook of

Antoine Coefficients for Vapor Pressure (Carl L. Yaws et al, 2009).

logl0Psat=A BC+ T (3-14)

Where A, B and C are Antoine Coefficients. They are specific and vary for each

component. T is temperature in °C.

17

3.2.3.Limiting Oxygen Concentration

Limiting Oxygen Concentration of each component, LOC, is estimated using the

method proposed by Hansen & Crowl (2010) in Eq. (3-15):

L0Ci=(LFLi-CUK;UFLi)(UFI,)1-C^ "UFL

(3-15)

Where LOC,- = Limitingoxygenconcentration for component i

LFL* = Lower flammable limit for component i

UFL, = Upper flammable limit for component i UFL

UFL0 = Oxygen concentration at the upper flammable limit (vol% oxygen

in air)

Cloc = fitting constant

UFL0is calculated using Eq. (3-16)

UFL0=0.21(100-LFL,) (3-16)

Cloc is a fitting constant whereby according to Crowl, Cloc = -1.11 is valid for most

hydrocarbons (Hansen & Crowl, 2010).

According to Crowl (2011), Cloc = -1.11 is valid for most hydrocarbons. This value

was obtained from data analysis ofvarious experimental values.

Limiting oxygen concentration for the mixture, LOCmix is estimated usingZlochower

and Green (2009) method (Eq. (3-17));

LOC .= ^2"V'

L0C< (3-17)

18

3.2.4. Flammability Diagram

Flammability diagram at each temperature (35, 40 and 50°C) will be constructed to

study effect of flammability range with respect with temperature. Flammability

diagram consists of three axes namely fuel, oxygen and nitrogen/inert material. It is

diagram which could represent the flammability ofgaseous mixture.

There are a few methods to construct the flammability diagram. Crowl & Louvar,

(2001) describe in detail on the construction of flammability diagram. In this work,

flammability limits and limiting oxygen concentration will be used mainly in the

construction ofthe flammability diagram.

19

3.3

.G

an

ttC

hart

FY

Pl

FY

P2

#A

cti

on

item

s1

23

45

67

89

10

11

12

13

14

12

34

56

7

a

89

10

11

12

13

14

1S

elec

tion

ofP

roje

ctT

itle

2P

reli

min

ary

Res

earc

hW

ork

3S

ub

mis

sio

no

fE

xte

nd

ed

Pro

posa

l

4P

ropo

salD

efen

ce

5P

roje

ctW

ork

•at

a £ CO u a

6S

ub

mis

sio

no

fIn

teri

mD

raft

1

7S

ubm

issi

ono

fIn

teri

mR

epor

t1M at

Mi

9S

ubm

issi

ono

fPro

gres

sR

epor

tI

10

Pre

-ED

X|i

11

Su

bm

issi

on

of

Dra

ftR

epor

t11

12

Su

bm

issi

on

of

Dis

sert

ati

on

13

Sub

mis

sion

of

Tec

hnic

alP

aper

14

Ora

lP

rese

nta

tio

n

15

Su

bm

issi

on

of

Hard

Bo

un

d

Dis

sert

ati

on

20

CHAPTER 4

RESULTS AND DISCUSSION

4.1. Components Identification

Chromatogram at eachtemperature; 35°C, 40°C and 50°C were studied. Components

with matchingof more than 95% with the Mass-Spectroscopy library were taken into

considerations. Figure 4, Figure 5 and Figure 6 shows the chromatogram at each

temperature that shows the retention time and peak abundance.

From the first analysis, the components' boiling points were gathered. Boilingpoints

must be increasing together with the retention time. Components with lowerboiling

point will be eluted first thus should be detected at an earlier retention time.

There are 11, 9 and 16 components identified respectively for 35°C, 40°C and 50°C.

Components for each temperature are listed in Table 4,

21

Table 5 and Table 6 below; Components presence in the vapour mixture ranging

from Ce to C9 at lower temperatures. Most components are from the family of

Alkanes and Alkenes.

The mass fraction of each component was calculated by dividing the respective peak

area with the total peak areas of all components. (Eq.(3-1))

22

cru

de

oii

72

_3

5c

100-

,19

81

.84

1.8

0

2.0

7

1.7

8

%-

2.6

0 2.9

1

4.7

0

3.5

74

.06

6.1

1

J[A

VA

AA

;I""

II'

•••

0.9

81

.98

2.9

83

.98

4.9

8

}3

5C

,0

.5m

lsa

mp

lecru

de,

29

-May

-20

12

+1

2:2

9:1

5S

can

EI+

TIC

9.8

2e9

11

.30

"I1

''•'P

'''I"

"I1

"'I•'•'TT'1"1I'

fFT'Tl'll

II|IIII

|III

I|II

IIITl

II|T111)ll'ITI-PIII|IIII|IIIl|

I

5.9

86

.98

7.9

88

.98

9.9

81

0.9

81

1.9

81

2.9

81

3.9

81

4.9

8

^A

/VT

ime

Fig

ure

4:C

hru

mat

og

ny

siat

35

°C

23

Cru

de

Oil

_7

2_

40

C

100-

,2-

11

%-

1.9

6

T-l-

T

2.1

9

2.5

2/ 2

.77

5.0

63

.11

6.5

6

•!••••!

3.7

5-8

.12

12

.02

8.7

5T

""

-^

13

.75

ii

l.i

ii

iI

ii

ii

Ii

i1

8.7

52

3.7

5

I40

C,

0.5

ml

sam

ple

cru

de,

28

-May

-20

12

+1

2:3

9:4

0S

can

EI+

TIC

7.8

6e9

28

.75

Ii

''

'I

''

33

.75

••!•'•

38

.75

''

i'

'i

ii

ii

Tim

e4

3.7

5

Fig

ure

?:

Ch

rom

ato

Km

raat

40

°C

24

tn

LU 1- oc Tas coo

CO

+

©

I

©

0)

Ias

If)

©

id

o

n

O

in

O

te

Tab

le4

:C

om

po

nen

tsat

35

°C

No

.

Peak

Nu

mb

er

inG

C-

HS

Rete

nti

on

Tim

e(m

in)

Co

mp

ou

nd

Data

base

mat

ch(%

)F

orm

ula

CA

S

Boi

ling

Po

int

(°C

)

Are

a

(Ab

un

dan

ce)

Mass

Fra

ctio

n,

x;

14

2.2

1B

UT

AN

E,2

,2-D

IME

TH

YL

-9

8.8

CfiH

147

5-8

3-2

49

.73

82

54

72

72

0.1

04

29

2.9

1C

YC

LO

PE

NT

AN

E,

ME

TH

YL

-9

8.4

C«H

«9

6-3

7-7

71

.81

17

78

02

14

40

.22

4

31

23

.75

CY

CL

OP

EN

TA

NE

,1

,3-D

IME

TH

YL

-,C

IS-

95

.4C7

H14

25

32

-58

-39

0.7

73

58

70

79

20

.04

5

41

33

.86

CY

CL

OP

EN

TA

NE

,1

,3-D

IME

TH

YL

-,T

RA

NS

-9

8.1

C7H

U1

75

9-5

8-6

91

.73

55

28

78

80

0.0

70

51

44

.05

6H

EP

TA

NE

97

.6C7

Hlfi

14

2-8

2-5

98

.43

95

42

70

48

0.1

20

61

54

.7C

YC

LO

HE

XA

NE

,M

ET

HY

L-

98

.4C7

HU

10

8-8

7-2

10

0.9

32

17

11

05

44

0.2

74

71

76

.37

3H

EP

TA

NE

,3

-ME

TH

YL

-9

8.7

CRH1

R5

89

-81

-11

18

.93

28

26

66

14

0.0

36

81

86

.64

1,3

-DIM

ET

HY

LC

YC

LO

HE

XA

NE

,C

&T

98

CsH

lfi5

91

-21

-91

19

.95

25

15

87

40

0.0

32

91

96

.74

5C

YC

LO

HE

XA

NE

,1

,4-D

IME

TH

YL

-9

8.5

CsH

lfi5

89

-90

-21

22

.75

11

22

89

74

0.0

14

10

20

7.5

9O

CT

AN

E9

7.1

C*H

,*1

11

-65

-91

25

.68

26

20

59

90

0.0

33

11

29

11

.29

9P

-XY

LE

NE

99

.2C8

H10

10

6-4

2-3

13

8.3

63

77

00

49

60

.04

8

79

26

06

49

41

26

Tab

le5:

C01

11J.

oo

en

tsat

40

CC

No

.

Peak

Nu

mb

er

in

GC

-HS

Rete

nti

on

Tim

e(m

in)

Com

poun

dD

ata

base

mat

ch(%

)F

orm

ula

CA

S

Boi

ling

Po

int

CO

Are

a

(Abu

ndan

ce)

Mass

Fra

ctio

n,

x5

14

2.3

43

BU

TA

NE

,2,2

-DIM

ET

HY

L9

8.2

C6H

U7

5-8

3-2

49

.73

38

40

07

44

0.0

95

08

29

3.1

1C

YC

LO

PE

NT

AN

E,

ME

TH

YL

-9

7.7

CfiH

„9

6-3

7-7

71

.81

95

32

23

52

0.2

36

02

31

34

.13

7C

YC

LO

PE

NT

AN

E,

1,3

-DIM

ET

HY

L-,

TR

AN

S-

97

.5C7

H,4

17

59

-58

-69

1.7

32

76

61

58

20

.06

84

9

41

44

.34

8H

EP

TA

NE

96

.9C7

H16

14

2-8

2-5

98

.43

48

02

23

52

0.1

18

91

51

55

.05

2C

YC

LO

HE

XA

NE

,M

ET

HY

L-

98

C7H1

41

08

-87

-21

00

.93

12

51

48

19

20

.30

98

7

61

76

.83

6H

EP

TA

NE

,3

-ME

TH

YL

-9

8C8

H18

58

9-8

1-1

11

8.9

31

43

57

16

20

.03

55

5

71

87

.11

71,

3-D

IME

TH

YL

CY

CL

OH

EX

AN

E,

C&

T9

6.5

C8H1

fi5

91

-21

-91

19

.95

14

24

62

67

0.0

35

27

82

08

.11

8O

CT

AN

E9

5.9

CRH!

R1

11

-65

-91

25

.68

16

83

48

20

0.0

41

68

92

91

2.0

18

P-X

YL

EN

E9

9CR

H]n

10

6-4

2-3

13

8.3

62

38

74

62

40

.05

91

1

40

38

68

09

51

27

Tab

le6:

Co

mp

on

en

tsat

50

°C

No

.

Peak

Nu

mb

er

inG

C-

HS

Rete

nti

on

Tim

e

(min

)C

om

po

un

dD

ata

base

mat

ch(%

)F

orm

ula

CA

S

Boi

ling

Po

int

(°C

)

Are

a

(Abu

ndan

ce)

Mass

Fra

ctio

n,X

i

15

2.1

7B

UT

AN

E,

2,2

-DIM

ET

HY

L-

96

.80

CfiH

u7

5-8

3-2

49

.65

68

78

02

64

.00

0.0

60

0

29

2.8

8C

YC

LO

PE

NT

AN

E,

ME

TH

YL

-9

7.8

0Cf

iH1?

96

-37

-77

1.8

12

26

16

47

84

.00

0.1

97

4

31

33

.82

CY

CL

OP

EN

TA

NE

,1

,3-D

IME

TH

YL

L-,

CIS

-9

6.3

0C7

H14

25

32

-58

-39

0.7

77

53

89

54

4.0

00

.06

58

41

44

.01

HE

PT

AN

E9

6.4

0C7

H1fi

14

2-8

2-5

98

.43

14

04

24

57

6.0

00

.12

26

51

54

.66

CY

CL

OH

EX

AN

E,

ME

TH

YL

-9

8.4

0C7

H1d

10

8-8

7-2

10

0.9

33

54

46

32

32

.00

0.3

09

4

61

76

.35

HE

PT

AN

E,

3-M

ET

HY

L-

97

.20

C«H

,85

89

-81

-11

18

.93

48

88

26

04

.00

0.0

42

7

71

96

.72

CY

CL

OH

EX

AN

E,

1,4

-DIM

ET

HY

L-T

RA

NS

-9

6.8

0CS

H,6

22

07

-04

-71

19

.36

18

17

46

10

.00

0.0

15

9

82

06

.94

CY

CL

OH

EX

AN

E,

1,1

-DIM

ET

HY

L-

98

.10

C8H

lfi5

90

-66

-91

19

.55

99

49

91

6.0

00

.00

87

92

17

.48

CY

CL

OH

EX

AN

E,

1,2

-DIM

ET

HY

L-,

TR

AN

S-

96

.80

C*H

,fi6

87

6-2

3-9

12

3.4

31

65

54

43

9.0

00

.01

45

10

22

7.5

8O

CT

AN

E9

5.3

0CS

H,8

11

1-6

5-9

12

5.6

86

13

83

06

0.0

00

.05

36

11

23

7.8

0C

YC

LO

HE

XA

NE

,1

,4-D

IME

TH

YL

-9

7.8

0<

W*

58

9-9

0-2

12

5.8

91

08

55

64

3.0

00

.00

95

12

26

9.3

7H

EP

TA

NE

,2

,5-D

IME

TH

YL

-9

7.4

0<

W.

22

16

-30

-01

36

.01

99

30

82

3.0

00

.00

87

13

28

10

.77

ET

HY

LB

EN

ZE

NE

97

.90

CA

ft1

00

-41

-41

36

.20

95

62

89

9.0

00

.00

83

14

29

11

.36

P-X

YL

EN

E9

9.0

0C«

H1B

10

6-4

2-3

13

8.3

66

76

84

35

2.0

00

.05

91

15

30

11

.61

OC

TA

NE

,3

-ME

TH

YL

-9

7.7

0C

,H,»

22

16

-33

-31

44

.23

10

18

94

68

.00

0.0

08

9

16

31

12

.75

O-X

YL

EN

E9

7.1

0Cf

tHm

95

-47

-61

44

.43

17

17

68

46

.00

0.0

15

0

11

45

56

70

60

.00

1.0

0

28

4.2. Mole Fraction in Vapour Phase

Quantitative analysis of crude oil sample was conducted. Fat for each components

were calculated using Antoine Equation (Eq. (3-14). Antoine Coefficients for each

components are provided in Appendix A. In this work, Psat are generally ranging

from 15.25 mmHg to 766.7 mmHg. The large range of Fat gives wide variation in

the values of mole fraction in vapour phase, yh Heavier components with low

volatilityhave smallervalue oiyt.

Graphical representation of the mole fraction in vapour phase of each component at

35°C, 40°C and 50°Care illustrated in Figure 7.

(a)

co

tsmfa.

u.

V

o

S

* .O'S^,*F^<>^^^

^&

tf&

<?

V^-i^-t,^oP& &*

tv*

x^' c»o*&*cy»

Compounds

Compounds

29

#

<? xP* "V?

^ *f C^ $

/>v.J5N

4? <f <f «T <f ^ J* <f ^ <& ^ <f ^

y y <^to* ^

J" jf >£

<?

V £

</#

Compounds

^

#•y

Figure 7: Mole fraction iii vapour phase at (a) 35°C (b) 40°C and (c) 50°C

4.3. LFL, UFL and LOC

LFL, and UFL, of each component i in vapour phase at 25°C had been calculated as

given in Eq. (3-11) and Eq. (3-12). LFL,- and UFL/ ofeach component i are compared

with literature value obtained from DIPPR Project 801 (AIChE, 2010a). Graphical

comparison of literature values of LFL, and UFL, with calculated values is included

in Appendix B. Some literature value of LFL and UFL that could not be found in

other database or published literatures are not indicated in the appendix.

LFL,- and UFL, of each component i at 35, 40 and 50°C are calculated using Eq. (3-

13) and Eq. (3-14). Net heat of combustion, AHc for each component that was

obtained from Yaws* Handbook of Thermodynamic and Physical Properties of

Chemical Compounds are listed in Appendix C. Flammability limits of each

component are listed in Table 7, Table 8 and Table 9.

As mentioned earlier, Le Chatelier equation (Eq. (3-12) and Eq. (3-13)) is used to

obtain the LFLmix and UFLmix in the vapour mixture. Values obtained from the

calculation are summarised in Table 10.

LOC, of each component i is calculated using the method proposed by Hansen and

Crowl (2010) in Eq. (3-17). LOC values ofeach component are illustrated in Table 7,

30

Table 8 and Table 9. Meanwhile LOCmix is obtained by applying Zlochower andGreen (2009)methodin Eq. (3-18).

Table 7: Flammabilitv Limits andLOCat35DCNo. Components LF^ LFL35 UFL25 UFL35 LOC35

1 BUTANE,2,2-DIMETHYL- 1.19 1.18 7.57 7.58 11.6442 CYCLOPENTANE, METHYL- 125 1.25 7.98 7.99 11.5923 CYCLOPENTANE, 1,3-DIMETHYL-, CIS- 1.08 1.07 6.87 6.87 11.7334 CYCLOPENTANE, 1,3-DIMETHYL-, TRANS- 1.08 1.07 6.87 6.87 11.7335 HEPTANE 1.03 1.02 6.56 6.57 11.7726 CYCLOHEXANE, METHYL- 1.08 1.07 6.87 6.87 11.7337 HEPTANE, 3-METHYL- 0.91 0.90 5.79 5.79 11.8698 1,3-DIMETHYLCYCLOHEXANE,C&T 0.95 0.94 6.02 6.03 11.8409 CYCLOHEXANE, 1,4-DIMETHYL- 0.95 0.94 6.02 6.03 11.84010 OCTANE 0.91 0.90 5.79 5.79 11.86911 P-XYLENE 1.08 1.07 6.87 6.87 11.733

Tabic 8: Flammabilitv Limits and LOC at 40°C

No. Components LFL* LFL40 UFL?5 UFL40 LOC501 BUTANE,2,2-DIMETHYL 1.190 1.178 7.572 7.585 11.6382 CYCLOPENTANE, METHYL- 1.255 1.242 7.984 7.996 11.5863 CYCLOPENTANE, 1,3-DIMETHYL-, TRANS- 1.079 1.068 6.865 6.876 11.7274 HEPTANE 1.031 1.020 6.559 6.570 11.7665 CYCLOHEXANE, METHYL- 1.079 1.068 6.865 6.876 11.7276 HEPTANE, 3-METHYL- 0.909 0.900 5.785 5.794 11.8637 1,3-DIMETHYLCYCLOHEXANE. C&T 0.946 0.937 6.022 6.032 11.8338 OCTANE 0.909 0.900 5.785 5.794 11.8639 P-XYLENE 1.079 1.068 6.865 6.876 11.727

Table 9: Flammabilitv Limits and LOC at 50°C

No. Components LFL25 LFL50 UFL25 UFL50 LOC501 BUTANE, 2,2-DIMETHYL- 1.19 1.17 7.57 7.59 11.6252 CYCLOPENTANE, METHYL- 1.25 1.23 7.98 8.00 11.5743 CYCLOPENTANE, 1,3-DIMETHYLL-, CIS- 1.08 1.06 6.87 6.88 11.7154 HEPTANE 1.03 1.01 6.56 6.58 11.7535 CYCLOHEXANE, METHYL- 1.08 1.06 6.87 6.88 11.7156 HEPTANE, 3-METHYL- 0.91 0.89 5.79 5.80 11.8517 CYCLOHEXANE, 1,4-DIMETHYL- TRANS- 0.95 0.93 6.02 6.04 11.8218 CYCLOHEXANE, 1,1-DIMETHYL- 0.95 0.93 6.02 6.04 11.8219 CYCLOHEXANE, 1,2-DIMETHYL-, TRANS- 0.95 0.93 6.02 6.04 11.82110 OCTANE 0.91 0.89 5.79 5.80 11.85111 CYCLOHEXANE, 1,4-DIMETHYL- 0.95 0.93 6.02 6.04 11.82312 HEPTANE, 2,5-DIMETHYL- 0.81 0.80 5.17 5.19 11.92813 ETHYLBENZENE 1.08 1.06 6.87 6.88 11.71514 P-XYLENE 1.08 1.06 6.87 6.88 11.71515 OCTANE, 3-METHYL- 0.81 0.80 5.17 5.19 11.92816 O-XYLENE 1.08 1.06 6.87 6.88 11.715

31

Results from calculation are tabulated in Table 10 and represented graphically in

Figure 9, Figure 8 and Figure 10.

Table 10: lFL, UFL and LOC at each temi. erature

T 35°C 40°C 50°C

LFLmix 1.16058 1.15783 1.13466

UFUh 7.444245 7.456181 7.365457

LOCh 11.65627 11.65363 11.65360

1.11

1.10

1.09

1.08

1.07

1.06

1.099

LFLmix vs. Temperature

1.096

1.068

5035 40 ,„TemperaturefC)

Figure S: UFL„,ix vs. Temperature

UFLmix vs. Temperature

7.10 •)7.048 7"058

7.05 -i* i

9 7.00 -jX 6.95 1

J iLL.

36.90 -j6.85 A— —„_ —„ -,.„_„____—™__ _.

6.934

UQ

35 40Temperature(°C)

Figure 9: LFL,!lK vs. Temperature

LOCmix vs. Temperature

11.65627

35 40Temperature(°C)

iT vs. lemperanire

32

50

11.65360

50

4.4. Flammability Diagram

Flammability diagram is represented by three axes: Fuel (hydrocarbon vapour

mixture), inert material (nitrogen) and oxidizer (oxygen). Concentrations of fuel,

nitrogen and oxygen are required in order to plot the triangular flammability

diagram.

Air line in the flammability diagram is plotter by taking the composition of air

(assumed as 21% oxygen, 79% nitrogen). The intersection of the stoichiometric line

with oxygen axis is given by 100(z/l+z). LOC,^ is drawn by locating the \X)Cmix

value on the oxygen axis then drawing a parallel line until it intersects with the

stoichiometric line. Meanwhile \J¥\,mix and LFLm/x are located at the air line and

extended to show the flammability range.

Flammability diagram at 35°C, 40°C and 50°C were constructed to assess the

flammability region of vapour mixture above crude oil at those three various

temperatures. As can be seen in Figure 11, Figure 12 and Figure 13, at each

temperature, the stoichiometric line goes through the flammable zone. This shows

that crude oil could give off enoughvapour that could be ignited in the vapourphase.

Therefore, vapour released from the storage tank poses potential hazard and can be

ignited with the presence of ignition source.

•fliim-eoRW-^, -S^

?i«ure i 1: F! a mmability Diagram at 35°C

33

&n%

:LOC / V.•a

« \

\e\*

/ \ \ \\\ \* 1

\ v\* i Xc • *

.Sfeichi

/ A^Uy*L/ 7y HIFL \ \ \ -,

so- at JO. jy 50 60- id ^ as «ra.

WBrogen % LFL

I1-OS6voJ&)

riyure 52: Flammabiiity Diagram at 40°C

LOC

imjes4-jDi&r"\ s

Figure 33: Flammability Diagram a

34

5UUC

4.5. Inherent Safety

Flammable liquids are considered to be more hazardous when compared to

combustible liquids. Flammable liquids will ignite and burn easily atnormal working

temperature compared to combustible liquids. Most fire cases are contributed by the

presence of flammable liquids. Flammable liquids will not burnt by itself. As

mentioned earlier, flammable liquids will give off vapour which will burn with the

presence of ignition source.

The wanner the liquid's temperature, themore vapour components and concentration

will begiven off. Vaporization of flammable liquids depends onits temperature and

vapour pressure.

At studied temperatures (35°C, 40°C and 50°C) crude oil (flammable liquids) can

give off enough vapour to form flammable mixtures with air. Extra caution on each

components' LFL and UFL need tobetaken to minimize thepossibility of fire. Table

7, Table 8 and Table 9 illustrated the list of LFL and UFL of each component. It is

important to keep the concentration of these components outside of the flammable

range. Outside of theflammable range, thevapour could notbe ignited. Meanwhile,

combustible liquids at temperature above its flash point will release enough vapourwhich will be flammable too.

High concentration of flammable vapour mixture from storage tank is favourable to

beprevented from being released to the atmosphere. Flammable vapour mixture was

released from thestorage tankcantravel to an ignition source andcause fire. Firecan

be prevented by controlling and minimizing the ignition source such as hot works

(cutting and welding, electrical sparks, heating equipment etc), hot surfaces,

lightning, and open flames nearby the storage area. Besides that, hazardous

chemicals storage tanks should be designed in a way where there should not be a

domino effect in the eventof fireor explosion.

From theresults presented in this study, the following recommendations canbe made

to prevent fire from occurring by thepresence of flammable vapours:

35

1. Operation at less than LFL is often considered to be safer than operation at

above UFL.

2. Eliminate ignition sources from areas where flammable vapours may be

present or near enough for vapour to travel.

3. Storage tank vents should not be in a confined space where vapour mixture

can be accumulated and forming flammable mixture.

4. Storage tank must be placed some distances away from critical process area.

5. Samples should be taken frequently for analysis on its flammability limits.

36

CHAPTERS

CONCLUSIONS AND RECOMMENDATIONS

Crude oil mixture consists of various hydrocarbon chain; different structures and

length. At low temperatures, the liquid mixture could give off enough vapour that

could be ignited with the presence of oxygen and ignition source at the flammable

range.

Flammability of crude oil was analysed by correlating the principle of

thermodynamics and process safety. Identification of significant components was

conducted using gas chromatography and analysis of the flammability of each

component will be carried out at low temperature. Flammability range of the

components mixture was represented in flammability diagram whereby flammable

range were identified.

This study could be used to recommend safety methods that could prevent fire and

explosion due to the presence of flammable mixture. This work can contribute to

minimizing the loss of production, property damage andriskof personnel injury.

3.1.Recommendations

It should be noted that when all the necessary actions have been taken to evaluate the

root cause of fire and explosion incident, the hazards shall be controlled and

minimized accordingly as to avoid recurrence. Therefore, industries should adapt a

more preventive strategy such as prevention of ignition sources/hot works nearby

area with possible traces of hydrocarbons. Besides that, implementing good plant

layout design could avoid any domino effect in the event of fire.

For future works, experimental works can be conducted to study on flammability

limit of mixture using 20-L combustion chamber. Results obtained can be compared

with the theoretical values calculatedin this study.

37

Mashuga, C. V., & Crowl, D. A. (1999). Flammability zone prediction using

calculated adiabatic flame temperatures, Process Safety Progress, 18(3), 127-134,

Petersen, R. L., Watson, K. D., & Roehner, R. (1997). LFL estimates for crude oil

vapors from relief tank vents. Process Safety Progress, 16(2), 72-77.

Razus, D., Molnarne, M., & FuAY, O. (2004). Limiting oxygen concentration

evaluation in flammable gaseous mixtures by means of calculated adiabatic flame

temperatures. Chemical Engineering andProcessing: Process Intensification, 43(6%

775-784.

Shaluf, I. M., Ahmadun, F. R., & Said, A. M. (2003). Fire incident at a refinery in

West Malaysia: the causes and lessons learned. Journal of Loss Prevention in the

Process Industries, 16(4), 297-303.

Speight, J. G., & Ebrary, I. (1999). The chemistry and technology ofpetroleum.

Vidal, M., Rogers, W. J., Holste, J. C, & Mannan, M. S. (2004). A review of

estimation methods for flash points and flammability limits. Process SafetyProgress,

23(1), 47-55.

Wang, J., Zhong, S., Li, X., & Wang, F. (2010). Development of20 liter spherical

explosion test apparatus based on LabVIEWand OPC technique.

Yaws, C. L. (2003). Yaws' Handbook of Thermodynamic and Physical Properties of

Chemical Compounds: Knovel.

Yaws, C. L., Narasimhan, P., K.; , & Gabbula, C. (2009). Yaws' Thermophysical

Properties ofChemicals and Hydrocarbons (Electronic Edition): Knovel.

Yaws, C. L., Narasimhan, P. K., & Gabbula, C. (2009). Yaws' Handbook of Antoine

Coefficients for Vapor Pressure (2nd Electronic Edition): Knovel.

Zabetakis, M. G. (1965). Flammability characteristics of combustible gases and

vapors: DTIC Document.

Zabetakis, M. G., Lambiris, S., & Scott, G. S. (1958). Flame temperatures of limit

mixtures.

40

Zlochower, I. A., & Green, G. M. (2009). The limiting oxygen concentration and

flammability limits of gases and gas mixtures. Journal of Loss Prevention in the

Process Industries, 22(4), 499-505.

41

APPENDICES

42

APPENDIX A: Antoine Coefficients

No ComponentsCAS

RegistryA B C

1 2,2-dimethylbutane 75-83-2 6.93954 1169.51 238.418

2 1,1-dimethylcyclohexane 590-66-9 6.86464 1363.21 222.636

3 trans-l,2-dimethylcyclohexane 6876-23-9 6.92012 1420.58 228.259

4 1,3-dimethylcyclohexane, cis and trans 591-21-9 7.07327 1436.62 222.718

5 1,4-dimethylcyclohexane,(cis+trans) 589-90-2 7.07703 1446.39 221.94

6 cis-1,3-dimethylcyclopentane 2532-58-3 7.14306 1421.81 242.813

7 trans-1,3-dimethylcyclopentane 1759-58-6 7.07206 1369.35 234.986

8 ethylbenzene 100-41-4 7.1561 1559.55 228.582

9 heptane 142-82-5 7.04605 1341.89 223.733

10 methylcyclohexane 108-87-2 7.00107 1375.13 232.819

11 methylcyclopentane 96-37-7 7.06372 1304.45 240.043

12 3-methylheptane 589-81-1 7.12293 1477.48 229.359

13 3-methyloctane 2216-33-3 7.17567 1594.94 227.131

14 octane 111-65-9 7.14462 1498.96 225.874

15 o-xylene 95-47-6 7.14914 1566.59 222.596

16 p-xylene 106-42-3 7.15471 1553.95 225.23

17 trans-1,4-dimethylcyclohexane 2207-04-7 6.87792 1369.34 223.224

18 2,5-dimethylheptane 2216-30-0 7.03511 1458.75 215.134

43

APPENDIX B: Flammability Limits' Literature and Calculated values at25°C

No ComponentsCAS

RegistryLFL/,, LFL25>Cflfc UFLffi UFL25,cofc

1 2,2-dimethvlbutane 75-83-2 1.20 1.19 7.00 7.57

?, 1,1-dimethylcyclohexane 590-66-9 1.00 0.95 6.10 6.02

3

trans-1,2-dimethylcvclohexane

6876-23-9 1.00 0.95 6.50 6.02

4

1,3-dimethylcyclohexane,cis and trans

591-21-9 N/A 0.95 N/A 6.02

5

1,4-dimethylcyclohexane,(cis+trans)

589-90-2 N/A 0.95 N/A 6.02

6

cis-1,3-dimethvlcvclopentane

2532-58-3 1.10 1.08 7.30 6.87

7

trans-1,3-dimethvlcvclopentane

1759-58-6 1.10 1.08 7.30 6.87

8 ethvlbenzene 100-41-4 1.00 1.08 6.70 6.87

9 heptane 142-82-5 1.05 1.03 6.70 6.56

10 methvlcvclohexane 108-87-2 1.15 1.08 6.70 6.87

11 methvlcyclopentane 96-37-7 1.20 1.25 8.40 7.98

12 3-methylheptane 589-81-1 0.79 0.91 5.80 5.79

13 3-methvloctane 2216-33-3 0.76 0.81 5.40 5.17

14 octane 111-65-9 0.96 0.91 6.50 5.79

15 o-xylene 95-47-6 1.10 1.08 6.40 6.87

16 D-xvlene 106-42-3 1.10 1.08 6.60 6.87

17

trans-1,4-dimethylcyclohexane

2207-04-7 1.00 0.95 6.50 6.02

18 2,5-dimethylheptane 2216-30-0 N/A 0.81 N/A 5.17

44

APPENDIX C: Net Heat of Combustion (AHC)

No ComponentsCAS

RegistryAHC

(kCal/mol)1 2,2-dimethylbutane 75-83-2 918.52

2 1,1-dimethylcyclohexane 590-66-9 1163.74

3 trans-1,2-dimethylcyclohexane 6876-23-9 1163.83

4 1,3-dimethylcyclohexane, cis and trans 591-21-9 1162.76

5 1,4-dimethylcyclohexane, (cis+trans) 589-90-2 1238.76

6 cis-1,3-dimethylcyclopentane 2532-58-3 1023.04

7 trans-1,3-dimethylcyclopentane 1759-58-6 1039.

8 ethylbenzene 100-41-4 1039.89

9 heptane 142-82-5 1068.19

10 methylcyclohexane 108-87-2 1018.28

11 methylcyclopentane 96-37-7 878.68

12 3-methylheptane 589-81-1 1213.57

13 3-methyloctane 2216-33-3 1359.77

14 octane 111-65-9 1214.26

15 o-xylene 95-47-6 1037.04

16 p-xylene 106-42-3 1037.07

17 trans-1,4-dimethylcyclohexane 2207-04-7 1161.63

18 2,5-dimethylheptane 2216-30-0 1358.28

45